Search

Cardiac arrhythmias are the leading cause of death and pose a huge health and economic burden globally. Electrocardiography (ECG) is an effective technique for the diagnosis of cardiovascular diseases because of its noninvasive and cost-effective advantages. However, traditional ECG analysis relies heavily on the clinical experience of physicians, which can be challenging and time-consuming to produce valid diagnostic results. This work proposes a new hybrid deep learning model that combines convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU) with multi-head attention (CBGM model). Specifically, the model consists of seven convolutional layers with varying filter sizes (4, 16, 32, and 64) and three pooling layers, respectively, while the BiGRU module includes two layers with 64 units each followed by multi-head attention (8-heads). The combination of CNN and BiGRU effectively captures spatio-temporal features of ECG signals, with multi-head attention comprehensively extracted global correlations among multiple segments of ECG signals. The validation in the MIT-BIH arrhythmia database achieved an accuracy of 99.41%, a precision of 99.15%, a specificity of 99.68%, and an F1-Score of 99.21%, indicating its robust performance across different evaluation metrics. Additionally, the model's performance was evaluated on the PTB Diagnostic ECG Database, where it achieved an accuracy of 98.82%, demonstrating its generalization capability. Comparative analysis against previous methods revealed that our proposed CBGM model exhibits more higher performance in automatic classification of arrhythmia and can be helpful for assisting clinicians by enabling real-time detection of cardiac arrhythmias during routine ECG screenings., (© 2024. The Author(s).)

Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary care physicians (PCPs), particularly in low-resource settings. Here, to bridge the gaps, we developed an integrated image-language system (DeepDR-LLM), combining a large language model (LLM module) and image-based deep learning (DeepDR-Transformer), to provide individualized diabetes management recommendations to PCPs. In a retrospective evaluation, the LLM module demonstrated comparable performance to PCPs and endocrinology residents when tested in English and outperformed PCPs and had comparable performance to endocrinology residents in Chinese. For identifying referable DR, the average PCP's accuracy was 81.0% unassisted and 92.3% assisted by DeepDR-Transformer. Furthermore, we performed a single-center real-world prospective study, deploying DeepDR-LLM. We compared diabetes management adherence of patients under the unassisted PCP arm (n = 397) with those under the PCP+DeepDR-LLM arm (n = 372). Patients with newly diagnosed diabetes in the PCP+DeepDR-LLM arm showed better self-management behaviors throughout follow-up (P < 0.05). For patients with referral DR, those in the PCP+DeepDR-LLM arm were more likely to adhere to DR referrals (P < 0.01). Additionally, DeepDR-LLM deployment improved the quality and empathy level of management recommendations. Given its multifaceted performance, DeepDR-LLM holds promise as a digital solution for enhancing primary diabetes care and DR screening., (© 2024. The Author(s).)

Microscopic cell detection is a challenging task due to significant inter-cell occlusions in dense clusters and diverse cell morphologies. This paper introduces a novel framework designed to enhance automated cell detection. The proposed approach integrates a deep learning model that produces an inverse distance transform-based detection map from the given image, accompanied by a secondary network designed to regress a cell density map from the same input. The inverse distance transform-based map effectively highlights each cell instance in the densely populated areas, while the density map accurately estimates the total cell count in the image. Then, a custom counting-aided cell center extraction strategy leverages the cell count obtained by integrating over the density map to refine the detection process, significantly reducing false responses and thereby boosting overall accuracy. The proposed framework demonstrated superior performance with F-scores of 96.93%, 91.21%, and 92.00% on the VGG, MBM, and ADI datasets, respectively, surpassing existing state-of-the-art methods. It also achieved the lowest distance error, further validating the effectiveness of the proposed approach. These results demonstrate significant potential for automated cell analysis in biomedical applications.

Aim: To develop and validate a deep learning (DL) algorithm for the automated detection and classification of carotid artery plaques (CAPs) on computed tomography angiography (CTA) images., Materials and Methods: This retrospective study enrolled 400 patients (300 in the Center Ⅰ and 100 in Ⅱ). Three radiologists co-labeled CAPs, and their revised calcification status (noncalcified, mixed, and calcified) was regarded as ground truth. Center Ⅰ patients were randomly divided into training and internal validation datasets, while Center Ⅱ patients served as the external validation dataset. Carotid artery regions were segmented using a modified 3D-UNet network, followed by CAPs detection and classification using a ResUNet-based architecture in a two-step DL system. The DL model's detection and classification performance were evaluated on the validation dataset using precision-recall curve, free-response receiver operating characteristic (fROC) curve, Cohen's kappa, and ROC curve analysis., Results: The DL model had achieved 83.4% sensitivity at 3.0 false positives (FPs)/CTA scan in internal validation and 78.9% in external validation. F1-scores were 0.764 and 0.769 at the optimal threshold, and area under fROC curves were 0.756 and 0.738, respectively, indicating good overall accuracy for CAP detection. The DL model also showed good performance for the ternary classification of CAPs, with Cohen's kappa achieved 0.728 and 0.703 in both validation datasets., Conclusion: This study demonstrated the feasibility of using a fully automated DL-based algorithm for the detection and ternary classification of CAPs, which could be helpful for the workloads of radiologists., (Copyright © 2024 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.)

Purpose: Neoadjuvant chemoradiotherapy has been the standard practice for patients with locally advanced rectal cancer. However, the treatment response varies greatly among individuals, how to select the optimal candidates for neoadjuvant chemoradiotherapy is crucial. This study aimed to develop an endoscopic image-based deep learning model for predicting the response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer., Methods: In this multicenter observational study, pre-treatment endoscopic images of patients from two Chinese medical centers were retrospectively obtained and a deep learning-based tumor regression model was constructed. Treatment response was evaluated based on the tumor regression grade and was defined as good response and non-good response. The prediction performance of the deep learning model was evaluated in the internal and external test sets. The main outcome was the accuracy of the treatment prediction model, measured by the AUC and accuracy., Results: This deep learning model achieved favorable prediction performance. In the internal test set, the AUC and accuracy were 0.867 (95% CI: 0.847-0.941) and 0.836 (95% CI: 0.818-0.896), respectively. The prediction performance was fully validated in the external test set, and the model had an AUC of 0.758 (95% CI: 0.724-0.834) and an accuracy of 0.807 (95% CI: 0.774-0.843)., Conclusion: The deep learning model based on endoscopic images demonstrated exceptional predictive power for neoadjuvant treatment response, highlighting its potential for guiding personalized therapy., (© 2024. The Author(s).)

Harmful algal blooms (HABs) are challenging to recognize because of their striped and uneven biomass distributions. To address this issue, a refined deep-learning algorithm termed HAB-Ne was developed for the recognition of HABs in GF-1 Wide Field of View (WFV) images using Noctiluca scintillans algal bloom as an example. First, a pretrained image super-resolution model was integrated to improve the spatial resolution of the GF-1 WFV images and minimize the impact of mixed pixels caused by the strip distribution. Side-window convolution was also explored to enhance the edge features of HABs and minimize the effects of uneven biomass distribution. In addition, a convolutional encoder-decoder network was constructed for threshold-free HAB recognition to address the dependence on thresholds in existing methods. HAB-Net effectively recognized HABs from GF-1 WFV images, achieving an average precision of 90.1% and an F1-score of 0.86. HAB-Net showed more fine-grained recognition results than those of existing methods, with over 4% improvement in the F1-Score, especially in the marginal areas of HAB distribution. The algorithm demonstrated its effectiveness in recognizing HABs in different marine environments, such as the Yellow Sea, East China Sea, and northern Vietnam. Additionally, the algorithm was proven suitable for detecting the macroalga Sargassum. This study demonstrates the potential of deep-learning-based fine-grained recognition of HABs, which can be extended to the recognition of other fine-scale and strip-distributed objects, such as oil spills and Ulva prolifera., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Quantitative phase contrast microscopy (QPCM) can realize high-quality imaging of sub-organelles inside live cells without fluorescence labeling, yet it requires at least three phase-shifted intensity images. Herein, we combine a novel convolutional neural network with QPCM to quantitatively obtain the phase distribution of a sample by only using two phase-shifted intensity images. Furthermore, we upgraded the QPCM setup by using a phase-type spatial light modulator (SLM) to record two phase-shifted intensity images in one shot, allowing for real-time quantitative phase imaging of moving samples or dynamic processes. The proposed technique was demonstrated by imaging the fine structures and fast dynamic behaviors of sub-organelles inside live COS7 cells and 3T3 cells, including mitochondria and lipid droplets, with a lateral spatial resolution of 245 nm and an imaging speed of 250 frames per second (FPS). We imagine that the proposed technique can provide an effective way for the high spatiotemporal resolution, high contrast, and label-free dynamic imaging of living cells.

