Search

This research introduces a Computer-Aided Diagnosis-system designed aimed at automated detections & classification of tomato leaf diseases, combining traditional handcrafted features with advanced deep learning techniques. The system's process encompasses preprocessing, feature extraction, feature fusion, and classification. It utilizes enhancement filters and segmentation algorithms to isolate with Regions-of-Interests (ROI) in images tomato leaves. These features based arranged in ABCD rule (Asymmetry, Borders, Colors, and Diameter) are integrated with outputs from a Convolutional Neural Network (CNN) pretrained on ImageNet. To address data imbalance, we introduced a novel evaluation method that has shown to improve classification accuracy by 15% compared to traditional methods, achieving an overall accuracy rate of 92% in field tests. By merging classical feature engineering with modern machine learning techniques under mutual information-based feature fusion, our system sets a new standard for precision in agricultural diagnostics. Specific performance metrics showcasing the effectiveness of our approach in automated detection and classifying of tomato leaf disease. [ABSTRACT FROM AUTHOR]

We introduce an advanced model for predicting protein–ligand interactions. Our approach combines the strengths of graph neural networks with physics-based scoring methods. Existing structure-based machine-learning models for protein–ligand binding prediction often fall short in practical virtual screening scenarios, hindered by the intricacies of binding poses, the chemical diversity of drug-like molecules, and the scarcity of crystallographic data for protein–ligand complexes. To overcome the limitations of existing machine learning-based prediction models, we propose a novel approach that fuses three independent neural network models. One classification model is designed to perform binary prediction of a given protein–ligand complex pose. The other two regression models are trained to predict the binding affinity and root-mean-square deviation of a ligand conformation from an input complex structure. We trained the model to account for both deviations in experimental and predicted binding affinities and pose prediction uncertainties. By effectively integrating the outputs of the triplet neural networks with a physics-based scoring function, our model showed a significantly improved performance in hit identification. The benchmark results with three independent decoy sets demonstrate that our model outperformed existing models in forward screening. Our model achieved top 1% enrichment factors of 32.7 and 23.1 with the CASF2016 and DUD-E benchmark sets, respectively. The benchmark results using the LIT-PCBA set further confirmed its higher average enrichment factors, emphasizing the model's efficiency and generalizability. The model's efficiency was further validated by identifying 23 active compounds from 63 candidates in experimental screening for autotaxin inhibitors, demonstrating its practical applicability in hit discovery. Scientific contribution Our work introduces a novel training strategy for a protein–ligand binding affinity prediction model by integrating the outputs of three independent sub-models and utilizing expertly crafted decoy sets. The model showcases exceptional performance across multiple benchmarks. The high enrichment factors in the LIT-PCBA benchmark demonstrate its potential to accelerate hit discovery. [ABSTRACT FROM AUTHOR]

This study introduces OpenMAP‐T1, a deep‐learning‐based method for rapid and accurate whole‐brain parcellation in T1‐ weighted brain MRI, which aims to overcome the limitations of conventional normalization‐to‐atlas‐based approaches and multi‐atlas label‐fusion (MALF) techniques. Brain image parcellation is a fundamental process in neuroscientific and clinical research, enabling a detailed analysis of specific cerebral regions. Normalization‐to‐atlas‐based methods have been employed for this task, but they face limitations due to variations in brain morphology, especially in pathological conditions. The MALF techniques improved the accuracy of the image parcellation and robustness to variations in brain morphology, but at the cost of high computational demand that requires a lengthy processing time. OpenMAP‐T1 integrates several convolutional neural network models across six phases: preprocessing; cropping; skull‐stripping; parcellation; hemisphere segmentation; and final merging. This process involves standardizing MRI images, isolating the brain tissue, and parcellating it into 280 anatomical structures that cover the whole brain, including detailed gray and white matter structures, while simplifying the parcellation processes and incorporating robust training to handle various scan types and conditions. The OpenMAP‐T1 was validated on the Johns Hopkins University atlas library and eight available open resources, including real‐world clinical images, and the demonstration of robustness across different datasets with variations in scanner types, magnetic field strengths, and image processing techniques, such as defacing. Compared with existing methods, OpenMAP‐T1 significantly reduced the processing time per image from several hours to less than 90 s without compromising accuracy. It was particularly effective in handling images with intensity inhomogeneity and varying head positions, conditions commonly seen in clinical settings. The adaptability of OpenMAP‐T1 to a wide range of MRI datasets and its robustness to various scan conditions highlight its potential as a versatile tool in neuroimaging. [ABSTRACT FROM AUTHOR]

The aurora arc is a separate auroral structure from the aurora oval, whose location and morphology are related to various solar-terrestrial circumstances. However, because of the low occurring frequency of aurora arc and the lack of the automatic identification technique, it can only be manually distinguished from a huge number of observed images, which is very inefficient. In order to improve the identification efficiency, we propose an identification algorithm based on YOLOX network and Convolutional Block Attention Module attention mechanism. Using the aurora images observed by Special Sensor Ultraviolet Spectrographic Imager carried by the Defense Meteorological Satellite Program F16-F19 satellites from 2013 to 2019, the automatic detection models for global and local areas were trained separately. The identification outputs will be integrated by calculating the intersection. According to the test results, the event identification precision is 86% and the position identification precision is 79%, both of which are greater than the results before integration. Therefore, the proposed method is not only able to identify whether the image contains the aurora arcs, but also accurately locate them, making it a highly effective tool for the advancement of future study. [ABSTRACT FROM AUTHOR]

RNA 5-methyluridine (m5U) sites play a significant role in understanding RNA modifications, which influence numerous biological processes such as gene expression and cellular functioning. Consequently, the identification of m5U sites can play a vital role in the integrity, structure, and function of RNA molecules. Therefore, this study introduces GRUpred-m5U, a novel deep learning-based framework based on a gated recurrent unit in mature RNA and full transcript RNA datasets. We used three descriptor groups: nucleic acid composition, pseudo nucleic acid composition, and physicochemical properties, which include five feature extraction methods ENAC, Kmer, DPCP, DPCP type 2, and PseDNC. Initially, we aggregated all the feature extraction methods and created a new merged set. Three hybrid models were developed employing deep-learning methods and evaluated through 10-fold cross-validation with seven evaluation metrics. After a comprehensive evaluation, the GRUpred-m5U model outperformed the other applied models, obtaining 98.41% and 96.70% accuracy on the two datasets, respectively. To our knowledge, the proposed model outperformed all the existing state-of-the-art technology. The proposed supervised machine learning model was evaluated using unsupervised machine learning techniques such as principal component analysis (PCA), and it was observed that the proposed method provided a valid performance for identifying m5U. Considering its multi-layered construction, the GRUpred-m5U model has tremendous potential for future applications in the biological industry. The model, which consisted of neurons processing complicated input, excelled at pattern recognition and produced reliable results. Despite its greater size, the model obtained accurate results, essential in detecting m5U. [ABSTRACT FROM AUTHOR]

Purpose Methods Results Conclusion To validate the performance of autonomous diabetic retinopathy (DR) grading by comparing a human grader and a self‐developed deep‐learning (DL) algorithm with gold‐standard evaluation.We included 500, 6‐field retinal images graded by an expert ophthalmologist (gold standard) according to the International Clinical Diabetic Retinopathy Disease Severity Scale as represented with DR levels 0–4 (97, 100, 100, 103, 100, respectively). Weighted kappa was calculated to measure the DR classification agreement for (1) a certified human grader without, and (2) with assistance from a DL algorithm and (3) the DL operating autonomously. Using any DR (level 0 vs. 1–4) as a cutoff, we calculated sensitivity, specificity, as well as positive and negative predictive values (PPV and NPV). Finally, we assessed lesion discrepancies between Model 3 and the gold standard.As compared to the gold standard, weighted kappa for Models 1–3 was 0.88, 0.89 and 0.72, sensitivities were 95%, 94% and 78% and specificities were 82%, 84% and 81%. Extrapolating to a real‐world DR prevalence of 23.8%, the PPV were 63%, 64% and 57% and the NPV were 98%, 98% and 92%. Discrepancies between the gold standard and Model 3 were mainly incorrect detection of artefacts (n = 49), missed microaneurysms (n = 26) and inconsistencies between the segmentation and classification (n = 51).While the autonomous DL algorithm for DR classification only performed on par with a human grader for some measures in a high‐risk population, extrapolations to a real‐world population demonstrated an excellent 92% NPV, which could make it clinically feasible to use autonomously to identify non‐DR patients. [ABSTRACT FROM AUTHOR]

