Search

Since 2000, there have been more than 8000 publications on radiology artificial intelligence (AI). AI breakthroughs allow complex tasks to be automated and even performed beyond human capabilities. However, the lack of details on the methods and algorithm code undercuts its scientific value. Many science subfields have recently faced a reproducibility crisis, eroding trust in processes and results, and influencing the rise in retractions of scientific papers. For the same reasons, conducting research in deep learning (DL) also requires reproducibility. Although several valuable manuscript checklists for AI in medical imaging exist, they are not focused specifically on reproducibility. In this study, we conducted a systematic review of recently published papers in the field of DL to evaluate if the description of their methodology could allow the reproducibility of their findings. We focused on the Journal of Digital Imaging (JDI), a specialized journal that publishes papers on AI and medical imaging. We used the keyword "Deep Learning" and collected the articles published between January 2020 and January 2022. We screened all the articles and included the ones which reported the development of a DL tool in medical imaging. We extracted the reported details about the dataset, data handling steps, data splitting, model details, and performance metrics of each included article. We found 148 articles. Eighty were included after screening for articles that reported developing a DL model for medical image analysis. Five studies have made their code publicly available, and 35 studies have utilized publicly available datasets. We provided figures to show the ratio and absolute count of reported items from included studies. According to our cross-sectional study, in JDI publications on DL in medical imaging, authors infrequently report the key elements of their study to make it reproducible. [ABSTRACT FROM AUTHOR]

Deep learning techniques have recently yielded remarkable results across various fields. However, the quality of these results depends heavily on the quality and quantity of data used during the training phase. One common issue in multi-class and multi-label classification is class imbalance, where one or several classes make up a substantial portion of the total instances. This imbalance causes the neural network to prioritize features of the majority classes during training, as their detection leads to higher scores. In the context of object detection, two types of imbalance can be identified: (1) an imbalance between the space occupied by the foreground and background and (2) an imbalance in the number of instances for each class. This paper aims to address the second type of imbalance without exacerbating the first. To achieve this, we propose a modification of the copy-paste data augmentation technique, combined with weight-balancing methods in the loss function. This strategy was specifically tailored to improve the performance in datasets with a high instance density, where instance overlap could be detrimental. To validate our methodology, we applied it to a highly unbalanced dataset focused on nuclei detection. The results show that this hybrid approach improves the classification of minority classes without significantly compromising the performance of majority classes. [ABSTRACT FROM AUTHOR]

Ensuring the transparency of machine learning models is vital for their ethical application in various industries. There has been a concurrent trend of distributed machine learning designed to limit access to training data for privacy concerns. Such models, trained over horizontally or vertically partitioned data, present a challenge for explainable AI because the explaining party may have a biased view of background data or a partial view of the feature space. As a result, explanations obtained from different participants of distributed machine learning might not be consistent with one another, undermining trust in the product. This paper presents an Explainable Data Collaboration Framework based on a model-agnostic additive feature attribution algorithm (KernelSHAP) and Data Collaboration method of privacy-preserving distributed machine learning. In particular, we present three algorithms for different scenarios of explainability in Data Collaboration and verify their consistency with experiments on open-access datasets. Our results demonstrated a significant (by at least a factor of 1.75) decrease in feature attribution discrepancies among the users of distributed machine learning. The proposed method improves consistency among explanations obtained from different participants, which can enhance trust in the product and enable ethical application in various industries. [ABSTRACT FROM AUTHOR]

Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most recent developments in artificial intelligence are coming from deep learning, which has proven revolutionary in almost all fields, from computer vision to health sciences. The effects of deep learning in medicine have changed the conventional ways of clinical application significantly. Although some sub-fields of medicine, such as pediatrics, have been relatively slow in receiving the critical benefits of deep learning, related research in pediatrics has started to accumulate to a significant level, too. Hence, in this paper, we review recently developed machine learning and deep learning-based solutions for neonatology applications. We systematically evaluate the roles of both classical machine learning and deep learning in neonatology applications, define the methodologies, including algorithmic developments, and describe the remaining challenges in the assessment of neonatal diseases by using PRISMA 2020 guidelines. To date, the primary areas of focus in neonatology regarding AI applications have included survival analysis, neuroimaging, analysis of vital parameters and biosignals, and retinopathy of prematurity diagnosis. We have categorically summarized 106 research articles from 1996 to 2022 and discussed their pros and cons, respectively. In this systematic review, we aimed to further enhance the comprehensiveness of the study. We also discuss possible directions for new AI models and the future of neonatology with the rising power of AI, suggesting roadmaps for the integration of AI into neonatal intensive care units. [ABSTRACT FROM AUTHOR]