DANCE is the first standard, generic, and extensible benchmark platform for accessing and evaluating computational methods across the spectrum of benchmark datasets for numerous single-cell analysis tasks. Currently, DANCE supports 3 modules and 8 popular tasks with 32 state-of-art methods on 21 benchmark datasets. People can easily reproduce the results of supported algorithms across major benchmark datasets via minimal efforts, such as using only one command line. In addition, DANCE provides an ecosystem of deep learning architectures and tools for researchers to facilitate their own model development. DANCE is an open-source Python package that welcomes all kinds of contributions., (© 2024. The Author(s).)

Background: Blood-invasive fungal infections can cause the death of patients, while diagnosis of fungal infections is challenging. Methods: A high-speed microscopy detection system was constructed that included a microfluidic system, a microscope connected to a high-speed camera and a deep learning analysis section. Results: For training data, the sensitivity and specificity of the convolutional neural network model were 93.5% (92.7-94.2%) and 99.5% (99.1-99.5%), respectively. For validating data, the sensitivity and specificity were 81.3% (80.0-82.5%) and 99.4% (99.2-99.6%), respectively. Cryptococcal cells were found in 22.07% of blood samples. Conclusion: This high-speed microscopy system can analyze fungal pathogens in blood samples rapidly with high sensitivity and specificity and can help dramatically accelerate the diagnosis of fungal infectious diseases.

The prevalence of metabolic diseases is increasing at a frightening rate year by year. The burgeoning development of deep learning enables drug design to be more efficient, selective, and structurally novel. The critical relevance of Histone deacetylase 11 (HDAC11) to the pathogenesis of several metabolic diseases makes it a promising drug target for curbing metabolic disorders. The present study aims to design new specific HDAC11 inhibitors for the treatment of metabolic diseases. Deep learning was performed to learn the properties of existing HDAC11 inhibitors and yield a novel compound library containing 23,122 molecules. Subsequently, the compound library was screened by ADMET properties, Lipinski & Veber rules, traditional machine classification models, and molecular docking, and 10 compounds were screened as candidate HDAC11 inhibitors. The stability of the 10 new molecules was further evaluated by deploying RMSD, RMSF, MM/GBSA, free energy landscape mapping, and PCA analysis in molecular dynamics simulations. As a result, ten compounds, Cpd&#95;17556, Cpd&#95;2184, Cpd&#95;8907, Cpd&#95;7771, Cpd&#95;14959, Cpd&#95;7108, Cpd&#95;12383, Cpd&#95;13153, Cpd&#95;14500and Cpd&#95;21811, were characterized as good HDAC11 inhibitors and are expected to be promising drug candidates for metabolic disorders, and further in vitro, in vivo and clinical trials to demonstrate in the future., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

Background: Automated surgical workflow recognition is the foundation for computational models of medical knowledge to interpret surgical procedures. The fine-grained segmentation of the surgical process and the improvement of the accuracy of surgical workflow recognition facilitate the realization of autonomous robotic surgery. This study aimed to construct a multigranularity temporal annotation dataset of the standardized robotic left lateral sectionectomy (RLLS) and develop a deep learning-based automated model for multilevel overall and effective surgical workflow recognition., Methods: From December 2016 to May 2019, 45 cases of RLLS videos were enrolled in our dataset. All frames of RLLS videos in this study are labeled with temporal annotations. The authors defined those activities that truly contribute to the surgery as effective frames, while other activities are labeled as under-effective frames. Effective frames of all RLLS videos are annotated with three hierarchical levels of 4 steps, 12 tasks, and 26 activities. A hybrid deep learning model were used for surgical workflow recognition of steps, tasks, activities, and under-effective frames. Moreover, the authors also carried out multilevel effective surgical workflow recognition after removing under-effective frames., Results: The dataset comprises 4 383 516 annotated RLLS video frames with multilevel annotation, of which 2 418 468 frames are effective. The overall accuracies of automated recognition for Steps, Tasks, Activities, and under-effective frames are 0.82, 0.80, 0.79, and 0.85, respectively, with corresponding precision values of 0.81, 0.76, 0.60, and 0.85. In multilevel effective surgical workflow recognition, the overall accuracies were increased to 0.96, 0.88, and 0.82 for Steps, Tasks, and Activities, respectively, while the precision values were increased to 0.95, 0.80, and 0.68., Conclusion: In this study, the authors created a dataset of 45 RLLS cases with multilevel annotations and developed a hybrid deep learning model for surgical workflow recognition. The authors demonstrated a fairly higher accuracy in multilevel effective surgical workflow recognition when under-effective frames were removed. Our research could be helpful in the development of autonomous robotic surgery., (Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc.)

Differences in gait patterns of children with Duchenne muscular dystrophy (DMD) and typically developing (TD) peers are visible to the eye, but quantifications of those differences outside of the gait laboratory have been elusive. In this work, we measured vertical, mediolateral, and anteroposterior acceleration using a waist-worn iPhone accelerometer during ambulation across a typical range of velocities. Fifteen TD and fifteen DMD children from 3 to 16 years of age underwent eight walking/running activities, including five 25 m walk/run speed-calibration tests at a slow walk to running speeds (SC-L1 to SC-L5), a 6-min walk test (6MWT), a 100 m fast walk/jog/run (100MRW), and a free walk (FW). For clinical anchoring purposes, participants completed a Northstar Ambulatory Assessment (NSAA). We extracted temporospatial gait clinical features (CFs) and applied multiple machine learning (ML) approaches to differentiate between DMD and TD children using extracted temporospatial gait CFs and raw data. Extracted temporospatial gait CFs showed reduced step length and a greater mediolateral component of total power (TP) consistent with shorter strides and Trendelenberg-like gait commonly observed in DMD. ML approaches using temporospatial gait CFs and raw data varied in effectiveness at differentiating between DMD and TD controls at different speeds, with an accuracy of up to 100%. We demonstrate that by using ML with accelerometer data from a consumer-grade smartphone, we can capture DMD-associated gait characteristics in toddlers to teens.

Diabetic retinopathy (DR) is the leading cause of preventable blindness worldwide. The risk of DR progression is highly variable among different individuals, making it difficult to predict risk and personalize screening intervals. We developed and validated a deep learning system (DeepDR Plus) to predict time to DR progression within 5 years solely from fundus images. First, we used 717,308 fundus images from 179,327 participants with diabetes to pretrain the system. Subsequently, we trained and validated the system with a multiethnic dataset comprising 118,868 images from 29,868 participants with diabetes. For predicting time to DR progression, the system achieved concordance indexes of 0.754-0.846 and integrated Brier scores of 0.153-0.241 for all times up to 5 years. Furthermore, we validated the system in real-world cohorts of participants with diabetes. The integration with clinical workflow could potentially extend the mean screening interval from 12 months to 31.97 months, and the percentage of participants recommended to be screened at 1-5 years was 30.62%, 20.00%, 19.63%, 11.85% and 17.89%, respectively, while delayed detection of progression to vision-threatening DR was 0.18%. Altogether, the DeepDR Plus system could predict individualized risk and time to DR progression over 5 years, potentially allowing personalized screening intervals., (© 2024. The Author(s).)

Pneumonia is a highly lethal disease, and research on its treatment and early screening tools has received extensive attention from researchers. Due to the maturity and cost reduction of chest X-ray technology, and with the development of artificial intelligence technology, pneumonia identification based on deep learning and chest X-ray has attracted attention from all over the world. Although the feature extraction capability of deep learning is strong, existing deep learning object detection frameworks are based on pre-defined anchors, which require a lot of tuning and experience to guarantee their excellent results in the face of new applications or data. To avoid the influence of anchor settings in pneumonia detection, this paper proposes an anchor-free object detection framework and RSNA dataset based on pneumonia detection. First, a data enhancement scheme is used to preprocess the chest X-ray images; second, an anchor-free object detection framework is used for pneumonia detection, which contains a feature pyramid, two-branch detection head, and focal loss. The average precision of 51.5 obtained by Intersection over Union (IoU) calculation shows that the pneumonia detection results obtained in this paper can surpass the existing classical object detection framework, providing an idea for future research and exploration., (© 2024. The Author(s).)