Tissue morphogenesis is intimately linked to the changes in shape and organisation of individual cells. In curved epithelia, cells can intercalate along their own apicobasal axes, adopting a shape named 'scutoid' that allows energy minimization in the tissue. Although several geometric and biophysical factors have been associated with this 3D reorganisation, the dynamic changes underlying scutoid formation in 3D epithelial packing remain poorly understood. Here, we use live imaging of the sea star embryo coupled with deep learningbased segmentation to dissect the relative contributions of cell density, tissue compaction and cell proliferation on epithelial architecture. We find that tissue compaction, which naturally occurs in the embryo, is necessary for the appearance of scutoids. Physical compression experiments identify cell density as the factor promoting scutoid formation at a global level. Finally, the comparison of the developing embryo with computational models indicates that the increase in the proportion of scutoids is directly associated with cell divisions. Our results suggest that apico-basal intercalations appearing immediately after mitosis may help accommodate the new cells within the tissue. We propose that proliferation in a compact epithelium induces 3D cell rearrangements during development. [ABSTRACT FROM AUTHOR]

A major concern in data-driven deep learning (DL) is how to maximize the capability of a model for limited datasets. The lack of high-performance datasets limits intelligent agriculture development. Recent studies have shown that image enhancement techniques can alleviate the limitations of datasets on model performance. Existing image enhancement algorithms mainly perform in the same category and generate highly correlated samples. Directly using authentic images to expand the dataset, the environmental noise in the image will seriously affect the model's accuracy. Hence, this paper designs an automatic leaf segmentation algorithm (AISG) based on the EISeg segmentation method, separating the leaf information with disease spot characteristics from the background noise in the picture. This algorithm enhances the network model's ability to extract disease features. In addition, the Cycle-GAN network is used for minor sample data enhancement to realize cross-category image transformation. Then, MobileNet was trained by transfer learning on an enhanced dataset. The experimental results reveal that the proposed method achieves a classification accuracy of 98.61% for the ten types of tomato diseases, surpassing the performance of other existing methods. Our method is beneficial in solving the problems of low accuracy and insufficient training data in tomato disease detection. This method can also provide a reference for the detection of other types of plant diseases. [ABSTRACT FROM AUTHOR]

By utilizing computed tomography (CT) technology, we can gain a comprehensive understanding of the specific details within the material. When combined with computational mechanics, this approach allows us to predict the structural response through numerical simulation, thereby avoiding the high experimental costs. In this study, the tensile cracking behavior of carbon–silicon carbide (C/SiC) composites is numerically simulated using the bond-based peridynamics model (BB-PD), which is based on geometric models derived from segmented images of three-dimensional (3D) CT data. To obtain results efficiently and accurately, we adopted a deep learning-based image recognition model to identify the kinds of material and then the pixel type that corresponds to the material point, which can be modeled by BB-PD for failure simulation. The numerical simulations of the composites indicate that the proposed image-based peridynamics (IB-PD) model can accurately reconstruct the actual composite microstructure. It can effectively simulate various fracture phenomena such as interfacial debonding, crack propagation affected by defects, and damage to the matrix. [ABSTRACT FROM AUTHOR]

Accurately predicting winter wheat yield before harvest could greatly benefit decision-makers when making management decisions. In this study, we utilized weather forecast (WF) data combined with Sentinel-2 data to establish the deep-learning network and achieved an in-season county-scale wheat yield prediction in China's main wheat-producing areas. We tested a combination of short-term WF data from the China Meteorological Administration to predict in-season yield at different forecast lengths. The results showed that explicitly incorporating WF data can improve the accuracy in crop yield predictions [Root Mean Square Error (RMSE) = 0.517 t/ha] compared to using only remote sensing data (RMSE = 0.624 t/ha). After comparing a series of WF data with different time series lengths, we found that adding 25 days of WF data can achieve the highest yield prediction accuracy. Specifically, the highest accuracy (RMSE = 0.496 t/ha) is achieved when predictions are made on Day of The Year (DOY) 215 (40 days before harvest). Our study established a deep-learning model which can be used for early yield prediction at the county level, and we have proved that weather forecast data can also be applied in data-driven deep-learning yield prediction tasks. [ABSTRACT FROM AUTHOR]

Objectives: Attenuation correction is a critical phenomenon in quantitative positron emission tomography (PET) imaging with its own special challenges. However, computed tomography (CT) modality which is used for attenuation correction and anatomical localization increases patient radiation dose. This study was aimed to develop a deep learning model for attenuation correction of whole-body 68Ga-DOTATATE PET images. Methods: Non-attenuation-corrected and computed tomography-based attenuation-corrected (CTAC) whole-body 68Ga-DOTATATE PET images of 118 patients from two different imaging centers were used. We implemented a residual deep learning model using the NiftyNet framework. The model was trained four times and evaluated six times using the test data from the centers. The quality of the synthesized PET images was compared with the PET-CTAC images using different evaluation metrics, including the peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), mean square error (MSE), and root mean square error (RMSE). Results: Quantitative analysis of four network training sessions and six evaluations revealed the highest and lowest PSNR values as (52.86±6.6) and (47.96±5.09), respectively. Similarly, the highest and lowest SSIM values were obtained (0.99±0.003) and (0.97±0.01), respectively. Additionally, the highest and lowest RMSE and MSE values fell within the ranges of (0.0117±0.003), (0.0015±0.000103), and (0.01072±0.002), (0.000121±5.07xe-5), respectively. The study found that using datasets from the same center resulted in the highest PSNR, while using datasets from different centers led to lower PSNR and SSIM values. In addition, scenarios involving datasets from both centers achieved the best SSIM and the lowest MSE and RMSE. Conclusion: Acceptable accuracy of attenuation correction on 68Ga-DOTATATE PET images using a deep learning model could potentially eliminate the need for additional X-ray imaging modalities, thereby imposing a high radiation dose on the patient. [ABSTRACT FROM AUTHOR]

Object detection adopting deep-learning has strongly promoted the development of intensive aquaculture. However, shrimp larvae, as an important aquatic organism, are more difficult to be detected than others. On the one hand, they have indeed small sizes, which will cause them to be easily ignored due to the background noise pollution. On the other hand, affected by environmental factors and the fact that shrimp larvae like to move fast as jumping, the images of shrimp larvae often appear blurry. In order to obtain better shrimp larvae detection performance, we propose an improved anchor-free method called CAGNet in this paper. Compared with YOLOX_s, three structures including backbone, neck, and head have been improved in the proposed method. Firstly, we ameliorate the backbone by adding a coordinate attention module to extract more location information and semantic information of shrimp larvae at different levels. Secondly, an adaptively spatial feature fusion module is introduced to the neck. It can adaptively integrate effective shrimp larvae features from different levels and suppress the interference of conflicting information arising from the background. Moreover, in the head, we use GIoU module instead of conventional IoU for more accurate bounding box regression. We conducted experiments by collecting shrimp larvae data from a real aquaculture farm. Compared with the general object detection methods and previous related research, CAGNet has achieved better performance in Precision, Recall, F1 Score, and AP@0.5:0.95. Hence, the proposed method can be effectively applied to shrimp larvae detection in intensive aquaculture. [ABSTRACT FROM AUTHOR]

Zenith Tropospheric Delay (ZTD) is one of the main atmospheric errors in the Global Navigation Satellite System (GNSS). In this study, we propose a novel ZTD forecasting model based on the deep-learning method named Informer-based ZTD (IBZTD) forecasting model using the European Centre for Medium-Range Weather Forecasts (ECMWF) fifth generation reanalysis data (ERA5) from 2011 to 2021. With 72-hour historical GNSS-derived ZTDs as prior information, the subsequent 24-hour ZTDs can be forecasted. The IBZTD forecasting model achieves the best regression fit with post GNSS-derived ZTDs compared with GPT3 (Global Pressure and Temperature 3) and HGPT2 (Hourly Global Pressure and Temperature 2) models, especially in winter with a Root Mean Square Error (RMSE) of 1.51 cm and a Mean Absolute Error (MAE) of 1.15 cm. With the post GNSS-derived ZTDs as reference, in terms of the overall 24-hour forecasting accuracy for 9 GNSS stations in 2022, IBZTD forecasting model achieves a MAE of 1.66 cm and a RMSE of 2.21 cm, significantly outperforming the GPT3 model (MAE: 2.60 cm, RMSE: 3.37 cm), HGPT2 model (MAE: 3.23 cm, RMSE: 4.03 cm) and Long Short-Term Memory (LSTM) model (MAE: 2.65 cm, RMSE: 3.65 cm). An average time improvement of 17.8% and comparable forecasting precisions are achieved in the IBZTD forecasting model compared with the Transformer-based ZTD (TBZTD) forecasting model. Using predicted ZTD as prior constraints in Precise Point Positioning (PPP), the vertical convergence speed exhibits a significant improvement of 14.20%, 20.24%, 18.48%, and 19.39% in four seasons. [ABSTRACT FROM AUTHOR]