AI-enabled ECGs have previously been shown to accurately predict patient sex in adults and correlate with sex hormone levels. We aimed to test the ability of AI-enabled ECGs to predict sex in the pediatric population and study the influence of pubertal development. AI-enabled ECG models were created using a convolutional neural network trained on pediatric 10-second, 12-lead ECGs. The first model was trained de novo using pediatric data. The second model used transfer learning from a previously validated adult data-derived algorithm. We analyzed the first ECG from 90,133 unique pediatric patients (aged ≤18 years) recorded between 1987–2022, and divided the cohort into training, validation, and testing datasets. Subgroup analysis was performed on prepubertal (0–7 years), peripubertal (8–14 years), and postpubertal (15–18 years) patients. The cohort was 46.7% male, with 21,678 prepubertal, 26,740 peripubertal, and 41,715 postpubertal children. The de novo pediatric model demonstrated 81% accuracy and an area under the curve (AUC) of 0.91. Model sensitivity was 0.79, specificity was 0.83, positive predicted value was 0.84, and the negative predicted value was 0.78, for the entire test cohort. The model's discriminatory ability was highest in postpubertal (AUC = 0.98), lower in the peripubertal age group (AUC = 0.91), and poor in the prepubertal age group (AUC = 0.67). There was no significant performance difference observed between the transfer learning and de novo models. AI-enabled interpretation of ECG can estimate sex in peripubertal and postpubertal children with high accuracy. [ABSTRACT FROM AUTHOR]

With the continuous mining and gradual reduction of shallow deposits, deep prospecting has become a new global prospecting trend. In addition, with the development of artificial intelligence, deep learning provides a favorable means for geological big data analysis. This paper, researches the No. II Orebody of the Xiongcun deposit. First, based on previous research results and metallogenic regularity, prospecting information, namely, lithology, Au-Ag-Cu chemical elements and wall rock alteration is extracted, and the block model is established by combining the Kriging interpolation structure. Second, the datasets are divided into dataset I and dataset II according to "randomness" and "depth". Third, deep prospecting prediction models based on deep neural networks (DNN) and the convolutional neural networks (CNN) is constructed, and the model parameters are optimized. Finally, the models are applied to the deep prediction of the Xiongcun No. II Orebody. The results show that the accuracy rate and recall rate of the prediction model based on the DNN algorithm are 96.15% and 89.23%, respectively, and the AUC is 96.39%, which are higher values than those of the CNN algorithm, indicating that the performance of the prediction model based on the DNN algorithm is better. The accuracy of prediction model based on dataset I is higher than that of dataset II. The accuracy of deep metallogenic prediction based on the DNN algorithm is approximately 89%, that based on the CNN is approximately 87%, and that based on prospecting information method is approximately 61.27%. The prediction results of the DNN algorithm are relatively consistent in the spatial location and scale of the orebody. Therefore, based on the work done in this paper, it is feasible to use a deep learning method to carry out deep mineral prediction. [ABSTRACT FROM AUTHOR]

With the continuous mining and gradual reduction of shallow deposits, deep prospecting has become a new global prospecting trend. In addition, with the development of artificial intelligence, deep learning provides a favorable means for geological big data analysis. This paper, researches the No. II Orebody of the Xiongcun deposit. First, based on previous research results and metallogenic regularity, prospecting information, namely, lithology, Au-Ag-Cu chemical elements and wall rock alteration is extracted, and the block model is established by combining the Kriging interpolation structure. Second, the datasets are divided into dataset I and dataset II according to "randomness" and "depth". Third, deep prospecting prediction models based on deep neural networks (DNN) and the convolutional neural networks (CNN) is constructed, and the model parameters are optimized. Finally, the models are applied to the deep prediction of the Xiongcun No. II Orebody. The results show that the accuracy rate and recall rate of the prediction model based on the DNN algorithm are 96.15% and 89.23%, respectively, and the AUC is 96.39%, which are higher values than those of the CNN algorithm, indicating that the performance of the prediction model based on the DNN algorithm is better. The accuracy of prediction model based on dataset I is higher than that of dataset II. The accuracy of deep metallogenic prediction based on the DNN algorithm is approximately 89%, that based on the CNN is approximately 87%, and that based on prospecting information method is approximately 61.27%. The prediction results of the DNN algorithm are relatively consistent in the spatial location and scale of the orebody. Therefore, based on the work done in this paper, it is feasible to use a deep learning method to carry out deep mineral prediction. [ABSTRACT FROM AUTHOR]

Healthcare datasets are becoming larger and more complex, necessitating the development of accurate and generalizable AI models for medical applications. Unstructured datasets, including medical imaging, electrocardiograms, and natural language data, are gaining attention with advancements in deep convolutional neural networks and large language models. However, estimating the generalizability of these models to new healthcare settings without extensive validation on external data remains challenging. In experiments across 13 datasets including X-rays, CTs, ECGs, clinical discharge summaries, and lung auscultation data, our results demonstrate that model performance is frequently overestimated by up to 20% on average due to shortcut learning of hidden data acquisition biases (DAB). Shortcut learning refers to a phenomenon in which an AI model learns to solve a task based on spurious correlations present in the data as opposed to features directly related to the task itself. We propose an open source, bias-corrected external accuracy estimate, PEst, that better estimates external accuracy to within 4% on average by measuring and calibrating for DAB-induced shortcut learning. [ABSTRACT FROM AUTHOR]