Decoding speech envelopes from electroencephalogram (EEG) signals holds potential as a research tool for objectively assessing auditory processing, which could contribute to future developments in hearing loss diagnosis. However, current methods struggle to meet both high accuracy and interpretability. We propose a deep learning model called the auditory decoding transformer (ADT) network for speech envelope reconstruction from EEG signals to address these issues. The ADT network uses spatio-temporal convolution for feature extraction, followed by a transformer decoder to decode the speech envelopes. Through anticausal masking, the ADT considers only the current and future EEG features to match the natural relationship of speech and EEG. Performance evaluation shows that the ADT network achieves average reconstruction scores of 0.168 and 0.167 on the SparrKULee and DTU datasets, respectively, rivaling those of other nonlinear models. Furthermore, by visualizing the weights of the spatio-temporal convolution layer as time-domain filters and brain topographies, combined with an ablation study of the temporal convolution kernels, we analyze the behavioral patterns of the ADT network in decoding speech envelopes. The results indicate that low- (0.5-8 Hz) and high-frequency (14-32 Hz) EEG signals are more critical for envelope reconstruction and that the active brain regions are primarily distributed bilaterally in the auditory cortex, consistent with previous research. Visualization of attention scores further validated previous research. In summary, the ADT network balances high performance and interpretability, making it a promising tool for studying neural speech envelope tracking., Competing Interests: Declaration of Conflicting InterestsThe authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Image reconstruction from limited and/or sparse data is known to be an ill-posed problem and a priori information/constraints have played an important role in solving the problem. Early constrained image reconstruction methods utilize image priors based on general image properties such as sparsity, low-rank structures, spatial support bound, etc. Recent deep learning-based reconstruction methods promise to produce even higher quality reconstructions by utilizing more specific image priors learned from training data. However, learning high-dimensional image priors requires huge amounts of training data that are currently not available in medical imaging applications. As a result, deep learning-based reconstructions often suffer from two known practical issues: a) sensitivity to data perturbations (e.g., changes in data sampling scheme), and b) limited generalization capability (e.g., biased reconstruction of lesions). This paper proposes a new method to address these issues. The proposed method synergistically integrates model-based and data-driven learning in three key components. The first component uses the linear vector space framework to capture global dependence of image features; the second exploits a deep network to learn the mapping from a linear vector space to a nonlinear manifold; the third is an unrolling-based deep network that captures local residual features with the aid of a sparsity model. The proposed method has been evaluated with magnetic resonance imaging data, demonstrating improved reconstruction in the presence of data perturbation and/or novel image features. The method may enhance the practical utility of deep learning-based image reconstruction.

Background: An emerging strategy to expand the donor pool is the use of a steatotic donor liver (SDLs; ≥ 30% macrosteatosis on biopsy). With the obesity epidemic and prevalence of nonalcoholic fatty liver disease, SDLs have been reported in 59% of all deceased donors. Many potential candidates need to decide whether to accept an SDL offer or remain on the waitlist for a nonsteatotic donor liver (non-SDL). The objective of this study was to compare the survival of accepting an SDL vs using a non-SDL after waiting various times., Methods: Using data from the United States' organ procurement and transplantation network, deep survival learning predictive models were built to compare post-decision survival after accepting an SDL vs waiting for a non-SDL. The comparison subjects contain simulated 20,000 different scenarios of a candidate either accepting an SDL immediately or receiving a non-SDL after waiting various times. The research variables were selected using the LASSO-Cox and Random Survival Forest (RSF) models. The Cox proportional hazards and RSF models were also comparatively included for survival prediction. In addition, personalized survival curves for randomly selected candidates were generated., Result: Deep survival learning outperformed Cox proportional hazards and RSF in predicting the survival of liver transplants. Among the simulations, 25% to 30% of scenarios demonstrated a higher 3-year survival post-decision for candidates accepting an SDL than waiting and receiving a non-SDL. The difference was only 1.43% in 3-year survival post-decision between accepting an SDL and waiting 260 days (mean waitlist time) for a non-SDL. As the number of days on the waitlist increases, the difference in survival between accepting SDLs and waiting for non-SDLs decreases., Conclusions: Appropriately used SDLs could expand the donor pool and relieve the candidates' unmet need for donor livers, which presents long-term survival benefits for recipients., Competing Interests: Declaration of Competing Interest The authors have no financial or proprietary interests in any material discussed in this article., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Introduction: This study aimed to develop and assess a deep-learning model based on CT images for distinguishing infectivity in patients with pulmonary tuberculosis (PTB)., Methods: We labeled all 925 patients from four centers with weak and strong infectivity based on multiple sputum smears within a month for our deep-learning model named TBINet's training. We compared TBINet's performance in identifying infectious patients to that of the conventional 3D ResNet model. For model explainability, we used gradient-weighted class activation mapping (Grad-CAM) technology to identify the site of lesion activation in the CT images., Results: The TBINet model demonstrated superior performance with an area under the curve (AUC) of 0.819 and 0.753 on the validation and external test sets, respectively, compared to existing deep learning methods. Furthermore, using Grad-CAM, we observed that CT images with higher levels of consolidation, voids, upper lobe involvement, and enlarged lymph nodes were more likely to come from patients with highly infectious forms of PTB., Conclusion: Our study proves the feasibility of using CT images to identify the infectivity of PTB patients based on the deep learning method., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Gao, Zhang, Hu, He, Fu, Lin, Liu, Fu, Liu, Sun, Chen, Yang and Zhou.)

Gestational diabetes mellitus (GDM) is a common complication of pregnancy, which has significant adverse effects on both the mother and fetus. The incidence of GDM is increasing globally, and early diagnosis is critical for timely treatment and reducing the risk of poor pregnancy outcomes. GDM is usually diagnosed and detected after 24 weeks of gestation, while complications due to GDM can occur much earlier. Copy number variations (CNVs) can be a possible biomarker for GDM diagnosis and screening in the early gestation stage. In this study, we proposed a machine-learning method to screen GDM in the early stage of gestation using cell-free DNA (cfDNA) sequencing data from maternal plasma. Five thousand and eighty-five patients from north regions of Mainland China, including 1942 GDM, were recruited. A non-overlapping sliding window method was applied for CNV coverage screening on low-coverage (~0.2×) sequencing data. The CNV coverage was fed to a convolutional neural network with attention architecture for the binary classification. The model achieved a classification accuracy of 88.14%, precision of 84.07%, recall of 93.04%, F1-score of 88.33% and AUC of 96.49%. The model identified 2190 genes associated with GDM, including DEFA1, DEFA3 and DEFB1. The enriched gene ontology (GO) terms and KEGG pathways showed that many identified genes are associated with diabetes-related pathways. Our study demonstrates the feasibility of using cfDNA sequencing data and machine-learning methods for early diagnosis of GDM, which may aid in early intervention and prevention of adverse pregnancy outcomes., (© The Author(s) 2024. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Accurate prediction of nucleic binding residues is essential for the understanding of transcription and translation processes. Integration of feature- and template-based strategies could improve the prediction of these key residues in proteins. Nevertheless, traditional hybrid algorithms have been surpassed by recently developed deep learning-based methods, and the possibility of integrating deep learning- and template-based approaches to improve performance remains to be explored. To address these issues, we developed a novel structure-based integrative algorithm called NABind that can accurately predict DNA- and RNA-binding residues. A deep learning module was built based on the diversified sequence and structural descriptors and edge aggregated graph attention networks, while a template module was constructed by transforming the alignments between the query and its multiple templates into features for supervised learning. Furthermore, the stacking strategy was adopted to integrate the above two modules for improving prediction performance. Finally, a post-processing module dependent on the random walk algorithm was proposed to further correct the integrative predictions. Extensive evaluations indicated that our approach could not only achieve excellent performance on both native and predicted structures but also outperformed existing hybrid algorithms and recent deep learning methods. The NABind server is available at http://liulab.hzau.edu.cn/NABind/., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2023 Jiang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Accurate gland segmentation is critical in determining adenocarcinoma. Automatic gland segmentation methods currently suffer from challenges such as less accurate edge segmentation, easy mis-segmentation, and incomplete segmentation. To solve these problems, this paper proposes a novel gland segmentation network Dual-branch Attention-guided Refinement and Multi-scale Features Fusion U-Net (DARMF-UNet), which fuses multi-scale features using deep supervision. At the first three layers of feature concatenation, a Coordinate Parallel Attention (CPA) is proposed to guide the network to focus on the key regions. A Dense Atrous Convolution (DAC) block is used in the fourth layer of feature concatenation to perform multi-scale features extraction and obtain global information. A hybrid loss function is adopted to calculate the loss of each segmentation result of the network to achieve deep supervision and improve the accuracy of segmentation. Finally, the segmentation results at different scales in each part of the network are fused to obtain the final gland segmentation result. The experimental results on the gland datasets Warwick-QU and Crag show that the network improves in terms of the evaluation metrics of F1 Score, Object Dice, Object Hausdorff, and the segmentation effect is better than the state-of-the-art network models., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Objective: The accurate diagnosis of mixed-type gastric cancer from pathology images presents a formidable challenge for pathologists, given its intricate features and resemblance to other subtypes of gastric cancer. Artificial Intelligence has the potential to overcome this hurdle. This study aimed to leverage deep machine learning techniques to establish a precise and efficient diagnostic approach for this cancer type which can also predict the metastatic risk using two software, U-Net and QuPath, which have not been trialled in gastric cancers., Methods: A U-Net neural network was trained to recognise, and segment differentiated components from 186 pathology images of mixed-type gastric cancer. Undifferentiated components in the same images were annotated using the open-source pathology imaging software QuPath. The outcomes from U-Net and QuPath were used to calculate the ratios of differentiation/undifferentiated components which were correlated to lymph node metastasis., Results: The models established by U-Net recognised ∼91% of the regions of interest, with precision, recall, and F1 values of 90.2%, 90.9% and 94.6%, respectively, indicating a high level of accuracy and reliability. Furthermore, the receiver operating characteristic curve analysis showed an area under the cure of 91%, indicating good performance. A bell-curve correlation between the differentiated/undifferentiated ratio and lymphatic metastasis was found (highest risk between 0.683 and 1.03), which is paradigm-shifting., Conclusion: U-Net and QuPath exhibit promising accuracy in the identification of differentiated and undifferentiated components in mixed-type gastric cancer, as well as paradigm-shifting prediction of metastasis. These findings bring us one step closer to their potential clinical application., Competing Interests: Declaration of Competing Interest The authors declare no conflict of interest., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Rationale and Objectives: Programmed Death-Ligand 1 (PD-L1) is an important biomarker for patient selection of immunotherapy in gastric cancer (GC). This study aimed to construct and validate a non-invasive virtual biopsy system based on radiological features and clinical factors to predict the PD-L1 expression level in GC., Materials and Methods: 217 patients who received gastrectomy for GC were consecutively enrolled in this study, with 157 patients from center 1 as the training cohort and 60 patients from center 2 as the external validation cohort. 1205 quantitative radiomics features were extracted from preprocessed pre-operative contrast-enhanced CT images of enrolled patients. A radiological signature was computed using a regression random forest model and was integrated with clinical factors in a multilayer perceptron. The performance of the digital biopsy system was evaluated by the receiver operating characteristic (ROC) curve and calibration curve in both the training and validation cohort., Results: 15 features were selected for the construction of radiological signature, which was significantly associated with expression levels of PD-L1 in both the training cohort (p<0.0001) and the external validation cohort (p<0.01). The hybrid deep learning model integrating the radiological signature and clinical factor could accurately distinguish GCs with high PD-L1 expression levels in both the training cohort (AUC = 0.806, 95%CI: 0.736-0.875) and the validation cohort (AUC = 0.784, 95%CI: 0.668-0.901)., Conclusions: Our results indicate that the combination of deep learning and quantitative radiological features are potential approaches for the non-invasive evaluation of PD-L1 expression levels in GC. The digital biopsy system could provide valuable suggestive information for clinical decision-making of immunotherapy in GC., (Copyright © 2022 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.)