Neural Machine Translation (NMT) has made significant strides in recent years, especially with the advent of deep learning, which has greatly enhanced performance across various Natural Language Processing (NLP) tasks. Despite these advances, NMT still falls short of perfect translation, facing ongoing challenges such as limited training data, handling rare words, and managing syntactic and semantic dependencies. This study introduces a multichannel character-level NMT model with hybrid attention for Arabic-English translation. The proposed approach addresses issues such as rare words and word alignment by encoding characters, incorporating Arabic word segmentation as handcrafted features, and using part-of-speech tagging in a multichannel CNN-BiLSTM encoder. The model then uses a Bi-LSTM decoder with hybrid attention to generate target language sentences. The proposed model was tested on a subset of the OPUS- 100 dataset, achieving promising results. [ABSTRACT FROM AUTHOR]

This study delves into the development of a novel 10 by 10 sensor array featuring 100 pressure sensor pixels, achieving remarkable sensitivity up to 888.79 kPa−1, through the innovative design of sensor structure. The critical challenge of strain sensitivity inherent is addressed in stretchable piezoresistive pressure sensors, a domain that has seen significant interest due to their potential for practical applications. This approach involves synthesizing and electrospinning polybutadiene‐urethane (PBU), a reversible cross‐linking polymer, subsequently coated with MXene nanosheets to create a conductive fabric. This fabrication technique strategically enhances sensor sensitivity by minimizing initial current values and incorporating semi‐cylindrical electrodes with Ag nanowires (AgNWs) selectively coated for optimal conductivity. The application of a pre‐strain method to electrode construction ensures strain immunity, preserving the sensor's electrical properties under expansion. The sensor array demonstrated remarkable sensitivity by consistently detecting even subtle airflow from an air gun in a wind sensing test, while a novel deep learning methodology significantly enhanced the long‐term sensing accuracy of polymer‐based stretchable mechanical sensors, marking a major advancement in sensor technology. This research presents a significant step forward in enhancing the reliability and performance of stretchable piezoresistive pressure sensors, offering a comprehensive solution to their current limitations. [ABSTRACT FROM AUTHOR]

This work aims at proposing an affordable, non-wearable system to detect falls of people in need of care. The proposal uses artificial vision based on deep learning techniques implemented on a Raspberry Pi4 4GB RAM with a High-Definition IR-CUT camera. The CNN architecture classifies detected people into five classes: fallen, crouching, sitting, standing, and lying down. When a fall is detected, the system sends an alert notification to mobile devices through the Telegram instant messaging platform. The system was evaluated considering real daily indoor activities under different conditions: outfit, lightning, and distance from camera. Results show a good trade-off between performance and cost of the system. Obtained performance metrics are: precision of 96.4 % , specificity of 96.6 % , accuracy of 94.8 % , and sensitivity of 93.1 % . Regarding privacy concerns, even though this system uses a camera, the video is not recorded or monitored by anyone, and pictures are only sent in case of fall detection. This work can contribute to reducing the fatal consequences of falls in people in need of care by providing them with prompt attention. Such a low-cost solution would be desirable, particularly in developing countries with limited or no medical alert systems and few resources. [ABSTRACT FROM AUTHOR]

The Industrial IoT system often struggles to identify malignant traffic and may cause disruption in the flow of work or even hazardous situations. The previously described techniques to identify such intrusions work well but not enough to be implemented in such environments where it is very difficult to identify malignant traffic in loads of benign ones. Hence, an intrusion detection system is needed that works well with very highly unbalanced datasets. Therefore, we developed a transformer model that gives a high accuracy and combined it with a boosting module that decreases false negatives, which is highly required. This model is applied to the UNSW-2018-IoT-Botnet dataset, which is publicly available in the cloudstor network. Thus, the classified traffic identified as malignant is eliminated from the system using prevention techniques. The paper also extends the model to classify among five different traffics for the same dataset, in which some of the traffics are very difficult to distinguish, such as DoS and DDoS traffic. The experiments on such data sets have shown much better results, which proves that the model classifies well and can be implemented practically as well. [ABSTRACT FROM AUTHOR]

Copyright of Arquivos Brasileiros de Oftalmologia is the property of Arquivos Brasileiros de Oftalmologia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Purpose: To develop and evaluate a deep learning (DL) ‐based rapid image reconstruction and motion correction technique for high‐resolution Cartesian first‐pass myocardial perfusion imaging at 3T with whole‐heart coverage for both single‐slice (SS) and simultaneous multi‐slice (SMS) acquisitions. Methods: 3D physics‐driven unrolled network architectures were utilized for the reconstruction of high‐resolution Cartesian perfusion imaging. The SS and SMS multiband (MB) = 2 networks were trained from 135 slices from 20 subjects. Structural similarity index (SSIM), peak SNR (PSNR), and normalized RMS error (NRMSE) were assessed, and prospective images were blindly graded by two experienced cardiologists (5, excellent; 1, poor). For respiratory motion correction, a 2D U‐Net based motion corrected network was proposed, and the temporal fidelity and second‐order derivative were calculated to assess the performance of the motion correction. Results: Excellent performance was demonstrated in the proposed technique with high SSIM and PSNR, and low NRMSE. Image quality scores were (4.3 [4.3, 4.4], 4.5 [4.4, 4.6], 4.3 [4.3, 4.4], and 4.5 [4.3, 4.5]) for SS DL and SS L1‐SENSE, MB = 2 DL and MB = 2 SMS‐L1‐SENSE, respectively, showing no statistically significant difference (p > 0.05 for SS and SMS) between (SMS)‐L1‐SENSE and the proposed DL technique. The network inference time was around 4 s per dynamic perfusion series with 40 frames while the time of (SMS)‐L1‐SENSE with GPU acceleration was approximately 30 min. Conclusion: The proposed DL‐based image reconstruction and motion correction technique enabled rapid and high‐quality reconstruction for SS and SMS MB = 2 high‐resolution Cartesian first‐pass perfusion imaging at 3T. [ABSTRACT FROM AUTHOR]

Background: Studies have shown that deep‐learning radiomics (DLR) could help differentiate glioblastoma (GBM) from solitary brain metastasis (SBM), but whether integrating demographic‐MRI and DLR features can more accurately distinguish GBM from SBM remains uncertain. Purpose: To construct and validate a demographic‐MRI deep‐learning radiomics nomogram (DDLRN) integrating demographic‐MRI and DLR signatures to differentiate GBM from SBM preoperatively. Study Type: Retrospective. Population: Two hundred and thirty‐five patients with GBM (N = 115) or SBM (N = 120), randomly divided into a training cohort (90 GBM and 98 SBM) and a validation cohort (25 GBM and 22 SBM). Field Strength/Sequence: Axial T2‐weighted fast spin‐echo sequence (T2WI), T2‐weighted fluid‐attenuated inversion recovery sequence (T2‐FLAIR), and contrast‐enhanced T1‐weighted spin‐echo sequence (CE‐T1WI) using 1.5‐T and 3.0‐T scanners. Assessment: The demographic‐MRI signature was constructed with seven imaging features ("pool sign," "irregular ring sign," "regular ring sign," "intratumoral vessel sign," the ratio of the area of peritumoral edema to the enhanced tumor, the ratio of the lesion area on T2‐FLAIR to CE‐T1WI, and the tumor location) and demographic factors (age and sex). Based on multiparametric MRI, radiomics and deep‐learning (DL) models, DLR signature, and DDLRN were developed and validated. Statistical Tests: The Mann–Whitney U test, Pearson test, least absolute shrinkage and selection operator, and support vector machine algorithm were applied for feature selection and construction of radiomics and DL models. Results: DDLRN showed the best performance in differentiating GBM from SBM with area under the curves (AUCs) of 0.999 and 0.947 in the training and validation cohorts, respectively. Additionally, the DLR signature (AUC = 0.938) outperformed the radiomics and DL models, and the demographic‐MRI signature (AUC = 0.775) was comparable to the T2‐FLAIR radiomics and DL models in the validation cohort (AUC = 0.762 and 0.749, respectively). Data Conclusion: DDLRN integrating demographic‐MRI and DLR signatures showed excellent performance in differentiating GBM from SBM. Level of Evidence: 3 Technical Efficacy: Stage 2 [ABSTRACT FROM AUTHOR]