Distribution shifts remain a problem for the safe application of regulated medical AI systems, and may impact their real-world performance if undetected. Postmarket shifts can occur for example if algorithms developed on data from various acquisition settings and a heterogeneous population are predominantly applied in hospitals with lower quality data acquisition or other centre-specific acquisition factors, or where some ethnicities are over-represented. Therefore, distribution shift detection could be important for monitoring AI-based medical products during postmarket surveillance. We implemented and evaluated three deep-learning based shift detection techniques (classifier-based, deep kernel, and multiple univariate kolmogorov-smirnov tests) on simulated shifts in a dataset of 130'486 retinal images. We trained a deep learning classifier for diabetic retinopathy grading. We then simulated population shifts by changing the prevalence of patients' sex, ethnicity, and co-morbidities, and example acquisition shifts by changes in image quality. We observed classification subgroup performance disparities w.r.t. image quality, patient sex, ethnicity and co-morbidity presence. The sensitivity at detecting referable diabetic retinopathy ranged from 0.50 to 0.79 for different ethnicities. This motivates the need for detecting shifts after deployment. Classifier-based tests performed best overall, with perfect detection rates for quality and co-morbidity subgroup shifts at a sample size of 1000. It was the only method to detect shifts in patient sex, but required large sample sizes ( > 3 0 ′ 000 ). All methods identified easier-to-detect out-of-distribution shifts with small (≤300) sample sizes. We conclude that effective tools exist for detecting clinically relevant distribution shifts. In particular classifier-based tests can be easily implemented components in the post-market surveillance strategy of medical device manufacturers. [ABSTRACT FROM AUTHOR]

The creation of an image from another and from different types of data including text, scene graph, and object layout, is one of the very challenging tasks in computer vision. In addition, capturing images from different views for generating an object or a product can be exhaustive and expansive to do manually. Now, using deep learning and artificial intelligence techniques, the generation of new images from different type of data has become possible. For that, a significant effort has been devoted recently to develop image generation strategies with a great achievement. To that end, we present in this paper, to the best of the authors' knowledge, the first comprehensive overview of existing image generation methods. Accordingly, a description of each image generation technique is performed based on the nature of the adopted algorithms, type of data used, and main objective. Moreover, each image generation category is discussed by presenting the proposed approaches. In addition, a presentation of existing image generation datasets is given. The evaluation metrics that are suitable for each image generation category are discussed and a comparison of the performance of existing solutions is provided to better inform the state-of-the-art and identify their limitations and strengths. Lastly, the current challenges that are facing this subject are presented. [ABSTRACT FROM AUTHOR]

E-commerce platforms apply recommendation technology to a wide variety of commercial websites to help users quickly find the product they need from a large number of products. At present, recommendation systems have been widely used in large-scale e-commerce websites, and most e-commerce shopping websites attract customers through the image information of goods. In this paper, the research status of content-based image retrieval algorithms is analysed, and how to use image content information to recommend goods is studied. A hierarchical commodity classification and retrieval system are designed. A class decision layer is used to determine the category of commodity images and then precisely retrieve the corresponding category of commodity image features. The corresponding relationship between different commodity visual features is used to recommend commodities, and a matching recommendation algorithm for commodities is proposed. Experiments show that the proposed image content-based recommendation method can coordinate with features, and its recommendation results have high accuracy in practical applications. [ABSTRACT FROM AUTHOR]

The monitoring of physiological parameters is a crucial topic in promoting human health and an indispensable approach for assessing physiological status and diagnosing diseases. Particularly, it holds significant value for patients who require long-term monitoring or with underlying cardiovascular disease. To this end, Visual Contactless Physiological Monitoring (VCPM) is capable of using videos recorded by a consumer camera to monitor blood volume pulse (BVP) signal, heart rate (HR), respiratory rate (RR), oxygen saturation (SpO2) and blood pressure (BP). Recently, deep learning-based pipelines have attracted numerous scholars and achieved unprecedented development. Although VCPM is still an emerging digital medical technology and presents many challenges and opportunities, it has the potential to revolutionize clinical medicine, digital health, telemedicine as well as other areas. The VCPM technology presents a viable solution that can be integrated into these systems for measuring vital parameters during video consultation, owing to its merits of contactless measurement, cost-effectiveness, user-friendly passive monitoring and the sole requirement of an off-the-shelf camera. In fact, the studies of VCPM technologies have been rocketing recently, particularly AI-based approaches, but few are employed in clinical settings. Here we provide a comprehensive overview of the applications, challenges, and prospects of VCPM from the perspective of clinical settings and AI technologies for the first time. The thorough exploration and analysis of clinical scenarios will provide profound guidance for the research and development of VCPM technologies in clinical settings. [ABSTRACT FROM AUTHOR]

Objective: Clinical and surgical decisions for glioblastoma patients depend on a tumor imaging-based evaluation. Artificial Intelligence (AI) can be applied to magnetic resonance imaging (MRI) assessment to support clinical practice, surgery planning and prognostic predictions. In a real-world context, the current obstacles for AI are low-quality imaging and postoperative reliability. The aim of this study is to train an automatic algorithm for glioblastoma segmentation on a clinical MRI dataset and to obtain reliable results both pre- and post-operatively. Methods: The dataset used for this study comprises 237 (71 preoperative and 166 postoperative) MRIs from 71 patients affected by a histologically confirmed Grade IV Glioma. The implemented U-Net architecture was trained by transfer learning to perform the segmentation task on postoperative MRIs. The training was carried out first on BraTS2021 dataset for preoperative segmentation. Performance is evaluated using DICE score (DS) and Hausdorff 95% (H95). Results: In preoperative scenario, overall DS is 91.09 (± 0.60) and H95 is 8.35 (± 1.12), considering tumor core, enhancing tumor and whole tumor (ET and edema). In postoperative context, overall DS is 72.31 (± 2.88) and H95 is 23.43 (± 7.24), considering resection cavity (RC), gross tumor volume (GTV) and whole tumor (WT). Remarkably, the RC segmentation obtained a mean DS of 63.52 (± 8.90) in postoperative MRIs. Conclusions: The performances achieved by the algorithm are consistent with previous literature for both pre-operative and post-operative glioblastoma's MRI evaluation. Through the proposed algorithm, it is possible to reduce the impact of low-quality images and missing sequences. [ABSTRACT FROM AUTHOR]