Objective: To evaluate the diagnostic accuracy of deep learning algorithms to identify age-related macular degeneration and to explore factors impacting the results for future model training., Methods: Diagnostic accuracy studies published in PubMed, EMBASE, the Cochrane Library, and ClinicalTrails.gov before 11 August 2022 which employed deep learning for age-related macular degeneration detection were identified and extracted by two independent researchers. Sensitivity analysis, subgroup, and meta-regression were performed by Review Manager 5.4.1, Meta-disc 1.4, and Stata 16.0. The risk of bias was assessed using QUADAS-2. The review was registered (PROSPERO CRD42022352753)., Results: The pooled sensitivity and specificity in this meta-analysis were 94% (P = 0, 95% CI 0.94-0.94, I2 = 99.7%) and 97% (P = 0, 95% CI 0.97-0.97, I2 = 99.6%), respectively. The pooled positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the area under the curve value were 21.77(95% CI 15.49-30.59), 0.06 (95% CI 0.04-0.09), 342.41 (95% CI 210.31-557.49), and 0.9925, respectively. Meta-regression indicated that types of AMD (P = 0.1882, RDOR = 36.03) and layers of the network (P = 0.4878, RDOR = 0.74) contributed to the heterogeneity., Conclusions: Convolutional neural networks are mostly adopted deep learning algorithms in age-related macular degeneration detection. Convolutional neural networks, especially ResNets, are effective in detecting age-related macular degeneration with high diagnostic accuracy. Types of age-related macular degeneration and layers of the network are the two essential factors that impact the model training process. Proper layers of the network will make the model more reliable. More datasets established by new diagnostic methods will be used to train deep learning models in the future, which will benefit for fundus application screening, long-range medical treatment, and reducing the workload of physicians., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2023 Leng et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Background and Aims: Distinguishing pancreatic cancer from nonneoplastic masses is critical and remains a clinical challenge. The study aims to construct a deep learning-based artificial intelligence system to facilitate pancreatic mass diagnosis, and to guide EUS-guided fine-needle aspiration (EUS-FNA) in real time., Methods: This is a prospective study. The CH-EUS MASTER system is composed of Model 1 (real-time capture and segmentation) and Model 2 (benign and malignant identification). It was developed using deep convolutional neural networks and Random Forest algorithm. Patients with pancreatic masses undergoing CH-EUS examinations followed by EUS-FNA were recruited. All patients underwent CH-EUS and were diagnosed both by endoscopists and CH-EUS MASTER. After diagnosis, they were randomly assigned to undergo EUS-FNA with or without CH-EUS MASTER guidance., Results: Compared with manual labeling by experts, the average overlap rate of Model 1 was 0.708. In the independent CH-EUS video testing set, Model 2 generated an accuracy of 88.9% in identifying malignant tumors. In clinical trial, the accuracy, sensitivity, and specificity for diagnosing pancreatic masses by CH-EUS MASTER were significantly better than that of endoscopists. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were respectively 93.8%, 90.9%, 100%, 100%, and 83.3% by CH-EUS MASTER guided EUS-FNA, and were not significantly different compared to the control group. CH-EUS MASTER-guided EUS-FNA significantly improved the first-pass diagnostic yield., Conclusion: CH-EUS MASTER is a promising artificial intelligence system diagnosing malignant and benign pancreatic masses and may guide FNA in real time., Trial Registration Number: NCT04607720., (© 2023 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.)

Objectives: To develop an ensemble multi-task deep learning (DL) framework for automatic and simultaneous detection, segmentation, and classification of primary bone tumors (PBTs) and bone infections based on multi-parametric MRI from multi-center., Methods: This retrospective study divided 749 patients with PBTs or bone infections from two hospitals into a training set (N = 557), an internal validation set (N = 139), and an external validation set (N = 53). The ensemble framework was constructed using T1-weighted image (T1WI), T2-weighted image (T2WI), and clinical characteristics for binary (PBTs/bone infections) and three-category (benign/intermediate/malignant PBTs) classification. The detection and segmentation performances were evaluated using Intersection over Union (IoU) and Dice score. The classification performance was evaluated using the receiver operating characteristic (ROC) curve and compared with radiologist interpretations., Result: On the external validation set, the single T1WI-based and T2WI-based multi-task models obtained IoUs of 0.71 ± 0.25/0.65 ± 0.30 for detection and Dice scores of 0.75 ± 0.26/0.70 ± 0.33 for segmentation. The framework achieved AUCs of 0.959 (95%CI, 0.955-1.000)/0.900 (95%CI, 0.773-0.100) and accuracies of 90.6% (95%CI, 79.7-95.9%)/78.3% (95%CI, 58.1-90.3%) for the binary/three-category classification. Meanwhile, for the three-category classification, the performance of the framework was superior to that of three junior radiologists (accuracy: 65.2%, 69.6%, and 69.6%, respectively) and comparable to that of two senior radiologists (accuracy: 78.3% and 78.3%)., Conclusion: The MRI-based ensemble multi-task framework shows promising performance in automatically and simultaneously detecting, segmenting, and classifying PBTs and bone infections, which was preferable to junior radiologists., Clinical Relevance Statement: Compared with junior radiologists, the ensemble multi-task deep learning framework effectively improves differential diagnosis for patients with primary bone tumors or bone infections. This finding may help physicians make treatment decisions and enable timely treatment of patients., Key Points: • The ensemble framework fusing multi-parametric MRI and clinical characteristics effectively improves the classification ability of single-modality models. • The ensemble multi-task deep learning framework performed well in detecting, segmenting, and classifying primary bone tumors and bone infections. • The ensemble framework achieves an optimal classification performance superior to junior radiologists' interpretations, assisting the clinical differential diagnosis of primary bone tumors and bone infections., (© 2023. The Author(s), under exclusive licence to European Society of Radiology.)