Breast cancer is a type of cancer responsible for higher mortality rates among women. The cruelty of breast cancer always requires a promising approach for its earlier detection. In light of this, the proposed research leverages the representation ability of pretrained EfficientNet-B0 model and the classification ability of the XGBoost model for the binary classification of breast tumors. In addition, the above transfer learning model is modified in such a way that it will focus more on tumor cells in the input mammogram. Accordingly, the work proposed an EfficientNet-B0 having a Spatial Attention Layer with XGBoost (ESA-XGBNet) for binary classification of mammograms. For this, the work is trained, tested, and validated using original and augmented mammogram images of three public datasets namely CBIS-DDSM, INbreast, and MIAS databases. Maximum classification accuracy of 97.585% (CBIS-DDSM), 98.255% (INbreast), and 98.91% (MIAS) is obtained using the proposed ESA-XGBNet architecture as compared with the existing models. Furthermore, the decision-making of the proposed ESA-XGBNet architecture is visualized and validated using the Attention Guided GradCAM-based Explainable AI technique. [ABSTRACT FROM AUTHOR]

If the development of machine learning and artificial intelligence plays a role in many fields of research and technology today, it has a special relationship with neurosciences. Indeed, historically inspired by our knowledge of the brain, deep learning shares some vocabularies with neurosciences and can sometimes be considered a brain's model. Taking the particular example of seizure, which can develop in any biological neural tissue, we question if and how the models used for deep learning can capture or model these pathological events. This particular example is a starting point to discuss the nature, limits, and functions of these models, and we discuss what we expect from a model of the brain. Finally, we argue that a pluralistic approach leading to the integrated coexistence of different models is necessary to study the brain in all its complexity. [ABSTRACT FROM AUTHOR]

Abstract This research introduces a Computer-Aided Diagnosis-system designed aimed at automated detections & classification of tomato leaf diseases, combining traditional handcrafted features with advanced deep learning techniques. The system’s process encompasses preprocessing, feature extraction, feature fusion, and classification. It utilizes enhancement filters and segmentation algorithms to isolate with Regions-of-Interests (ROI) in images tomato leaves. These features based arranged in ABCD rule (Asymmetry, Borders, Colors, and Diameter) are integrated with outputs from a Convolutional Neural Network (CNN) pretrained on ImageNet. To address data imbalance, we introduced a novel evaluation method that has shown to improve classification accuracy by 15% compared to traditional methods, achieving an overall accuracy rate of 92% in field tests. By merging classical feature engineering with modern machine learning techniques under mutual information-based feature fusion, our system sets a new standard for precision in agricultural diagnostics. Specific performance metrics showcasing the effectiveness of our approach in automated detection and classifying of tomato leaf disease.

Abstract RNA 5-methyluridine (m5U) sites play a significant role in understanding RNA modifications, which influence numerous biological processes such as gene expression and cellular functioning. Consequently, the identification of m5U sites can play a vital role in the integrity, structure, and function of RNA molecules. Therefore, this study introduces GRUpred-m5U, a novel deep learning-based framework based on a gated recurrent unit in mature RNA and full transcript RNA datasets. We used three descriptor groups: nucleic acid composition, pseudo nucleic acid composition, and physicochemical properties, which include five feature extraction methods ENAC, Kmer, DPCP, DPCP type 2, and PseDNC. Initially, we aggregated all the feature extraction methods and created a new merged set. Three hybrid models were developed employing deep-learning methods and evaluated through 10-fold cross-validation with seven evaluation metrics. After a comprehensive evaluation, the GRUpred-m5U model outperformed the other applied models, obtaining 98.41% and 96.70% accuracy on the two datasets, respectively. To our knowledge, the proposed model outperformed all the existing state-of-the-art technology. The proposed supervised machine learning model was evaluated using unsupervised machine learning techniques such as principal component analysis (PCA), and it was observed that the proposed method provided a valid performance for identifying m5U. Considering its multi-layered construction, the GRUpred-m5U model has tremendous potential for future applications in the biological industry. The model, which consisted of neurons processing complicated input, excelled at pattern recognition and produced reliable results. Despite its greater size, the model obtained accurate results, essential in detecting m5U.

Weimin Cai,1,* Xiao Wu,1,* Kun Guo,2 Yongxian Chen,3 Yubo Shi,4 Xinran Lin1 1Department of Gastroenterology and Hepatology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325000, People’s Republic of China; 2Department of Cardiology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325000, People’s Republic of China; 3Department of Chest Cancer, Xiamen Second People’s Hospital, Xiamen, 36100, People’s Republic of China; 4Department of Pulmonary, Yueqing People’s Hospital, Wenzhou, 325000, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xinran Lin, Department of Gastroenterology and Hepatology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325000, People’s Republic of China, Tel +86 18857838243, Fax +86 0576 87755312, Email lxr190910@163.comBackground: Accurate prediction of treatment response in Crohn’s disease (CD) patients undergoing infliximab (IFX) therapy is essential for clinical decision-making. Our goal was to compare the performance of the clinical characteristics, radiomics and deep learning model from computed tomography enterography (CTE) for identifying individuals at high risk of IFX treatment failure.Methods: This retrospective study enrolled 263 CD patients from three medical centers between 2017 and 2023 patients received CTE examinations within 1 month before IFX commencement. A training cohort was recruited from center 1 (n=166), while test cohort from centers 2 and 3 (n=97). The deep learning model and radiomics were constructed based on CTE images of lesion. The clinical model was developed using clinical characteristics. Two fusion methods were used to create fusion model: the feature-based early fusion model and the decision-based late fusion model. The performances of the predictive models were evaluated.Results: The early fusion model achieved the highest area under characteristics curve (AUC) (0.85– 0.91) among all patient cohorts, significantly outperforming deep learning model (AUC=0.72– 0.82, p=0.06– 0.03, Delong test) and radiomics model (AUC=0.72– 0.78, p=0.06– 0.01). Compared to early fusion model, the AUC values for the clinical and late fusion models were 0.71– 0.91 (p=0.01– 0.41), and 0.81– 0.88 (p=0.49– 0.37) in the test and training set, respectively. Moreover, the early fusion had the lowest value of Brier’s score 0.15– 0.12 in all patient set.Conclusion: The early fusion model, which integrates deep learning, radiomics, and clinical data, can be utilized to predict the response to IFX treatment in CD patients and illustrated clinical decision-making utility.Keywords: deep-learning, radiomics, Crohn’s disease, infliximab, prediction

Plant physiological status is the interaction between the plant genome and the prevailing growth conditions. Accurate characterization of plant physiology is, therefore, fundamental to effective plant phenotyping studies; particularly those focused on identifying traits associated with improved yield, lower input requirements, and climate resilience. Here, we outline the approaches used to assess plant physiology and how these techniques of direct empirical observations of processes such as photosynthetic CO2 assimilation, stomatal conductance, photosystem II electron transport, or the effectiveness of protective energy dissipation mechanisms are unsuited to high-throughput phenotyping applications. Novel optical sensors, remote/proximal sensing (multi- and hyperspectral reflectance, infrared thermography, sun-induced fluorescence), LiDAR, and automated analyses of below-ground development offer the possibility to infer plant physiological status and growth. However, there are limitations to such indirect approaches to gauging plant physiology. These methodologies that are appropriate for the rapid high temporal screening of a number of crop varieties over a wide spatial scale do still require calibration or validation with direct empirical measurement of plant physiological status. The use of deep-learning and artificial intelligence approaches may enable the effective synthesis of large multivariate datasets to more accurately quantify physiological characters rapidly in high numbers of replicate plants. Advances in automated data collection and subsequent data processing represent an opportunity for plant phenotyping efforts to fully integrate fundamental physiological data into vital efforts to ensure food and agro-economic sustainability.