Abdominal ultrasonography has become an integral component of the evaluation of trauma patients. Internal hemorrhage can be rapidly diagnosed by finding free fluid with point-of-care ultrasound (POCUS) and expedite decisions to perform lifesaving interventions. However, the widespread clinical application of ultrasound is limited by the expertise required for image interpretation. This study aimed to develop a deep learning algorithm to identify the presence and location of hemoperitoneum on POCUS to assist novice clinicians in accurate interpretation of the Focused Assessment with Sonography in Trauma (FAST) exam. We analyzed right upper quadrant (RUQ) FAST exams obtained from 94 adult patients (44 confirmed hemoperitoneum) using the YoloV3 object detection algorithm. Exams were partitioned via fivefold stratified sampling for training, validation, and hold-out testing. We assessed each exam image-by-image using YoloV3 and determined hemoperitoneum presence for the exam using the detection with highest confidence score. We determined the detection threshold as the score that maximizes the geometric mean of sensitivity and specificity over the validation set. The algorithm had 95% sensitivity, 94% specificity, 95% accuracy, and 97% AUC over the test set, significantly outperforming three recent methods. The algorithm also exhibited strength in localization, while the detected box sizes varied with a 56% IOU averaged over positive cases. Image processing demonstrated only 57-ms latency, which is adequate for real-time use at the bedside. These results suggest that a deep learning algorithm can rapidly and accurately identify the presence and location of free fluid in the RUQ of the FAST exam in adult patients with hemoperitoneum. [ABSTRACT FROM AUTHOR]

Communication between sensors spread everywhere in healthcare systems may cause some missing in the transferred features. Repairing the data problems of sensing devices by artificial intelligence technologies have facilitated the Medical Internet of Things (MIoT) and its emerging applications in Healthcare. MIoT has great potential to affect the patient's life. Data collected from smart wearable devices size dramatically increases with data collected from millions of patients who are suffering from diseases such as diabetes. However, sensors or human errors lead to missing some values of the data. The major challenge of this problem is how to predict this value to maintain the data analysis model performance within a good range. In this paper, a complete healthcare system for diabetics has been used, as well as two new algorithms are developed to handle the crucial problem of missed data from MIoT wearable sensors. The proposed work is based on the integration of Random Forest, mean, class' mean, interquartile range (IQR), and Deep Learning to produce a clean and complete dataset. Which can enhance any machine learning model performance. Moreover, the outliers repair technique is proposed based on dataset class detection, then repair it by Deep Learning (DL). The final model accuracy with the two steps of imputation and outliers repair is 97.41% and 99.71% Area Under Curve (AUC). The used healthcare system is a web-based diabetes classification application using flask to be used in hospitals and healthcare centers for the patient diagnosed with an effective fashion. [ABSTRACT FROM AUTHOR]

Nowadays, Artificial Intelligence (AI) is growing by leaps and bounds in almost all fields of technology, and Autonomous Vehicles (AV) research is one more of them. This paper proposes the using of algorithms based on Deep Learning (DL) in the control layer of an autonomous vehicle. More specifically, Deep Reinforcement Learning (DRL) algorithms such as Deep Q-Network (DQN) and Deep Deterministic Policy Gradient (DDPG) are implemented in order to compare results between them. The aim of this work is to obtain a trained model, applying a DRL algorithm, able of sending control commands to the vehicle to navigate properly and efficiently following a determined route. In addition, for each of the algorithms, several agents are presented as a solution, so that each of these agents uses different data sources to achieve the vehicle control commands. For this purpose, an open-source simulator such as CARLA is used, providing to the system with the ability to perform a multitude of tests without any risk into an hyper-realistic urban simulation environment, something that is unthinkable in the real world. The results obtained show that both DQN and DDPG reach the goal, but DDPG obtains a better performance. DDPG perfoms trajectories very similar to classic controller as LQR. In both cases RMSE is lower than 0.1m following trajectories with a range 180-700m. To conclude, some conclusions and future works are commented. [ABSTRACT FROM AUTHOR]

Digital pathology is vital for the correct diagnosis of kidney before transplantation or kidney disease identification. One of the key challenges in kidney diagnosis is glomerulus detection in kidney tissue segments. In this study, we propose a deep learning–based method for glomerulus detection from digitized kidney slide segments. The proposed method applies models based on convolutional neural networks to detect image segments containing the glomerulus region. We employ various networks such as ResNets, UNet, LinkNet, and EfficientNet to train the models. In our experiments on a network trained on the NIH HuBMAP kidney whole slide image dataset, the proposed method achieves the highest scores with Dice coefficient of 0.942. [ABSTRACT FROM AUTHOR]