Ischemic stroke is a cerebrovascular disease with a high morbidity and mortality rate, which poses a serious challenge to human health and life. Meanwhile, the management of ischemic stroke remains highly dependent on manual visual analysis of noncontrast computed tomography (CT) or magnetic resonance imaging (MRI). However, artifacts and noise of the equipment as well as the radiologist experience play a significant role on diagnostic accuracy. To overcome these defects, the number of computer-aided diagnostic (CAD) methods for ischemic stroke is increasing substantially during the past decade. Particularly, deep learning models with massive data learning capabilities are recognized as powerful auxiliary tools for the acute intervention and guiding prognosis of ischemic stroke. To select appropriate interventions, facilitate clinical practice, and improve the clinical outcomes of patients, this review firstly surveys the current state-of-the-art deep learning technology. Then, we summarized the major applications in acute ischemic stroke imaging, particularly in exploring the potential function of stroke diagnosis and multimodal prognostication. Finally, we sketched out the current problems and prospects., Competing Interests: The authors declare that there is no conflict of interest regarding the publication of this paper., (Copyright © 2022 Liyuan Cui et al.)

Background and Objective: Deep learning techniques are powerful tools for image analysis. However, the lack of programming experience makes it difficult for novice users to apply this technology. This project aims to lower the barrier for clinical users to implement deep learning methods in microscopic image classification., Methods: In this study, an out-of-the-box software, AIMIC (artificial intelligence-based microscopy image classifier), was developed for users to apply deep learning technology in a code-free manner. The platform was equipped with state-of-the-art deep learning techniques and data preprocessing approaches. Furthermore, we evaluated the built-in networks on four benchmark microscopy image datasets to assist entry-level practitioners in selecting a suitable algorithm., Results: The entire deep learning pipeline, from training a new network to inferring unseen samples using the trained model, could be implemented on the proposed platform without the need for programming. In the evaluation experiments, the ResNeXt-50-32×4d outperformed other competitor algorithms in terms of average accuracy (96.83%) and average F1-score (96.82%). In addition, the MobileNet-V2 achieved a good balance between the performance (accuracy of 95.72%) and computational cost (inference time of 0.109s for identifying one sample)., Conclusions: The proposed AI platform allows people without programming experience to use artificial intelligence methods in microscopy image analysis. Besides, the ResNeXt-50-32×4d is a preferable solution for microscopic image classification, and MobileNet-V2 is most likely to be an alternative selection for the scenario when computing resources are limited., Competing Interests: Declaration of Competing Interest The authors declare that there are no conflicts of interest., (Copyright © 2022 Elsevier B.V. All rights reserved.)

This paper proposes corresponding teaching methods and instructional modes based on predecessors' research on mathematics instructional mode and the current state of mathematics teaching. In addition, this paper constructs a teaching evaluation model based on DL algorithm based on an in-depth study of DL-related theories in order to accurately and scientifically analyze the problems that exist in mathematics teaching. This paper constructs an instructional quality evaluation index system based on rationality and fairness, and uses the BPNN evaluation model to train and study a set of instructional quality data. Finally, the experimental results show that this system has a high level of stability, with a 96.37 percent stability rate and a 95.42 percent evaluation accuracy rate. The results of this paper's evaluation of the mathematical instructional quality model are objective and reasonable. It can accurately assess instructional quality while also assessing problems in the teaching process based on the instructional quality scores and making reasonable recommendations for teaching improvement based on the weak links in the teaching process. It has the potential to provide a workable system for assessing instructional quality., Competing Interests: The author does not have any possible conflicts of interest., (Copyright © 2022 Rui Liu.)

Purpose: We aimed to predict the prognosis of advanced nasopharyngeal carcinoma (stage Ⅲ-Ⅳa) using Pre- and Post-treatment MR images based on deep learning (DL)., Methods: A total of 206 patients with primary nasopharyngeal carcinoma who were diagnosed and treated at the Renmin Hospital of Wuhan University between June 2012 and January 2018 were retrospectively selected. A rectangular region of interest (ROI), which included the tumor area, surrounding tissues and organs, was delineated on each Pre- and Post-treatment MR image. Two Inception-Resnet-V2 based transfer learning models, named Pre-model and Post-model, were trained with the Pre-treatment images and the Post-treatment images, respectively. In addition, an ensemble learning model based on the Pre-model and Post-models was established. The three established models were evaluated by receiver operating characteristic curve (ROC), confusion matrix, and Harrell's concordance indices (C-index). High-risk-related gradient-weighted class activation mapping (Grad-CAM) images were developed according to the DL models., Results: The Pre-model, Post-model, and ensemble model displayed a C-index of 0.717 (95% CI: 0.639 to 0.795), 0.811 (95% CI: 0.745-0.877), 0.830 (95% CI: 0.767-0.893), and AUC of 0.741 (95% CI: 0.584-0.900), 0.806 (95% CI: 0.670-0.942), and 0.842 (95% CI: 0.718-0.967) for the test cohort, respectively. In comparison with the models, the performance of Post-model was better than the performance of Pre-model, which indicated the importance of Post-treatment images for prognosis prediction. All three DL models performed better than the TNM staging system (0.723, 95% CI: 0.567-0.879). The captured features presented on Grad-CAM images suggested that the areas around the tumor and lymph nodes were related to the prognosis of the tumor., Conclusions: The three established DL models based on Pre- and Post-treatment MR images have a better performance than TNM staging. Post-treatment MR images are of great significance for prognosis prediction and could contribute to clinical decision-making., Competing Interests: Declaration of Competing Interest All authors declared that there were no competing interests., (Copyright © 2022 The Authors. Published by Elsevier B.V. All rights reserved.)

This study explores the memory characteristics of elderly individuals to design effective smart devices based on smart memory storage solutions under deep learning to improve the learning efficiency of elderly individuals. The different memory formation stages in the existing human brain are analysed. A smart memory storage solution based on memory-enhanced embedded learning is constructed based on meta-learning under deep learning, which reduces the cost of learning new tasks to the greatest extent. Finally, the performance of the proposed solution is verified using different datasets. The results reveal that the solution based on deep learning has obvious effects on different datasets, with an average accuracy rate of 99.7%. By synthesizing a large number of target sample features, this solution can lower the learning difficulty and improve the learning effect. The proposed elderly oriented smart device effectively reduces the shortcomings in the current market and lowers the learning difficulty, which provides an important reference for further enriching devices in the ageing market., Competing Interests: The authors declare that they have no conflicts of interest., (Copyright © 2022 Chen Gong et al.)

The excessive content of additives in food is a radical problem that affects human health. However, traditional chemical methods are limited by a long cycle, low accuracy, and strong destructiveness, so a fast and accurate alternative is urgently needed. This paper proposes a prediction model introducing near-infrared spectroscopy and deep learning to perform fast and accurate non-destructive detection of artificial bright blue pigment in cream. The model results show that R 2 is 0.9638, RMSEP is 0.0157, and RPD is 4.4022. In the preprocessing part, this paper compares the traditional preprocessing methods (SNV, MSC, SG) horizontally and innovatively proposes the use of autoencoders to mitigate the dimensionality of data, which has immensely improved the follow-up prediction effect. In addition, it tries to perform regression prediction on spectral data and establish a fully connected convolutional neural network model through deep learning, whose result indicators prove better than those of traditional methods such as PLSR and MLR. When constructing the deep learning model, this paper applies knowledge evolution to compress the model to achieve a lower calculation cost and higher accuracy. Compared with the traditional methods, the model proposed in this paper has greater accuracy and higher speed with samples undamaged, which is worth popularizing., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Objectives: To build and validate deep learning and machine learning fusion models to classify benign, malignant, and intermediate bone tumors based on patient clinical characteristics and conventional radiographs of the lesion., Methods: In this retrospective study, data were collected with pathologically confirmed diagnoses of bone tumors between 2012 and 2019. Deep learning and machine learning fusion models were built to classify tumors as benign, malignant, or intermediate using conventional radiographs of the lesion and potentially relevant clinical data. Five radiologists compared diagnostic performance with and without the model. Diagnostic performance was evaluated using the area under the curve (AUC)., Results: A total of 643 patients' (median age, 21 years; interquartile range, 12-38 years; 244 women) 982 radiographs were included. In the test set, the binary category classification task, the radiological model of classification for benign/not benign, malignant/nonmalignant, and intermediate/not intermediate had AUCs of 0.846, 0.827, and 0.820, respectively; the fusion models had an AUC of 0.898, 0.894, and 0.865, respectively. In the three-category classification task, the radiological model achieved a macro average AUC of 0.813, and the fusion model had a macro average AUC of 0.872. In the observation test, the mean macro average AUC of all radiologists was 0.819. With the three-category classification fusion model support, the macro AUC improved by 0.026., Conclusion: We built, validated, and tested deep learning and machine learning models that classified bone tumors at a level comparable with that of senior radiologists. Model assistance may somewhat help radiologists' differential diagnoses of bone tumors., Key Points: • The deep learning model can be used to classify benign, malignant, and intermediate bone tumors. • The machine learning model fusing information from radiographs and clinical characteristics can improve the classification capacity for bone tumors. • The diagnostic performance of the fusion model is comparable with that of senior radiologists and is potentially useful as a complement to radiologists in a bone tumor differential diagnosis., (© 2021. European Society of Radiology.)