Deep learning is a powerful branch of machine learning, which presents a promising new approach for diagnose diseases. However, the deep learning for detecting anterior cruciate ligament still limits to the evaluation of whether there are injuries. The accuracy of the deep learning model is not high, and the parameters are complex. In this study, we have developed a deep learning model based on ResNet-18 to detect ACL conditions. The results suggest that there is no significant difference between our proposed model and two orthopaedic surgeons and radiologists in diagnosing ACL conditions. [ABSTRACT FROM AUTHOR]

Single Molecule Localisation Microscopy (SMLM) is becoming a widely used technique in cell biology. After processing the images, the molecular localisations are typically stored in a table as xy (or xyz) coordinates, with additional information, such as number of photons, etc. This set of coordinates can be used to generate an image to visualise the molecular distribution, for example, a 2D or 3D histogram of localisations. Many different methods have been devised to analyse SMLM data, among which cluster analysis of the localisations is popular. However, it can be useful to first segment the data, to extract the localisations in a specific region of a cell or in individual cells, prior to downstream analysis. Here we describe a pipeline for annotating localisations in an SMLM dataset in which we compared membrane segmentation approaches, including Otsu thresholding and machine learning models, and subsequent cell segmentation. We used an SMLM dataset derived from dSTORM images of sectioned cell pellets, stained for the membrane proteins EGFR (epidermal growth factor receptor) and EREG (epiregulin) as a test dataset. We found that a Cellpose model retrained on our data performed the best in the membrane segmentation task, allowing us to perform downstream cluster analysis of membrane versus cell interior localisations. We anticipate this will be generally useful for SMLM analysis. LAY DESCRIPTION: A form of microscopy called single‐molecule localisation microscopy (SMLM) allows researchers to locate the positions of proteins with very high precision inside cells. The organisation of these proteins is typically important for their function. Once the imaging data is acquired, it is useful to segment (select areas) regions of the cells to find out how proteins are organised in these specific regions. Here we have developed an approach that can select localisations of proteins within the plasma membrane as well as localisations within the cell, for downstream analysis. We used this data to determine how clusters of molecules varied between the plasma membrane and the interior of the cell. This approach could be generally useful for segmenting SMLM datasets. [ABSTRACT FROM AUTHOR]

Predicting wind speed is a complex undertaking influenced not only by the wind-speed sequence itself but also by various meteorological factors. This paper introduces a novel multivariate deep-learning neural network prediction model that takes into account not only historical wind-speed data but also a series of meteorological features relevant to wind speed. The meteorological features associated with wind speed are initially extracted using the random forest algorithm (RF). Subsequently, Variational Mode Decomposition and Autocorrelation Function analysis are employed for noise reduction in the wind-speed series. Finally, the wind-speed series are predicted using a Gated Recurrent Unit (GRU) deep-learning neural network, and an Improved Sparrow Search Algorithm is proposed to optimize the four parameters of the GRU. To validate the predictive performance of the model, experimental data from three cities in China, Shenyang, Dalian, and Yingkou, are utilized. The experimental results demonstrate that our proposed model outperforms other models, as evidenced by four key performance indicators. [ABSTRACT FROM AUTHOR]

Abstract A recent global health crisis, COVID-19 is a significant global health crisis that has profoundly affected lifestyles. The detection of such diseases from similar thoracic anomalies using medical images is a challenging task. Thus, the requirement of an end-to-end automated system is vastly necessary in clinical treatments. In this way, the work proposes a Squeeze-and-Excitation Attention-based ResNet50 (SEA-ResNet50) model for detecting COVID-19 utilizing chest X-ray data. Here, the idea lies in improving the residual units of ResNet50 using the squeeze-and-excitation attention mechanism. For further enhancement, the Ranger optimizer and adaptive Mish activation function are employed to improve the feature learning of the SEA-ResNet50 model. For evaluation, two publicly available COVID-19 radiographic datasets are utilized. The chest X-ray input images are augmented during experimentation for robust evaluation against four output classes namely normal, pneumonia, lung opacity, and COVID-19. Then a comparative study is done for the SEA-ResNet50 model against VGG-16, Xception, ResNet18, ResNet50, and DenseNet121 architectures. The proposed framework of SEA-ResNet50 together with the Ranger optimizer and adaptive Mish activation provided maximum classification accuracies of 98.38% (multiclass) and 99.29% (binary classification) as compared with the existing CNN architectures. The proposed method achieved the highest Kappa validation scores of 0.975 (multiclass) and 0.98 (binary classification) over others. Furthermore, the visualization of the saliency maps of the abnormal regions is represented using the explainable artificial intelligence (XAI) model, thereby enhancing interpretability in disease diagnosis.

Abstract Accurate recognition of hand-drawn chemical structures is crucial for digitising hand-written chemical information in traditional laboratory notebooks or facilitating stylus-based structure entry on tablets or smartphones. However, the inherent variability in hand-drawn structures poses challenges for existing Optical Chemical Structure Recognition (OCSR) software. To address this, we present an enhanced Deep lEarning for Chemical ImagE Recognition (DECIMER) architecture that leverages a combination of Convolutional Neural Networks (CNNs) and Transformers to improve the recognition of hand-drawn chemical structures. The model incorporates an EfficientNetV2 CNN encoder that extracts features from hand-drawn images, followed by a Transformer decoder that converts the extracted features into Simplified Molecular Input Line Entry System (SMILES) strings. Our models were trained using synthetic hand-drawn images generated by RanDepict, a tool for depicting chemical structures with different style elements. A benchmark was performed using a real-world dataset of hand-drawn chemical structures to evaluate the model's performance. The results indicate that our improved DECIMER architecture exhibits a significantly enhanced recognition accuracy compared to other approaches. Scientific contribution The new DECIMER model presented here refines our previous research efforts and is currently the only open-source model tailored specifically for the recognition of hand-drawn chemical structures. The enhanced model performs better in handling variations in handwriting styles, line thicknesses, and background noise, making it suitable for real-world applications. The DECIMER hand-drawn structure recognition model and its source code have been made available as an open-source package under a permissive license. Graphical Abstract

This paper presents an innovative deep-learning model for Automatic Modulation Classification (AMC) in wireless communication systems. The proposed architecture integrates Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) networks, augmented by a Gaussian noise layer to mitigate overfitting. The integration of both networks seeks to enhance classification accuracy and performance by leveraging the unique capabilities of CNNs and LSTMs in capturing spatial and temporal features, respectively. The model is expected to distinguish between eight digital and two analog modulation modes. Experimental evaluation on the RadioML2016.10b dataset demonstrates a peak recognition accuracy of 93.2% at 18 dB SNR. Comparative analyses validate the superior performance of the proposed architecture. The Gaussian noise layer contributes significantly to a 3% performance improvement at 18 dB SNR. The model achieves recognition accuracy exceeding 96% for most modulation modes, highlighting its robustness. Finally, computational complexity analysis underscores the efficiency of the proposed architecture, reinforcing its practical viability.

Abstract 3D reconstruction of human brain volumes at high resolution is now possible thanks to advancements in tissue clearing methods and fluorescence microscopy techniques. Analyzing the massive data produced with these approaches requires automatic methods able to perform fast and accurate cell counting and localization. Recent advances in deep learning have enabled the development of various tools for cell segmentation. However, accurate quantification of neurons in the human brain presents specific challenges, such as high pixel intensity variability, autofluorescence, non-specific fluorescence and very large size of data. In this paper, we provide a thorough empirical evaluation of three techniques based on deep learning (StarDist, CellPose and BCFind-v2, an updated version of BCFind) using a recently introduced three-dimensional stereological design as a reference for large-scale insights. As a representative problem in human brain analysis, we focus on a $$4~\text {-cm}^3$$ 4 -cm 3 portion of the Broca’s area. We aim at helping users in selecting appropriate techniques depending on their research objectives. To this end, we compare methods along various dimensions of analysis, including correctness of the predicted density and localization, computational efficiency, and human annotation effort. Our results suggest that deep learning approaches are very effective, have a high throughput providing each cell 3D location, and obtain results comparable to the estimates of the adopted stereological design.