The bold vision of AI-driven pervasive physiological monitoring, through the proliferation of off-the-shelf wearables that began a decade ago, has created immense opportunities to extract actionable information for precision medicine. These AI algorithms model input-output relationships of a system that, in many cases, exhibits complex nature and personalization requirements. A particular example is cuffless blood pressure estimation using wearable bioimpedance. However, these algorithms need training over significant amount of ground truth data. In the context of biomedical applications, collecting ground truth data, particularly at the personalized level is challenging, burdensome, and in some cases infeasible. Our objective is to establish physics-informed neural network (PINN) models for physiological time series data that would use minimal ground truth information to extract complex cardiovascular information. We achieve this by building Taylor's approximation for gradually changing known cardiovascular relationships between input and output (e.g., sensor measurements to blood pressure) and incorporating this approximation into our proposed neural network training. The effectiveness of the framework is demonstrated through a case study: continuous cuffless BP estimation from time series bioimpedance data. We show that by using PINNs over the state-of-the-art time series models tested on the same datasets, we retain high correlations (systolic: 0.90, diastolic: 0.89) and low error (systolic: 1.3 ± 7.6 mmHg, diastolic: 0.6 ± 6.4 mmHg) while reducing the amount of ground truth training data on average by a factor of 15. This could be helpful in developing future AI algorithms to help interpret pervasive physiologic data using minimal amount of training data. [ABSTRACT FROM AUTHOR]

Rising incidence and mortality of cancer have led to an incremental amount of research in the field. To learn from preexisting data, it has become important to capture maximum information related to disease type, stage, treatment, and outcomes. Medical imaging reports are rich in this kind of information but are only present as free text. The extraction of information from such unstructured text reports is labor-intensive. The use of Natural Language Processing (NLP) tools to extract information from radiology reports can make it less time-consuming as well as more effective. In this study, we have developed and compared different models for the classification of lung carcinoma reports using clinical concepts. This study was approved by the institutional ethics committee as a retrospective study with a waiver of informed consent. A clinical concept-based classification pipeline for lung carcinoma radiology reports was developed using rule-based as well as machine learning models and compared. The machine learning models used were XGBoost and two more deep learning model architectures with bidirectional long short-term neural networks. A corpus consisting of 1700 radiology reports including computed tomography (CT) and positron emission tomography/computed tomography (PET/CT) reports were used for development and testing. Five hundred one radiology reports from MIMIC-III Clinical Database version 1.4 was used for external validation. The pipeline achieved an overall F1 score of 0.94 on the internal set and 0.74 on external validation with the rule-based algorithm using expert input giving the best performance. Among the machine learning models, the Bi-LSTM_dropout model performed better than the ML model using XGBoost and the Bi-LSTM_simple model on internal set, whereas on external validation, the Bi-LSTM_simple model performed relatively better than other 2. This pipeline can be used for clinical concept-based classification of radiology reports related to lung carcinoma from a huge corpus and also for automated annotation of these reports. [ABSTRACT FROM AUTHOR]

Under the black-box nature in the deep learning model, it is uncertain how the change in contrast level and format affects the performance. We aimed to investigate the effect of contrast level and image format on the effectiveness of deep learning for diagnosing pneumothorax on chest radiographs. We collected 3316 images (1016 pneumothorax and 2300 normal images), and all images were set to the standard contrast level (100%) and stored in the Digital Imaging and Communication in Medicine and Joint Photographic Experts Group (JPEG) formats. Data were randomly separated into 80% of training and 20% of test sets, and the contrast of images in the test set was changed to 5 levels (50%, 75%, 100%, 125%, and 150%). We trained the model to detect pneumothorax using ResNet-50 with 100% level images and tested with 5-level images in the two formats. While comparing the overall performance between each contrast level in the two formats, the area under the receiver-operating characteristic curve (AUC) was significantly different (all p < 0.001) except between 125 and 150% in JPEG format (p = 0.382). When comparing the two formats at same contrast levels, AUC was significantly different (all p < 0.001) except 50% and 100% (p = 0.079 and p = 0.082, respectively). The contrast level and format of medical images could influence the performance of the deep learning model. It is required to train with various contrast levels and formats of image, and further image processing for improvement and maintenance of the performance. [ABSTRACT FROM AUTHOR]

Obesity and overweight has increased in the last year and has become a pandemic disease, the result of sedentary lifestyles and unhealthy diets rich in sugars, refined starches, fats and calories. Machine learning (ML) has proven to be very useful in the scientific community, especially in the health sector. With the aim of providing useful tools to help nutritionists and dieticians, research focused on the development of ML and Deep Learning (DL) algorithms and models is searched in the literature. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol has been used, a very common technique applied to carry out revisions. In our proposal, 17 articles have been filtered in which ML and DL are applied in the prediction of diseases, in the delineation of treatment strategies, in the improvement of personalized nutrition and more. Despite expecting better results with the use of DL, according to the selected investigations, the traditional methods are still the most used and the yields in both cases fluctuate around positive values, conditioned by the databases (transformed in each case) to a greater extent than by the artificial intelligence paradigm used. Conclusions: An important compilation is provided for the literature in this area. ML models are time-consuming to clean data, but (like DL) they allow automatic modeling of large volumes of data which makes them superior to traditional statistics. [ABSTRACT FROM AUTHOR]