Purpose: Luminance contrast is the fundamental building block of human spatial vision. Therefore contrast sensitivity, the reciprocal of contrast threshold required for target detection, has been a barometer of human visual function. Although retinal ganglion cells (RGCs) are known to be involved in contrast coding, it still remains unknown whether the retinal layers containing RGCs are linked to a person's contrast sensitivity (e.g., Pelli-Robson contrast sensitivity) and, if so, to what extent the retinal layers are related to behavioral contrast sensitivity. Thus the current study aims to identify the retinal layers and features critical for predicting a person's contrast sensitivity via deep learning., Methods: Data were collected from 225 subjects including individuals with either glaucoma, age-related macular degeneration, or normal vision. A deep convolutional neural network trained to predict a person's Pelli-Robson contrast sensitivity from structural retinal images measured with optical coherence tomography was used. Then, activation maps that represent the critical features learned by the network for the output prediction were computed., Results: The thickness of both ganglion cell and inner plexiform layers, reflecting RGC counts, were found to be significantly correlated with contrast sensitivity (r = 0.26 ∼ 0.58, Ps < 0.001 for different eccentricities). Importantly, the results showed that retinal layers containing RGCs were the critical features the network uses to predict a person's contrast sensitivity (an average R2 = 0.36 ± 0.10)., Conclusions: The findings confirmed the structure and function relationship for contrast sensitivity while highlighting the role of RGC density for human contrast sensitivity.

Purpose: To develop a deep learning-based reconstruction framework for ultrafast and robust diffusion tensor imaging and fiber tractography., Methods: SuperDTI was developed to learn the nonlinear relationship between DWIs and the corresponding diffusion tensor parameter maps. It bypasses the tensor fitting procedure, which is highly susceptible to noises and motions in DWIs. The network was trained and tested using data sets from the Human Connectome Project and patients with ischemic stroke. Results from SuperDTI were compared against widely used methods for tensor parameter estimation and fiber tracking., Results: Using training and testing data acquired using the same protocol and scanner, SuperDTI was shown to generate fractional anisotropy and mean diffusivity maps, as well as fiber tractography, from as few as six raw DWIs, with a quantification error of less than 5% in all white-matter and gray-matter regions of interest. It was robust to noises and motions in the testing data. Furthermore, the network trained using healthy volunteer data showed no apparent reduction in lesion detectability when directly applied to stroke patient data., Conclusions: Our results demonstrate the feasibility of superfast DTI and fiber tractography using deep learning with as few as six DWIs directly, bypassing tensor fitting. Such a significant reduction in scan time may allow the inclusion of DTI into the clinical routine for many potential applications., (© 2021 International Society for Magnetic Resonance in Medicine.)

Deep learning offers new approaches to investigate the mechanisms underlying complex biological phenomena, such as subgenome dominance. Subgenome dominance refers to the dominant expression and/or biased fractionation of genes in one subgenome of allopolyploids, which has shaped the evolution of a large group of plants. However, the underlying cause of subgenome dominance remains elusive. Here, we adopt deep learning to construct two convolutional neural network (CNN) models, binary expression model (BEM) and homoeolog contrast model (HCM), to investigate the mechanism underlying subgenome dominance using DNA sequence and methylation sites. We apply these CNN models to analyze three representative polyploidization systems, Brassica, Gossypium, and Cucurbitaceae, each with available ancient and neo/synthetic polyploidized genomes. The BEM shows that DNA sequence of the promoter region can accurately predict whether a gene is expressed or not. More importantly, the HCM shows that the DNA sequence of the promoter region predicts dominant expression status between homoeologous gene pairs retained from ancient polyploidizations, thus predicting subgenome dominance associated with these events. However, HCM fails to predict gene expression dominance between new homoeologous gene pairs arising from the neo/synthetic polyploidizations. These results are consistent across the three plant polyploidization systems, indicating broad applicability of our models. Furthermore, the two models based on methylation sites produce similar results. These results show that subgenome dominance is associated with long-term sequence differentiation between the promoters of homoeologs, suggesting that subgenome expression dominance precedes and is the driving force or even the determining factor for sequence divergence between subgenomes following polyploidization., (© 2024 The Author(s). The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd.)

Photoacoustic computed tomography (PACT) has centimeter-level imaging ability and can be used to detect the human body. However, strong photoacoustic signals from skin cover deep tissue information, hindering the frontal display and analysis of photoacoustic images of deep regions of interest. Therefore, we propose a 2.5 D deep learning model based on feature pyramid structure and single-type skin annotation to extract the skin region, and design a mask generation algorithm to remove skin automatically. PACT imaging experiments on the human periphery blood vessel verified the correctness our proposed skin-removal method. Compared with previous studies, our method exhibits high robustness to the uneven illumination, irregular skin boundary, and reconstruction artifacts in the images, and the reconstruction errors of PACT images decreased by 20% ~ 90% with a 1.65 dB improvement in the signal-to-noise ratio at the same time. This study may provide a promising way for high-definition PACT imaging of deep tissues., (© 2024 Wiley‐VCH GmbH.)

In recent years, significant advancements have been made in the field of smart greenhouses, particularly in the application of computer vision and robotics for pollinating flowers. Robotic pollination offers several benefits, including reduced labor requirements and preservation of costly pollen through artificial tomato pollination. However, previous studies have primarily focused on the labeling and detection of tomato flowers alone. Therefore, the objective of this study was to develop a comprehensive methodology for simultaneously labeling, training, and detecting tomato flowers specifically tailored for robotic pollination. To achieve this, transfer learning techniques were employed using well-known models, namely YOLOv5 and the recently introduced YOLOv8, for tomato flower detection. The performance of both models was evaluated using the same image dataset, and a comparison was made based on their Average Precision (AP) scores to determine the superior model. The results indicated that YOLOv8 achieved a higher mean AP (mAP) of 92.6% in tomato flower and bud detection, outperforming YOLOv5 with 91.2%. Notably, YOLOv8 also demonstrated an inference speed of 0.7 ms when considering an image size of 1920 × 1080 pixels resized to 640 × 640 pixels during detection. The image dataset was acquired during both morning and evening periods to minimize the impact of lighting conditions on the detection model. These findings highlight the potential of YOLOv8 for real-time detection of tomato flowers and buds, enabling further estimation of flower blooming peaks and facilitating robotic pollination. In the context of robotic pollination, the study also focuses on the deployment of the proposed detection model on the 3P2R gantry robot. The study introduces a kinematic model and a modified circuit for the gantry robot. The position-based visual servoing method is employed to approach the detected flower during the pollination process. The effectiveness of the proposed visual servoing approach is validated in both un-clustered and clustered plant environments in the laboratory setting. Additionally, this study provides valuable theoretical and practical insights for specialists in the field of greenhouse systems, particularly in the design of flower detection algorithms using computer vision and its deployment in robotic systems used in greenhouses., (© 2024. The Author(s).)

The neural network algorithm of deep learning was applied to optimize and improve color Doppler ultrasound images, which was used for the research on elderly patients with chronic heart failure (CHF) complicated with sarcopenia, so as to analyze the effect of the deep-learning-based color Doppler ultrasound image on the diagnosis of CHF. 259 patients were selected randomly in this study, who were admitted to hospital from October 2017 to March 2020 and were diagnosed with sarcopenia. Then, all of them underwent cardiac ultrasound examination and were divided into two groups according to whether deep learning technology was used for image processing or not. A group of routine unprocessed images was set as the control group, and the images processed by deep learning were set as the experimental group. The results of color Doppler images before and after processing were analyzed and compared; that is, the processed images of the experimental group were clearer and had higher resolution than the unprocessed images of the control group, with the peak signal-to-noise ratio (PSNR) = 20 and structural similarity index measure (SSIM) = 0.09; the similarity between the final diagnosis results and the examination results of the experimental group (93.5%) was higher than that of the control group (87.0%), and the comparison was statistically significant ( P < 0.05); among all the patients diagnosed with sarcopenia, 88.9% were also eventually diagnosed with CHF and only a small part of them were diagnosed with other diseases, with statistical significance ( P < 0.05). In conclusion, deep learning technology had certain application value in processing color Doppler ultrasound images. Although there was no obvious difference between the color Doppler ultrasound images before and after processing, they could all make a better diagnosis. Moreover, the research results showed the correlation between CHF and sarcopenia., Competing Interests: The authors declare no conflicts of interest., (Copyright © 2021 Peng Bian et al.)