Abstract Background For dosimetry, the demand for whole-body SPECT/CT imaging, which require long acquisition durations with dual-head Anger cameras, is increasing. Here we evaluated sparsely acquired projections and assessed whether the addition of deep-learning-generated synthetic intermediate projections (SIPs) could improve the image quality while preserving dosimetric accuracy. Methods This study included 16 patients treated with 177Lu-DOTATATE with SPECT/CT imaging (120 projections, 120P) at four time points. Deep neural networks (CUSIPs) were designed and trained to compile 90 SIPs from 30 acquired projections (30P). The 120P, 30P, and three different CUSIP sets (30P + 90 SIPs) were reconstructed using Monte Carlo-based OSEM reconstruction (yielding 120P_rec, 30P_rec, and CUSIP_recs). The noise levels were visually compared. Quantitative measures of normalised root mean square error, normalised mean absolute error, peak signal-to-noise ratio, and structural similarity were evaluated, and kidney and bone marrow absorbed doses were estimated for each reconstruction set. Results The use of SIPs visually improved noise levels. All quantitative measures demonstrated high similarity between CUSIP sets and 120P. Linear regression showed nearly perfect concordance of the kidney and bone marrow absorbed doses for all reconstruction sets, compared to the doses of 120P_rec (R2 ≥ 0.97). Compared to 120P_rec, the mean relative difference in kidney absorbed dose, for all reconstruction sets, was within 3%. For bone marrow absorbed doses, there was a higher dissipation in relative differences, and CUSIP_recs outperformed 30P_rec in mean relative difference (within 4% compared to 9%). Kidney and bone marrow absorbed doses for 30P_rec were statistically significantly different from those of 120_rec, as opposed to the absorbed doses of the best performing CUSIP_rec, where no statistically significant difference was found. Conclusion When performing SPECT/CT reconstruction, the use of SIPs can substantially reduce acquisition durations in SPECT/CT imaging, enabling acquisition of multiple fields of view of high image quality with satisfactory dosimetric accuracy.

Abstract Various types of motion introduced into a solution can affect, among other factors, the alignment and positioning of molecules, the agglomeration of large molecules, oxidation processes, and the production of microparticles and microbubbles. We employed turbulent mixing vs. laminar flow induced by a vortex vs. diffusion-based mixing during the production of Viscum album Quercus L. 10−3 following the guidelines for manufacturing homeopathic preparations. The differently mixed preparation variants were analyzed using the droplet evaporation method. The crystalline structures formed in dried droplets were photographed and analyzed using computer-supported image analysis and deep learning. Computer-supported evaluation and deep learning revealed that the patterns of the variant succussed under turbulence are characterized by lower complexity, whereas those obtained from the vortex-mixed variant are characterized by greater complexity compared to the diffusion-based mixed control variant. The droplet evaporation method could provide a relatively inexpensive means of testing the effects of liquid flow and serve as an alternative to currently used methods.

Proteins perform their biological functions through motion. Although high throughput prediction of the three-dimensional static structures of proteins has proved feasible using deep-learning-based methods, predicting the conformational motions remains a challenge. Purely data-driven machine learning methods encounter difficulty for addressing such motions because available laboratory data on conformational motions are still limited. In this work, we develop a method for generating protein allosteric motions by integrating physical energy landscape information into deep-learning-based methods. We show that local energetic frustration, which represents a quantification of the local features of the energy landscape governing protein allosteric dynamics, can be utilized to empower AlphaFold2 (AF2) to predict protein conformational motions. Starting from ground state static structures, this integrative method generates alternative structures as well as pathways of protein conformational motions, using a progressive enhancement of the energetic frustration features in the input multiple sequence alignment sequences. For a model protein adenylate kinase, we show that the generated conformational motions are consistent with available experimental and molecular dynamics simulation data. Applying the method to another two proteins KaiB and ribose-binding protein, which involve large-amplitude conformational changes, can also successfully generate the alternative conformations. We also show how to extract overall features of the AF2 energy landscape topography, which has been considered by many to be black box. Incorporating physical knowledge into deeplearning-based structure prediction algorithms provides a useful strategy to address the challenges of dynamic structure prediction of allosteric proteins. [ABSTRACT FROM AUTHOR]

A recent global health crisis, COVID-19 is a significant global health crisis that has profoundly affected lifestyles. The detection of such diseases from similar thoracic anomalies using medical images is a challenging task. Thus, the requirement of an end-to-end automated system is vastly necessary in clinical treatments. In this way, the work proposes a Squeeze-and-Excitation Attention-based ResNet50 (SEA-ResNet50) model for detecting COVID-19 utilizing chest X-ray data. Here, the idea lies in improving the residual units of ResNet50 using the squeeze-and-excitation attention mechanism. For further enhancement, the Ranger optimizer and adaptive Mish activation function are employed to improve the feature learning of the SEA-ResNet50 model. For evaluation, two publicly available COVID-19 radiographic datasets are utilized. The chest X-ray input images are augmented during experimentation for robust evaluation against four output classes namely normal, pneumonia, lung opacity, and COVID-19. Then a comparative study is done for the SEA-ResNet50 model against VGG-16, Xception, ResNet18, ResNet50, and DenseNet121 architectures. The proposed framework of SEA-ResNet50 together with the Ranger optimizer and adaptive Mish activation provided maximum classification accuracies of 98.38% (multiclass) and 99.29% (binary classification) as compared with the existing CNN architectures. The proposed method achieved the highest Kappa validation scores of 0.975 (multiclass) and 0.98 (binary classification) over others. Furthermore, the visualization of the saliency maps of the abnormal regions is represented using the explainable artificial intelligence (XAI) model, thereby enhancing interpretability in disease diagnosis. [ABSTRACT FROM AUTHOR]

In the current digital era, images have become one of the key channels for communication and information. There are multiple platforms where digital images are used as an essential identity, like social media platforms, chat applications, electronic and print media, medical science, forensics and criminal investigation, the court of law, and many more. Alternation of digital images becomes easy because multiple image editing software applications are accessible freely on the internet. These modified images can create severe problems in the field where the correctness of the image is essential. In such situations, the authenticity of the digital images from the bare eye is almost impossible. To prove the validity of the digital images, we have only one option: Digital Image Forensics (DIF). This study reviewed various image forgery and image forgery detection methods based on blind forgery detection techniques mainly. We describe the essential components of these approaches, as well as the datasets used to train and verify them. Performance analysis of these methods on various metrics is also discussed here. [ABSTRACT FROM AUTHOR]

Background/Aim: To evaluate the performance of Chat Generative Pre-trained Transformer (ChatGPT), a large language model trained by Open artificial intelligence. Materials and Methods: This study has three main steps to evaluate the effectiveness of ChatGPT in the urologic field. The first step involved 35 questions from our institution's experts, who have at least 10 years of experience in their fields. The responses of ChatGPT versions were qualitatively compared with the responses of urology residents to the same questions. The second step assesses the reliability of ChatGPT versions in answering current debate topics. The third step was to assess the reliability of ChatGPT versions in providing medical recommendations and directives to patients' commonly asked questions during the outpatient and inpatient clinic. Results: In the first step, version 4 provided correct answers to 25 questions out of 35 while version 3.5 provided only 19 (71.4% vs 54%). It was observed that residents in their last year of education in our clinic also provided a mean of 25 correct answers, and 4th year residents provided a mean of 19.3 correct responses. The second step involved evaluating the response of both versions to debate situations in urology, and it was found that both versions provided variable and inappropriate results. In the last step, both versions had a similar success rate in providing recommendations and guidance to patients based on expert ratings. Conclusion: The difference between the two versions of the 35 questions in the first step of the study was thought to be due to the improvement of ChatGPT's literature and data synthesis abilities. It may be a logical approach to use ChatGPT versions to inform the nonhealth care providers' questions with quick and safe answers but should not be used to as a diagnostic tool or make a choice among different treatment modalities. [ABSTRACT FROM AUTHOR]