Today, the internet of things (IoT) is becoming more common and finds applications in several domains, especially in the healthcare sector. Due to the rising demands of IoT, a massive quantity of sensing data gets generated from diverse sensing devices. Artificial intelligence (AI) techniques are vital for providing a scalable and precise analysis of data in real time. But the design and development of a useful big data analysis technique face a few challenges, like centralized architecture, security, and privacy, resource constraints, and the lack of adequate training data. On the other hand, the rising blockchain technology offers a decentralized architecture. It enables secure sharing of data and resources to the different nodes of the IoT network and is promoted for removing centralized control and resolving the problems of AI. This study develops an optimal deep-learning-based secure blockchain (ODLSB) enabled intelligent IoT and healthcare diagnosis model. The proposed model involves three major processes: secure transaction, hash value encryption, and medical diagnosis. The ODLSB technique comprises the orthogonal particle swarm optimization (OPSO) algorithm for the secret sharing of medical images. In addition, the hash value encryption process takes place using neighborhood indexing sequence (NIS) algorithm. At last, the optimal deep neural network (ODNN) is applied as a classification model to diagnose the diseases. The utilization of OPSO algorithm for secret sharing and optimal parameter tuning process shows the novelty of the work. We carried out detailed experiments to validate the outcome of the proposed method, and several aspects of the results are considered. At the time of the diagnosis process, the OPSO-DNN model has yielded superior results, with the highest sensitivity (92.75%), specificity (91.42%), and accuracy (93.68%). [ABSTRACT FROM AUTHOR]

There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. However, there are few clinical language models, the largest of which trained in the clinical domain is comparatively small at 110 million parameters (compared with billions of parameters in the general domain). It is not clear how large clinical language models with billions of parameters can help medical AI systems utilize unstructured EHRs. In this study, we develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text) and systematically evaluate it on five clinical NLP tasks including clinical concept extraction, medical relation extraction, semantic textual similarity, natural language inference (NLI), and medical question answering (MQA). We examine how (1) scaling up the number of parameters and (2) scaling up the size of the training data could benefit these NLP tasks. GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks (e.g., 9.6% and 9.5% improvement in accuracy for NLI and MQA), which can be applied to medical AI systems to improve healthcare delivery. The GatorTron models are publicly available at: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/models/gatortron_og. [ABSTRACT FROM AUTHOR]

The development of medical image analysis algorithm is a complex process including the multiple sub-steps of model training, data visualization, human–computer interaction and graphical user interface (GUI) construction. To accelerate the development process, algorithm developers need a software tool to assist with all the sub-steps so that they can focus on the core function implementation. Especially, for the development of deep learning (DL) algorithms, a software tool supporting training data annotation and GUI construction is highly desired. In this work, we constructed AnatomySketch, an extensible open-source software platform with a friendly GUI and a flexible plugin interface for integrating user-developed algorithm modules. Through the plugin interface, algorithm developers can quickly create a GUI-based software prototype for clinical validation. AnatomySketch supports image annotation using the stylus and multi-touch screen. It also provides efficient tools to facilitate the collaboration between human experts and artificial intelligent (AI) algorithms. We demonstrate four exemplar applications including customized MRI image diagnosis, interactive lung lobe segmentation, human-AI collaborated spine disc segmentation and Annotation-by-iterative-Deep-Learning (AID) for DL model training. Using AnatomySketch, the gap between laboratory prototyping and clinical testing is bridged and the development of MIA algorithms is accelerated. The software is opened at https://github.com/DlutMedimgGroup/AnatomySketch-Software. [ABSTRACT FROM AUTHOR]

This work presents a high-performance approach for 3D lung segmentation tasks in computer tomography images using a new intelligent machine learning algorithm called Floor of Log(FoL). The Support Vector Machine was used to learn the better parameter of the FoL algorithm using the parenchyma and its border as labels. Sensitivity, Matthews Correlation Coefficient (MCC), Hausdorff Distance (HD), Dice, Accuracy (ACC), and Jaccard were used to evaluate the proposed algorithm. The FoL was compared with recent 3D region growing, 3D Adaptive Crisp Active Contour, 3D OsiriX toolbox, and Level-set algorithm based on the coherent propagation method algorithms. The FoL algorithm achieves good results with approximately 19 s in the most significant result in an exam with 430 slices and presents similarity indexes achieving HD 3.5, DICE 83.63, and Jaccard 99.73 and qualitative indexes achieving Sensitivity 83.87, MCC 83.08, and ACC 99.62. The proposed approach of this work showed a simple and powerful algorithm to segment lung in computer tomography images of the chest region by combining similar textures, highlighting the lung structure. The FoL was presented as a new supervised clustering algorithm which can be trained to achieve better results and attached to other approaches as Convolutional Deep Neural Networks applications. [ABSTRACT FROM AUTHOR]