Retinal screening contributes to early detection of diabetic retinopathy and timely treatment. To facilitate the screening process, we develop a deep learning system, named DeepDR, that can detect early-to-late stages of diabetic retinopathy. DeepDR is trained for real-time image quality assessment, lesion detection and grading using 466,247 fundus images from 121,342 patients with diabetes. Evaluation is performed on a local dataset with 200,136 fundus images from 52,004 patients and three external datasets with a total of 209,322 images. The area under the receiver operating characteristic curves for detecting microaneurysms, cotton-wool spots, hard exudates and hemorrhages are 0.901, 0.941, 0.954 and 0.967, respectively. The grading of diabetic retinopathy as mild, moderate, severe and proliferative achieves area under the curves of 0.943, 0.955, 0.960 and 0.972, respectively. In external validations, the area under the curves for grading range from 0.916 to 0.970, which further supports the system is efficient for diabetic retinopathy grading.

Microfluidic technologies have long enabled the manipulation of flow-driven cells en masse under a variety of force fields with the goal of characterizing them or discriminating the pathogenic ones. On the other hand, a microfluidic platform is typically designed to function under optimized conditions, which rarely account for specimen heterogeneity and internal/external perturbations. In this work, we demonstrate a proof-of-principle adaptive microfluidic system that consists of an integrated network of distributed electrical sensors for on-chip tracking of cells and closed-loop feedback control that modulates chip parameters based on the sensor data. In our system, cell flow speed is measured at multiple locations throughout the device, the data is interpreted in real-time via deep learning-based algorithms, and a proportional-integral feedback controller updates a programmable pressure pump to maintain a desired cell flow speed. We validate the adaptive microfluidic system with both static and dynamic targets and also observe a fast convergence of the system under continuous external perturbations. With an ability to sustain optimal processing conditions in unsupervised settings, adaptive microfluidic systems would be less prone to artifacts and could eventually serve as reliable standardized biomedical tests at the point of care.

Background: Accurate and robust pathological image analysis for colorectal cancer (CRC) diagnosis is time-consuming and knowledge-intensive, but is essential for CRC patients' treatment. The current heavy workload of pathologists in clinics/hospitals may easily lead to unconscious misdiagnosis of CRC based on daily image analyses., Methods: Based on a state-of-the-art transfer-learned deep convolutional neural network in artificial intelligence (AI), we proposed a novel patch aggregation strategy for clinic CRC diagnosis using weakly labeled pathological whole-slide image (WSI) patches. This approach was trained and validated using an unprecedented and enormously large number of 170,099 patches, > 14,680 WSIs, from > 9631 subjects that covered diverse and representative clinical cases from multi-independent-sources across China, the USA, and Germany., Results: Our innovative AI tool consistently and nearly perfectly agreed with (average Kappa statistic 0.896) and even often better than most of the experienced expert pathologists when tested in diagnosing CRC WSIs from multicenters. The average area under the receiver operating characteristics curve (AUC) of AI was greater than that of the pathologists (0.988 vs 0.970) and achieved the best performance among the application of other AI methods to CRC diagnosis. Our AI-generated heatmap highlights the image regions of cancer tissue/cells., Conclusions: This first-ever generalizable AI system can handle large amounts of WSIs consistently and robustly without potential bias due to fatigue commonly experienced by clinical pathologists. It will drastically alleviate the heavy clinical burden of daily pathology diagnosis and improve the treatment for CRC patients. This tool is generalizable to other cancer diagnosis based on image recognition.

Purpose: To develop a deep learning-based approach to reduce the scan time of multipool CEST MRI for Parkinson's disease (PD) while maintaining sufficient prediction accuracy., Method: A deep learning approach based on a modified one-dimensional U-Net, termed Z-spectral compressed sensing (CS), was proposed to recover dense Z-spectra from sparse ones. The neural network was trained using simulated Z-spectra generated by the Bloch equation with various parameter settings. Its feasibility and effectiveness were validated through numerical simulations and in vivo rat brain experiments, compared with commonly used linear, pchip, and Lorentzian interpolation methods. The proposed method was applied to detect metabolism-related changes in the 6-hydroxydopamine PD model with multipool CEST MRI, including APT, CEST@2 ppm, nuclear Overhauser enhancement, direct saturation, and magnetization transfer, and the prediction performance was evaluated by area under the curve., Results: The numerical simulations and in vivo rat-brain experiments demonstrated that the proposed method could yield superior fidelity in retrieving dense Z-spectra compared with existing methods. Significant differences were observed in APT, CEST@2 ppm, nuclear Overhauser enhancement, and direct saturation between the striatum regions of wild-type and PD models, whereas magnetization transfer exhibited no significant difference. Receiver operating characteristic analysis demonstrated that multipool CEST achieved better predictive performance compared with individual pools. Combined with Z-spectral CS, the scan time of multipool CEST MRI can be reduced to 33% without distinctly compromising prediction accuracy., Conclusion: The integration of Z-spectral CS with multipool CEST MRI can enhance the prediction accuracy of PD and maintain the scan time within a reasonable range., (© 2024 International Society for Magnetic Resonance in Medicine.)

Chemical exchange saturation transfer (CEST) MRI at 3 T suffers from low specificity due to overlapping CEST effects from multiple metabolites, while higher field strengths (B 0 ) allow for better separation of Z-spectral "peaks," aiding signal interpretation and quantification. However, data acquisition at higher B 0 is restricted by equipment access, field inhomogeneity and safety issues. Herein, we aim to synthesize higher-B 0 Z-spectra from readily available data acquired with 3 T clinical scanners using a deep learning framework. Trained with simulation data using models based on Bloch-McConnell equations, this framework comprised two deep neural networks (DNNs) and a singular value decomposition (SVD) module. The first DNN identified B 0 shifts in Z-spectra and aligned them to correct frequencies. After B 0 correction, the lower-B 0 Z-spectra were streamlined to the second DNN, casting into the key feature representations of higher-B 0 Z-spectra, obtained through SVD truncation. Finally, the complete higher-B 0 Z-spectra were recovered from inverse SVD, given the low-rank property of Z-spectra. This study constructed and validated two models, a phosphocreatine (PCr) model and a pseudo-in-vivo one. Each experimental dataset, including PCr phantoms, egg white phantoms, and in vivo rat brains, was sequentially acquired on a 3 T human and a 9.4 T animal scanner. Results demonstrated that the synthetic 9.4 T Z-spectra were almost identical to the experimental ground truth, showing low RMSE (0.11% ± 0.0013% for seven PCr tubes, 1.8% ± 0.2% for three egg white tubes, and 0.79% ± 0.54% for three rat slices) and high R 2 (>0.99). The synthesized amide and NOE contrast maps, calculated using the Lorentzian difference, were also well matched with the experiments. Additionally, the synthesis model exhibited robustness to B 0 inhomogeneities, noise, and other acquisition imperfections. In conclusion, the proposed framework enables synthesis of higher-B 0 Z-spectra from lower-B 0 ones, which may facilitate CEST MRI quantification and applications., (© 2024 John Wiley & Sons Ltd.)

Objectives/hypothesis: To develop a deep-learning-based computer-aided diagnosis system for distinguishing laryngeal neoplasms (benign, precancerous lesions, and cancer) and improve the clinician-based accuracy of diagnostic assessments of laryngoscopy findings., Study Design: Retrospective study., Methods: A total of 24,667 laryngoscopy images (normal, vocal nodule, polyps, leukoplakia and malignancy) were collected to develop and test a convolutional neural network (CNN)-based classifier. A comparison between the proposed CNN-based classifier and the clinical visual assessments (CVAs) by 12 otolaryngologists was conducted., Results: In the independent testing dataset, an overall accuracy of 96.24% was achieved; for leukoplakia, benign, malignancy, normal, and vocal nodule, the sensitivity and specificity were 92.8% vs. 98.9%, 97% vs. 99.7%, 89% vs. 99.3%, 99.0% vs. 99.4%, and 97.2% vs. 99.1%, respectively. Furthermore, when compared with CVAs on the randomly selected test dataset, the CNN-based classifier outperformed physicians for most laryngeal conditions, with striking improvements in the ability to distinguish nodules (98% vs. 45%, P < .001), polyps (91% vs. 86%, P < .001), leukoplakia (91% vs. 65%, P < .001), and malignancy (90% vs. 54%, P < .001)., Conclusions: The CNN-based classifier can provide a valuable reference for the diagnosis of laryngeal neoplasms during laryngoscopy, especially for distinguishing benign, precancerous, and cancer lesions., Level of Evidence: NA Laryngoscope, 130:E686-E693, 2020., (© 2020 The American Laryngological, Rhinological and Otological Society, Inc.)