Background: Direction Modulated Brachytherapy (DMBT) enables conformal dose distributions. However, clinicians may face challenges in creating viable treatment plans within a fast‐paced clinical setting, especially for a novel technology like DMBT, where cumulative clinical experience is limited. Deep learning‐based dose prediction methods have emerged as effective tools for enhancing efficiency. Purpose: To develop a voxel‐wise dose prediction model using an attention‐gating mechanism and a 3D UNET for cervical cancer high‐dose‐rate (HDR) brachytherapy treatment planning with DMBT six‐groove tandems with ovoids or ring applicators. Methods: A multi‐institutional cohort of 122 retrospective clinical HDR brachytherapy plans treated to a prescription dose in the range of 4.8–7.0 Gy/fraction was used. A DMBT tandem model was constructed and incorporated onto a research version of BrachyVision Treatment Planning System (BV‐TPS) as a 3D solid model applicator and retrospectively re‐planned all cases by seasoned experts. Those plans were randomly divided into 64:16:20 as training, validating, and testing cohorts, respectively. Data augmentation was applied to the training and validation sets to increase the size by a factor of 4. An attention‐gated 3D UNET architecture model was developed to predict full 3D dose distributions based on high‐risk clinical target volume (CTVHR) and organs at risk (OARs) contour information. The model was trained using the mean absolute error loss function, Adam optimization algorithm, a learning rate of 0.001, 250 epochs, and a batch size of eight. In addition, a baseline UNET model was trained similarly for comparison. The model performance was evaluated on the testing dataset by analyzing the outcomes in terms of mean dose values and derived dose‐volume‐histogram indices from 3D dose distributions and comparing the generated dose distributions against the ground‐truth dose distributions using dose statistics and clinically meaningful dosimetric indices. Results: The proposed attention‐gated 3D UNET model showed competitive accuracy in predicting 3D dose distributions that closely resemble the ground‐truth dose distributions. The average values of the mean absolute errors were 1.82 ± 29.09 Gy (vs. 6.41 ± 20.16 Gy for a baseline UNET) in CTVHR, 0.89 ± 1.25 Gy (vs. 0.94 ± 3.96 Gy for a baseline UNET) in the bladder, 0.33 ± 0.67 Gy (vs. 0.53 ± 1.66 Gy for a baseline UNET) in the rectum, and 0.55 ± 1.57 Gy (vs. 0.76 ± 2.89 Gy for a baseline UNET) in the sigmoid. The results showed that the mean absolute error (MAE) for the bladder, rectum, and sigmoid were 0.22 ± 1.22 Gy (3.62%) (p = 0.015), 0.21 ± 1.06 Gy (2.20%) (p = 0.172), and ‐0.03 ± 0.54 Gy (1.13%) (p = 0.774), respectively. The MAE for D90, V100%, and V150% of the CTVHR were 0.46 ± 2.44 Gy (8.14%) (p = 0.018), 0.57 ± 11.25% (5.23%) (p = 0.283), and ‐0.43 ± 19.36% (4.62%) (p = 0.190), respectively. The proposed model needs less than 5 s to predict a full 3D dose distribution of 64 × 64 × 64 voxels for any new patient plan, thus making it sufficient for near real‐time applications and aiding with decision‐making in the clinic. Conclusions: Attention gated 3D‐UNET model demonstrated a capability in predicting voxel‐wise dose prediction, in comparison to 3D UNET, for DMBT intracavitary brachytherapy planning. The proposed model could be used to obtain dose distributions for near real‐time decision‐making before DMBT planning and quality assurance. This will guide future automated planning, making the workflow more efficient and clinically viable. [ABSTRACT FROM AUTHOR]

Machine learning approaches have become a crucial tool in graph analysis. Despite the accurate results of the existing approaches, most of them are not scalable enough to be used in real-world problems. Networks provide two different kinds of information, nodes contents and nodes relations (network structure). Training deep graph neural networks (GNN) over large-scale graphs is challenging due to the limitation of the message passing framework. Graph Convolutional Networks (GCN) work on all node neighbours at once. Furthermore, it is usual to transform node features with a deep neural network before the GC operation. Therefore, the deep transform operation may apply up to hundreds of times for each target node which is heavy computation and hard to batch. This paper presents an abstract framework with two embedding components, the first component embeds node relations, and the second one embeds node contents. The model makes predictions by aggregating these embeddings through a combination component. The presented approach limits the deep transform only to the target node and uses random walk-based embedding instead of the GC operator to reduce the cost. The main goal of the proposed approach is to provide a light framework for the task. To this aim, node relations are embedded based on node neighbourhood structure by a biased variant of the DeepWalk model, called GuidedWalk, and an autoencoder embeds node contents. The experimental results on three well-known datasets show the superiority of the proposed model compared to the state-of-the-art GraphSAGE and TADW models with less computational complexity. On the Citeseer, Cora, and PubMed datasets, the model has achieved 3.23%, 0.88%, and 7.63% improvement in Macro-F1 and 3.25%, 0.7%, and 6.34% improvement in Micro-F1, respectively. Although GNNs are state-of-the-art models, considering node content is their main advantage. This paper shows that even a simple integration of node content to available random walk-based methods improves their performance up to GCNs without increasing the complexity. [ABSTRACT FROM AUTHOR]

Patients with hypertrophic cardiomyopathy (HCM) are at risk of sudden death, and individuals with ≥1 major risk markers are considered for primary prevention implantable cardioverter-defibrillators. Guidelines recommend cardiac magnetic resonance (CMR) imaging to identify high-risk imaging features. However, CMR imaging is resource intensive and is not widely accessible worldwide. The purpose of this study was to develop electrocardiogram (ECG) deep-learning (DL) models for the identification of patients with HCM and high-risk imaging features. Patients with HCM evaluated at Tufts Medical Center (N = 1930; Boston, MA) were used to develop ECG-DL models for the prediction of high-risk imaging features: systolic dysfunction, massive hypertrophy (≥30 mm), apical aneurysm, and extensive late gadolinium enhancement. ECG-DL models were externally validated in a cohort of patients with HCM from the Amrita Hospital HCM Center (N = 233; Kochi, India). ECG-DL models reliably identified high-risk features (systolic dysfunction, massive hypertrophy, apical aneurysm, and extensive late gadolinium enhancement) during holdout testing (c-statistic 0.72, 0.83, 0.93, and 0.76) and external validation (c-statistic 0.71, 0.76, 0.91, and 0.68). A hypothetical screening strategy using echocardiography combined with ECG-DL–guided selective CMR use demonstrated a sensitivity of 97% for identifying patients with high-risk features while reducing the number of recommended CMRs by 61%. The negative predictive value with this screening strategy for the absence of high-risk features in patients without ECG-DL recommendation for CMR was 99.5%. In HCM, novel ECG-DL models reliably identified patients with high-risk imaging features while offering the potential to reduce CMR testing requirements in underresourced areas. [Display omitted] [ABSTRACT FROM AUTHOR]

Purpose: To evaluate organ at risk (OAR) auto-segmentation in the head and neck region of computed tomography images using two different commercially available deep-learning-based auto-segmentation (DLAS) tools in a single institutional clinical applications. Methods: Twenty-two OARs were manually contoured by clinicians according to published guidelines on planning computed tomography (pCT) images for 40 clinical head and neck cancer (HNC) cases. Automatic contours were generated for each patient using two deep-learning-based auto-segmentation models--Manteia AccuContour and MIM Prote'ge'AI. The accuracy and integrity of autocontours (ACs) were then compared to expert contours (ECs) using the Sørensen-Dice similarity coefficient (DSC) and Mean Distance (MD) metrics. Results: ACs were generated for 22 OARs using AccuContour and 17 OARs using Prote'ge'AI with average contour generation time of 1 min/patient and 5 min/patient respectively. EC and AC agreement was highest for the mandible (DSC 0.90 ± 0.16) and (DSC 0.91 ± 0.03), and lowest for the chiasm (DSC 0.28 ± 0.14) and (DSC 0.30 ± 0.14) for AccuContour and Prote'ge'AI respectively. Using AccuContour, the average MD was<1mm for 10 of the 22 OARs contoured, 1-2mm for 6 OARs, and 2-3mm for 6 OARs. For Prote'ge'AI, the average mean distance was<1mm for 8 out of 17 OARs, 1-2mm for 6 OARs, and 2-3mm for 3 OARs. Conclusions: Both DLAS programs were proven to be valuable tools to significantly reduce the time required to generate large amounts of OAR contours in the head and neck region, even though manual editing of ACs is likely needed prior to implementation into treatment planning. The DSCs and MDs achieved were similar to those reported in other studies that evaluated various other DLAS solutions. Still, small volume structures with nonideal contrast in CT images, such as nerves, are very challenging and will require additional solutions to achieve sufficient results. [ABSTRACT FROM AUTHOR]