Cardiac structure contouring is a time consuming and tedious manual activity used for radiotherapeutic dose toxicity planning. We developed an automatic cardiac structure segmentation pipeline for use in low-dose non-contrast planning CT based on deep learning algorithms for small datasets. Fifty CT scans were retrospectively selected and the whole heart, ventricles and atria were contoured. A two stage deep learning pipeline was trained on 41 non contrast planning CTs, tuned with 3 CT scans and validated on 6 CT scans. In the first stage, An InceptionResNetV2 network was used to identify the slices that contained cardiac structures. The second stage consisted of three deep learning models trained on the images containing cardiac structures to segment the structures. The three deep learning models predicted the segmentations/contours on axial, coronal and sagittal images and are combined to create the final prediction. The final accuracy of the pipeline was quantified on 6 volumes by calculating the Dice similarity coefficient (DC), 95% Hausdorff distance (95% HD) and volume ratios between predicted and ground truth volumes. Median DC and 95% HD of 0.96, 0.88, 0.92, 0.80 and 0.82, and 1.86, 2.98, 2.02, 6.16 and 6.46 were achieved for the whole heart, right and left ventricle, and right and left atria respectively. The median differences in volume were -4, -1, + 5, -16 and -20% for the whole heart, right and left ventricle, and right and left atria respectively. The automatic contouring pipeline achieves good results for whole heart and ventricles. Robust automatic contouring with deep learning methods seems viable for local centers with small datasets. [ABSTRACT FROM AUTHOR]

Preparing radiology examinations for interpretation requires prefetching relevant prior examinations and implementing hanging protocols to optimally display the examination along with comparisons. Body part is a critical piece of information to facilitate both prefetching and hanging protocols, but body part information encoded using the Digital Imaging and Communications in Medicine (DICOM) standard is widely variable, error-prone, not granular enough, or missing altogether. This results in inappropriate examinations being prefetched or relevant examinations left behind; hanging protocol optimization suffers as well. Modern artificial intelligence (AI) techniques, particularly when harnessing federated deep learning techniques, allow for highly accurate automatic detection of body part based on the image data within a radiological examination; this allows for much more reliable implementation of this categorization and workflow. Additionally, new avenues to further optimize examination viewing such as dynamic hanging protocol and image display can be implemented using these techniques. [ABSTRACT FROM AUTHOR]

Objectives: To investigate the image quality and perception of a sinogram-based deep learning image reconstruction (DLIR) algorithm for single-energy abdominal CT compared to standard-of-care strength of ASIR-V. Methods: In this retrospective study, 50 patients (62% F; 56.74 ± 17.05 years) underwent portal venous phase. Four reconstructions (ASIR-V at 40%, and DLIR at three strengths: low (DLIR-L), medium (DLIR-M), and high (DLIR-H)) were generated. Qualitative and quantitative image quality analysis was performed on the 200 image datasets. Qualitative scores were obtained for image noise, contrast, small structure visibility, sharpness, and artifact by three blinded radiologists on a 5-point scale (1, excellent; 5, very poor). Radiologists also indicated image preference on a 3-point scale (1, most preferred; 3, least preferred). Quantitative assessment was performed by measuring image noise and contrast-to-noise ratio (CNR). Results: DLIR had better image quality scores compared to ASIR-V. Scores on DLIR-H for noise (1.40 ± 0.53), contrast (1.41 ± 0.55), small structure visibility (1.51 ± 0.61), and sharpness (1.60 ± 0.54) were the best (p < 0.05) followed by DLIR-M (1.85 ± 0.52, 1.66 ± 0.57, 1.69 ± 0.59, 1.68 ± 0.46), DLIR-L (2.29 ± 0.58, 1.96 ± 0.61, 1.90 ± 0.65, 1.86 ± 0.46), and ASIR-V (2.86 ± 0.67, 2.55 ± 0.58, 2.34 ± 0.66, 2.01 ± 0.36). Ratings for artifacts were similar for all reconstructions (p > 0.05). DLIRs did not influence subjective textural perceptions and were preferred over ASIR-V from the beginning. All DLIRs had a higher CNR (26.38–102.30%) and lower noise (20.64–48.77%) than ASIR-V. DLIR-H had the best objective scores. Conclusion: Sinogram-based deep learning image reconstructions were preferred over iterative reconstruction subjectively and objectively due to improved image quality and lower noise, even in large patients. Use in clinical routine may allow for radiation dose reduction. Key Points: • Deep learning image reconstructions (DLIRs) have a higher contrast-to-noise ratio compared to medium-strength hybrid iterative reconstruction techniques. • DLIR may be advantageous in patients with large body habitus due to a lower image noise. • DLIR can enable further optimization of radiation doses used in abdominal CT. [ABSTRACT FROM AUTHOR]

Purpose: A deep learning artificial intelligence (AI) algorithm has been demonstrated to outperform embryologists in identifying euploid embryos destined to implant with an accuracy of 75.3% (1). Our aim was to evaluate the performance of highly trained embryologists in selecting top quality day 5 euploid blastocysts with and without the aid of a deep learning algorithm. Materials and methods: A non-overlapping series of 200 sets of day 5 euploid embryo images with known implantation outcomes was distributed to 17 highly trained embryologists. One embryo in each set was known to have implanted and one failed implantation. They were asked to select which embryo to transfer from each set. The same 200 sets of embryos, with indication of which embryo in each set had been identified by the algorithm as more likely to implant was then distributed. Chi-squared, t-test, and receiver operating curves were performed to compare the embryologist performeance with and without AI. Results: Fourteen embryologists completed both assessments. Embryologists provided with AI results selected successfully implanted embryos in 73.6% of cases compared to 65.5% for those selected using visual assessments alone (p < 0.001). All embryologists improved in their ability to select embryos with the aid of the AI algorithm with a mean percent improvement of 11.1% (range 1.4% to 15.5%). There were no differences in degree of improvement by embryologist level of experience (junior, intermediate, senior). Conclusions: The incorporation of an AI framework for blastocyst selection enhanced the performance of trained embryologists in identifying PGT-A euploid embryos destined to implant. [ABSTRACT FROM AUTHOR]