Shortness of breath is a major reason that patients present to the emergency department (ED) and point-of-care ultrasound (POCUS) has been shown to aid in diagnosis, particularly through evaluation for artifacts known as B-lines. B-line identification and quantification can be a challenging skill for novice ultrasound users, and experienced users could benefit from a more objective measure of quantification. We sought to develop and test a deep learning (DL) algorithm to quantify the assessment of B-lines in lung ultrasound. We utilized ultrasound clips ( n = 400 ) from an existing database of ED patients to provide training and test sets to develop and test the DL algorithm based on deep convolutional neural networks. Interpretations of the images by algorithm were compared to expert human interpretations on binary and severity (a scale of 0-4) classifications. Our model yielded a sensitivity of 93% (95% confidence interval (CI) 81%-98%) and a specificity of 96% (95% CI 84%-99%) for the presence or absence of B-lines compared to expert read, with a kappa of 0.88 (95% CI 0.79-0.97). Model to expert agreement for severity classification yielded a weighted kappa of 0.65 (95% CI 0.56-074). Overall, the DL algorithm performed well and could be integrated into an ultrasound system in order to help diagnose and track B-line severity. The algorithm is better at distinguishing the presence from the absence of B-lines but can also be successfully used to distinguish between B-line severity. Such methods could decrease variability and provide a standardized method for improved diagnosis and outcome.

Aims/hypothesis: Corneal confocal microscopy is a rapid non-invasive ophthalmic imaging technique that identifies peripheral and central neurodegenerative disease. Quantification of corneal sub-basal nerve plexus morphology, however, requires either time-consuming manual annotation or a less-sensitive automated image analysis approach. We aimed to develop and validate an artificial intelligence-based, deep learning algorithm for the quantification of nerve fibre properties relevant to the diagnosis of diabetic neuropathy and to compare it with a validated automated analysis program, ACCMetrics., Methods: Our deep learning algorithm, which employs a convolutional neural network with data augmentation, was developed for the automated quantification of the corneal sub-basal nerve plexus for the diagnosis of diabetic neuropathy. The algorithm was trained using a high-end graphics processor unit on 1698 corneal confocal microscopy images; for external validation, it was further tested on 2137 images. The algorithm was developed to identify total nerve fibre length, branch points, tail points, number and length of nerve segments, and fractal numbers. Sensitivity analyses were undertaken to determine the AUC for ACCMetrics and our algorithm for the diagnosis of diabetic neuropathy., Results: The intraclass correlation coefficients for our algorithm were superior to those for ACCMetrics for total corneal nerve fibre length (0.933 vs 0.825), mean length per segment (0.656 vs 0.325), number of branch points (0.891 vs 0.570), number of tail points (0.623 vs 0.257), number of nerve segments (0.878 vs 0.504) and fractals (0.927 vs 0.758). In addition, our proposed algorithm achieved an AUC of 0.83, specificity of 0.87 and sensitivity of 0.68 for the classification of participants without (n = 90) and with (n = 132) neuropathy (defined by the Toronto criteria)., Conclusions/interpretation: These results demonstrated that our deep learning algorithm provides rapid and excellent localisation performance for the quantification of corneal nerve biomarkers. This model has potential for adoption into clinical screening programmes for diabetic neuropathy., Data Availability: The publicly shared cornea nerve dataset (dataset 1) is available at http://bioimlab.dei.unipd.it/Corneal%20Nerve%20Tortuosity%20Data%20Set.htm and http://bioimlab.dei.unipd.it/Corneal%20Nerve%20Data%20Set.htm.

Deep learning models, such as deep convolutional neural network and deep long-short term memory model, have achieved great successes in many pattern classification applications over shadow machine learning models with hand-crafted features. The main reason is the ability of deep learning models to automatically extract hierarchical features from massive data by multiple layers of neurons. However, in many other situations, existing deep learning models still cannot gain satisfying results due to the limitation of the inputs of models. The existing deep learning models only take the data instances of an input point but completely ignore the other data points in the dataset, which potentially provides critical insight for the classification of the given input. To overcome this gap, in this paper, we show that the neighboring data points besides the input data point itself can boost the deep learning model's performance significantly and design a novel deep learning model which takes both the data instances of an input point and its neighbors' classification responses as inputs. In addition, we develop an iterative algorithm which updates the neighbors of data points according to the deep representations output by the deep learning model and the parameters of the deep learning model alternately. The proposed algorithm, named "Iterative Deep Neighborhood (IDN)," shows its advantages over the state-of-the-art deep learning models over tasks of image classification, text sentiment analysis, property price trend prediction, etc., Competing Interests: The authors declare that they have no conflicts of interest., (Copyright © 2020 Rong Liu et al.)

Objectives: Our study aims to investigate the impact of B-mode ultrasound (B-US) imaging, color Doppler flow imaging (CDFI), strain elastography (SE), and patient age on the prediction of molecular subtypes in breast lesions., Methods: Totally 2272 multimodal ultrasound imaging was collected from 198 patients. The ResNet-18 network was employed to predict four molecular subtypes from B-US imaging, CDFI, and SE of patients with different ages. All the images were split into training and testing datasets by the ratio of 80%:20%. The predictive performance on testing dataset was evaluated through 5 metrics including mean accuracy, precision, recall, F1-scores, and confusion matrix., Results: Based on B-US imaging, the test mean accuracy is 74.50%, the precision is 74.84%, the recall is 72.48%, and the F1-scores is 0.73. By combining B-US imaging with CDFI, the results were increased to 85.41%, 85.03%, 85.05%, and 0.84, respectively. With the integration of B-US imaging and SE, the results were changed to 75.64%, 74.69%, 73.86%, and 0.74, respectively. Using images from patients under 40 years old, the results were 90.48%, 90.88%, 88.47%, and 0.89. When images from patients who are above 40 years old, they were changed to 81.96%, 83.12%, 80.5%, and 0.81, respectively., Conclusion: Multimodal ultrasound imaging can be used to accurately predict the molecular subtypes of breast lesions. In addition to B-US imaging, CDFI rather than SE contribute further to improve predictive performance. The predictive performance is notably better for patients under 40 years old compared with those who are 40 years old and above., (© 2024 American Institute of Ultrasound in Medicine.)

Purpose: To develop a highly accelerated CEST Z-spectral acquisition method using a specifically-designed k-space sampling pattern and corresponding deep-learning-based reconstruction., Methods: For k-space down-sampling, a customized pattern was proposed for CEST, with the randomized probability following a frequency-offset-dependent (FOD) function in the direction of saturation offset. For reconstruction, the convolution network (CNN) was enhanced with a Partially Separable (PS) function to optimize the spatial domain and frequency domain separately. Retrospective experiments on a self-acquired human brain dataset (13 healthy adults and 15 brain tumor patients) were conducted using k-space resampling. The prospective performance was also assessed on six healthy subjects., Results: In retrospective experiments, the combination of FOD sampling and PS network (FOD + PSN) showed the best quantitative metrics for reconstruction, outperforming three other combinations of conventional sampling with varying density and a regular CNN (nMSE and SSIM, p < 0.001 for healthy subjects). Across all acceleration factors from 4 to 14, the FOD + PSN approach consistently outperformed the comparative methods in four contrast maps including MTR asym , MTR rex , as well as the Lorentzian Difference maps of amide and nuclear Overhauser effect (NOE). In the subspace replacement experiment, the error distribution demonstrated the denoising benefits achieved in the spatial subspace. Finally, our prospective results obtained from healthy adults and brain tumor patients (14×) exhibited the initial feasibility of our method, albeit with less accurate reconstruction than retrospective ones., Conclusion: The combination of FOD sampling and PSN reconstruction enabled highly accelerated CEST MRI acquisition, which may facilitate CEST metabolic MRI for brain tumor patients., (© 2024 International Society for Magnetic Resonance in Medicine.)