Counting nematodes is a labor-intensive and time-consuming task, yet it is a pivotal step in various quantitative nematological studies; preparation of initial population densities and final population densities in pot, micro-plot and field trials for different objectives related to management including sampling and location of nematode infestation foci. Nematologists have long battled with the complexities of nematode counting, leading to several research initiatives aimed at automating this process. However, these research endeavors have primarily focused on identifying single-class objects within individual images. To enhance the practicality of this technology, there’s a pressing need for an algorithm that cannot only detect but also classify multiple classes of objects concurrently. This study endeavors to tackle this challenge by developing a user-friendly Graphical User Interface (GUI) that comprises multiple deep learning algorithms, allowing simultaneous recognition and categorization of nematode eggs and second stage juveniles of Meloidogyne spp. In total of 650 images for eggs and 1339 images for juveniles were generated using two distinct imaging systems, resulting in 8655 eggs and 4742 Meloidogyne juveniles annotated using bounding box and segmentation, respectively. The deep-learning models were developed by leveraging the Convolutional Neural Networks (CNNs) machine learning architecture known as YOLOv8x. Our results showed that the models correctly identified eggs as eggs and Meloidogyne juveniles as Meloidogyne juveniles in 94% and 93% of instances, respectively. The model demonstrated higher than 0.70 coefficient correlation between model predictions and observations on unseen images. Our study has showcased the potential utility of these models in practical applications for the future. The GUI is made freely available to the public through the author’s GitHub repository (https://github.com/bresilla/ nematode_counting). While this study currently focuses on one genus, there are plans to expand the GUI’s capabilities to include other economically significant genera of plant parasitic nematodes. Achieving these objectives, including enhancing the models’ accuracy on different imaging systems, may necessitate collaboration among multiple nematology teams and laboratories, rather than being the work of a single entity. With the increasing interest among nematologists in harnessing machine learning, the authors are confident in the potential development of a universal automated nematode counting system accessible to all. This paper aims to serve as a framework and catalyst for initiating global collaboration toward this important goal. [ABSTRACT FROM AUTHOR]

Accurate recognition of hand-drawn chemical structures is crucial for digitising hand-written chemical information in traditional laboratory notebooks or facilitating stylus-based structure entry on tablets or smartphones. However, the inherent variability in hand-drawn structures poses challenges for existing Optical Chemical Structure Recognition (OCSR) software. To address this, we present an enhanced Deep lEarning for Chemical ImagE Recognition (DECIMER) architecture that leverages a combination of Convolutional Neural Networks (CNNs) and Transformers to improve the recognition of hand-drawn chemical structures. The model incorporates an EfficientNetV2 CNN encoder that extracts features from hand-drawn images, followed by a Transformer decoder that converts the extracted features into Simplified Molecular Input Line Entry System (SMILES) strings. Our models were trained using synthetic hand-drawn images generated by RanDepict, a tool for depicting chemical structures with different style elements. A benchmark was performed using a real-world dataset of hand-drawn chemical structures to evaluate the model's performance. The results indicate that our improved DECIMER architecture exhibits a significantly enhanced recognition accuracy compared to other approaches. Scientific contribution: The new DECIMER model presented here refines our previous research efforts and is currently the only open-source model tailored specifically for the recognition of hand-drawn chemical structures. The enhanced model performs better in handling variations in handwriting styles, line thicknesses, and background noise, making it suitable for real-world applications. The DECIMER hand-drawn structure recognition model and its source code have been made available as an open-source package under a permissive license. [ABSTRACT FROM AUTHOR]

Adolescent idiopathic scoliosis (AIS) necessitates accurate spinal curvature assessment for effective clinical management. Traditional two-dimensional (2D) Cobb angle measurements have been the standard, but the emergence of three-dimensional (3D) automatic measurement techniques, such as those using weight-bearing 3D imaging (WR3D), presents an opportunity to enhance the accuracy and comprehensiveness of AIS evaluation. This study aimed to compare traditional 2D Cobb angle measurements with 3D automatic measurements utilizing the WR3D imaging technique in patients with AIS. A cohort of 53 AIS patients was recruited, encompassing 88 spinal curves, for comparative analysis. The patient sample consisted of 53 individuals diagnosed with AIS. Cobb angles were calculated using the conventional 2D method and three different 3D methods: the Analytical Method (AM), the Plane Intersecting Method (PIM), and the Plane Projection Method (PPM). The 2D cobb angle was manually measured by 3 experienced clinicians with 2D frontal whole-spine radiographs. For 3D cobb angle measurements, the spine and femoral heads were segmented from the WR3D images using a 3D-UNet deep-learning model, and the automatic calculations of the angles were performed with the 3D slicer software. AM and PIM estimates were found to be significantly larger than 2D measurements. Conversely, PPM results showed no statistical difference compared to the 2D method. These findings were consistent in a subgroup analysis based on 2D Cobb angles. Each 3D measurement method provides a unique assessment of spinal curvature, with PPM offering values closely resembling 2D measurements, while AM and PIM yield larger estimations. The utilization of WR3D technology alongside deep learning segmentation ensures accuracy and efficiency in comparative analyses. However, additional studies, particularly involving patients with severe curves, are required to validate and expand on these results. This study emphasizes the importance of selecting an appropriate measurement method considering the imaging modality and clinical context when assessing AIS, and it also underlines the need for continuous refinement of these techniques for optimal use in clinical decision-making and patient management. [ABSTRACT FROM AUTHOR]

This paper presents an innovative deep-learning model for Automatic Modulation Classification (AMC) in wireless communication systems. The proposed architecture integrates Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) networks, augmented by a Gaussian noise layer to mitigate overfitting. The integration of both networks seeks to enhance classification accuracy and performance by leveraging the unique capabilities of CNNs and LSTMs in capturing spatial and temporal features, respectively. The model is expected to distinguish between eight digital and two analog modulation modes. Experimental evaluation on the RadioML2016.10b dataset demonstrates a peak recognition accuracy of 93.2% at 18 dB SNR. Comparative analyses validate the superior performance of the proposed architecture. The Gaussian noise layer contributes significantly to a 3% performance improvement at 18 dB SNR. The model achieves recognition accuracy exceeding 96% for most modulation modes, highlighting its robustness. Finally, computational complexity analysis underscores the efficiency of the proposed architecture, reinforcing its practical viability. [ABSTRACT FROM AUTHOR]

This study describes improving network security by implementing and assessing an intrusion detection system (IDS) based on deep neural networks (DNNs). The paper investigates contemporary technical ways for enhancing intrusion detection performance, given the vital relevance of safeguarding computer networks against harmful activity. The DNN-based IDS is trained and validated by the model using the NSL-KDD dataset, a popular benchmark for IDS research. The model performs well in both the training and validation stages, with 91.30% training accuracy and 94.38% validation accuracy. Thus, the model shows good learning and generalization capabilities with minor losses of 0.22 in training and 0.1553 in validation. Furthermore, for both macro and micro averages across class 0 (normal) and class 1 (anomalous) data, the study evaluates the model using a variety of assessment measures, such as accuracy scores, precision, recall, and F1 scores. The macro-average recall is 0.9422, the macro-average precision is 0.9482, and the accuracy scores are 0.942. Furthermore, macro-averaged F1 scores of 0.9245 for class 1 and 0.9434 for class 0 demonstrate the model's ability to precisely identify anomalies precisely. The research also highlights how real-time threat monitoring and enhanced resistance against new online attacks may be achieved by DNN-based intrusion detection systems, which can significantly improve network security. The study underscores the critical function of DNN-based IDS in contemporary cybersecurity procedures by setting the foundation for further developments in this field. Upcoming research aims to enhance intrusion detection systems by examining cooperative learning techniques and integrating up-to-date threat knowledge. [ABSTRACT FROM AUTHOR]

Determining the script of historical manuscripts is pivotal for understanding historical narratives, providing historians with vital insights into the past. In this study, our focus lies in developing an automated system for effectively identifying the script of historical documents using a deep learning approach. Leveraging the ClAMM dataset as the foundation for our system, we initiate the system with dataset preprocessing, employing two fundamental techniques: denoising through non-local means denoising and binarization using Canny-edge detection. These techniques prepare the document for keypoint detection facilitated by the Harris-corner detector, a feature-detection method. Subsequently, we cluster these keypoints utilizing the k-means algorithm and extract patches based on the identified features. The final step involves training these patches on deep learning models, with a comparative analysis between two architectures: Convolutional Neural Networks (CNN) and Vision Transformers (ViT). Given the absence of prior studies investigating the performance of vision transformers on historical manuscripts, our research fills this gap. The system undergoes a series of experiments to fine-tune its parameters for optimal performance. Our conclusive results demonstrate an average accuracy of 89.2 and 91.99% respectively of the CNN and ViT based proposed framework, surpassing the state of the art in historical script classification so far, and affirming the effectiveness of our automated script identification system. [ABSTRACT FROM AUTHOR]