The objective of the study was to determine if the pathology depicted on a mammogram is either benign or malignant (ductal or non-ductal carcinoma) using deep learning and artificial intelligence techniques. A total of 559 patients underwent breast ultrasound, mammography, and ultrasound-guided breast biopsy. Based on the histopathological results, the patients were divided into three categories: benign, ductal carcinomas, and non-ductal carcinomas. The mammograms in the cranio-caudal view underwent pre-processing and segmentation. Given the large variability of the areola, an algorithm was used to remove it and the adjacent skin. Therefore, patients with breast lesions close to the skin were removed. The remaining breast image was resized on the Y axis to a square image and then resized to 512 × 512 pixels. A variable square of 322,622 pixels was searched inside every image to identify the lesion. Each image was rotated with no information loss. For data augmentation, each image was rotated 360 times and a crop of 227 × 227 pixels was saved, resulting in a total of 201,240 images. The reason why our images were cropped at this size is because the deep learning algorithm transfer learning used from AlexNet network has an input image size of 227 × 227. The mean accuracy was 95.8344% ± 6.3720% and mean AUC 0.9910% ± 0.0366%, computed on 100 runs of the algorithm. Based on the results, the proposed solution can be used as a non-invasive and highly accurate computer-aided system based on deep learning that can classify breast lesions based on changes identified on mammograms in the cranio-caudal view. [ABSTRACT FROM AUTHOR]

This study aimed to develop a method for detection of femoral neck fracture (FNF) including displaced and non-displaced fractures using convolutional neural network (CNN) with plain X-ray and to validate its use across hospitals through internal and external validation sets. This is a retrospective study using hip and pelvic anteroposterior films for training and detecting femoral neck fracture through residual neural network (ResNet) 18 with convolutional block attention module (CBAM) + +. The study was performed at two tertiary hospitals between February and May 2020 and used data from January 2005 to December 2018. Our primary outcome was favorable performance for diagnosis of femoral neck fracture from negative studies in our dataset. We described the outcomes as area under the receiver operating characteristic curve (AUC), accuracy, Youden index, sensitivity, and specificity. A total of 4,189 images that contained 1,109 positive images (332 non-displaced and 777 displaced) and 3,080 negative images were collected from two hospitals. The test values after training with one hospital dataset were 0.999 AUC, 0.986 accuracy, 0.960 Youden index, and 0.966 sensitivity, and 0.993 specificity. Values of external validation with the other hospital dataset were 0.977, 0.971, 0.920, 0.939, and 0.982, respectively. Values of merged hospital datasets were 0.987, 0.983, 0.960, 0.973, and 0.987, respectively. A CNN algorithm for FNF detection in both displaced and non-displaced fractures using plain X-rays could be used in other hospitals to screen for FNF after training with images from the hospital of interest. [ABSTRACT FROM AUTHOR]

Objectives: To develop and evaluate a deep learning–based model capable of detecting primary hepatic malignancies in multiphase CT images of patients at high risk for hepatocellular carcinoma (HCC). Methods: A total of 1350 multiphase CT scans of 1280 hepatic malignancies (1202 HCCs and 78 non-HCCs) in 1320 patients at high risk for HCC were retrospectively analyzed. Following the delineation of the focal hepatic lesions according to reference standards, the CT scans were categorized randomly into the training (568 scans), tuning (193 scans), and test (589 scans) sets. Multiphase CT information was subjected to multichannel integration, and livers were automatically segmented before model development. A deep learning–based model capable of detecting malignancies was developed using a mask region–based convolutional neural network. The thresholds of the prediction score and the intersection over union were determined on the tuning set corresponding to the highest sensitivity with < 5 false-positive cases per CT scan. The sensitivity and the number of false-positives of the proposed model on the test set were calculated. Potential causes of false-negatives and false-positives on the test set were analyzed. Results: This model exhibited a sensitivity of 84.8% with 4.80 false-positives per CT scan on the test set. The most frequent potential causes of false-negatives and false-positives were determined to be atypical enhancement patterns for HCC (71.7%) and registration/segmentation errors (42.7%), respectively. Conclusions: The proposed deep learning–based model developed to automatically detect primary hepatic malignancies exhibited an 84.8% of sensitivity with 4.80 false-positives per CT scan in the test set. Key Points: • Image processing, including multichannel integration of multiphase CT and automatic liver segmentation, enabled the application of a deep learning–based model to detect primary hepatic malignancy. • Our model exhibited a sensitivity of 84.8% with a false-positive rate of 4.80 per CT scan. [ABSTRACT FROM AUTHOR]