State-of-the-art review on deep learning in medical imaging

Convolutional neural network with parallel convolution scale attention module and ResCBAM for breast histology image classification

Breast cancer is the most common cause of female morbidity and death worldwide. Compared with other cancers, early detection of breast cancer is more helpful to improve the prognosis of patients. In order to achieve early diagnosis and treatment, clinical treatment requires rapid and accurate diagnosis. Therefore, the development of an automatic detection system for breast cancer suitable for patient imaging is of great significance for assisting clinical treatment. Accurate classification of pathological images plays a key role in computer-aided medical diagnosis and prognosis. However, in the automatic recognition and classification methods of breast cancer pathological images, the scale information, the loss of image information caused by insufficient feature fusion, and the enormous structure of the model may lead to inaccurate or inefficient classification. To minimize the impact, we proposed a lightweight PCSAM-ResCBAM model based on two-stage convolutional neural network. The model included a Parallel Convolution Scale Attention Module network (PCSAM-Net) and a Residual Convolutional Block Attention Module network (ResCBAM-Net). The first-level convolutional network was built through a 4-layer PCSAM module to achieve prediction and classification of patches extracted from images. To optimize the network's ability to represent global features of images, we proposed a tiled feature fusion method to fuse patch features from the same image, and proposed a residual convolutional attention module. Based on the above, the second-level convolutional network was constructed to achieve predictive classification of images. We evaluated the performance of our proposed model on the ICIAR2018 dataset and the BreakHis dataset, respectively. Furthermore, through model ablation studies, we found that scale attention and dilated convolution play an important role in improving model performance. Our proposed model outperforms the existing state-of-the-art models on 200 × and 400 × magnification datasets with a maximum accuracy of 98.74 %.

Polygenic Risk Score for Cardiovascular Diseases in Artificial Intelligence Paradigm: A Review

Cardiovascular disease (CVD) related mortality and morbidity heavily strain society. The relationship between external risk factors and our genetics have not been well established. It is widely acknowledged that environmental influence and individual behaviours play a significant role in CVD vulnerability, leading to the development of polygenic risk scores (PRS). We employed the PRISMA search method to locate pertinent research and literature to extensively review artificial intelligence (AI)-based PRS models for CVD risk prediction. Furthermore, we analyzed and compared conventional vs. AI-based solutions for PRS. We summarized the recent advances in our understanding of the use of AI-based PRS for risk prediction of CVD. Our study proposes three hypotheses: i) Multiple genetic variations and risk factors can be incorporated into AI-based PRS to improve the accuracy of CVD risk predicting. ii) AI-based PRS for CVD circumvents the drawbacks of conventional PRS calculators by incorporating a larger variety of genetic and non-genetic components, allowing for more precise and individualised risk estimations. iii) Using AI approaches, it is possible to significantly reduce the dimensionality of huge genomic datasets, resulting in more accurate and effective disease risk prediction models. Our study highlighted that the AI-PRS model outperformed traditional PRS calculators in predicting CVD risk. Furthermore, using AI-based methods to calculate PRS may increase the precision of risk predictions for CVD and have significant ramifications for individualized prevention and treatment plans.

Artificial intelligence-based preventive, personalized and precision medicine for cardiovascular disease/stroke risk assessment in rheumatoid arthritis patients: a narrative review

The challenges associated with diagnosing and treating cardiovascular disease (CVD)/Stroke in Rheumatoid arthritis (RA) arise from the delayed onset of symptoms. Existing clinical risk scores are inadequate in predicting cardiac events, and conventional risk factors alone do not accurately classify many individuals at risk. Several CVD biomarkers consider the multiple pathways involved in the development of atherosclerosis, which is the primary cause of CVD/Stroke in RA. To enhance the accuracy of CVD/Stroke risk assessment in the RA framework, a proposed approach involves combining genomic-based biomarkers (GBBM) derived from plasma and/or serum samples with innovative non-invasive radiomic-based biomarkers (RBBM), such as measurements of synovial fluid, plaque area, and plaque burden. This review presents two hypotheses: (i) RBBM and GBBM biomarkers exhibit a significant correlation and can precisely detect the severity of CVD/Stroke in RA patients. (ii) Artificial Intelligence (AI)-based preventive, precision, and personalized (aiP3) CVD/Stroke risk AtheroEdge™ model (AtheroPoint™, CA, USA) that utilizes deep learning (DL) to accurately classify the risk of CVD/stroke in RA framework. The authors conducted a comprehensive search using the PRISMA technique, identifying 153 studies that assessed the features/biomarkers of RBBM and GBBM for CVD/Stroke. The study demonstrates how DL models can be integrated into the AtheroEdge™-aiP3 framework to determine the risk of CVD/Stroke in RA patients. The findings of this review suggest that the combination of RBBM with GBBM introduces a new dimension to the assessment of CVD/Stroke risk in the RA framework. Synovial fluid levels that are higher than normal lead to an increase in the plaque burden. Additionally, the review provides recommendations for novel, unbiased, and pruned DL algorithms that can predict CVD/Stroke risk within a RA framework that is preventive, precise, and personalized.

Combining Molecular and Radiomic Features for Risk Assessment in Breast Cancer

Breast cancer risk is highly variable within the population and current research is leading the shift toward personalized medicine. By accurately assessing an individual woman's risk, we can reduce the risk of over/undertreatment by preventing unnecessary procedures or by elevating screening procedures. Breast density measured from conventional mammography has been established as one of the most dominant risk factors for breast cancer; however, it is currently limited by its ability to characterize more complex breast parenchymal patterns that have been shown to provide additional information to strengthen cancer risk models. Molecular factors ranging from high penetrance, or high likelihood that a mutation will show signs and symptoms of the disease, to combinations of gene mutations with low penetrance have shown promise for augmenting risk assessment. Although imaging biomarkers and molecular biomarkers have both individually demonstrated improved performance in risk assessment, few studies have evaluated them together. This review aims to highlight the current state of the art in breast cancer risk assessment using imaging and genetic biomarkers.

Ensemble Deep Learning Derived from Transfer Learning for Classification of COVID-19 Patients on Hybrid Deep-Learning-Based Lung Segmentation: A Data Augmentation and Balancing Framework

BACKGROUND AND MOTIVATION

Lung computed tomography (CT) techniques are high-resolution and are well adopted in the intensive care unit (ICU) for COVID-19 disease control classification. Most artificial intelligence (AI) systems do not undergo generalization and are typically overfitted. Such trained AI systems are not practical for clinical settings and therefore do not give accurate results when executed on unseen data sets. We hypothesize that ensemble deep learning (EDL) is superior to deep transfer learning (TL) in both non-augmented and augmented frameworks.

METHODOLOGY

The system consists of a cascade of quality control, ResNet-UNet-based hybrid deep learning for lung segmentation, and seven models using TL-based classification followed by five types of EDL's. To prove our hypothesis, five different kinds of data combinations (DC) were designed using a combination of two multicenter cohorts-Croatia (80 COVID) and Italy (72 COVID and 30 controls)-leading to 12,000 CT slices. As part of generalization, the system was tested on unseen data and statistically tested for reliability/stability.

RESULTS

Using the K5 (80:20) cross-validation protocol on the balanced and augmented dataset, the five DC datasets improved TL mean accuracy by 3.32%, 6.56%, 12.96%, 47.1%, and 2.78%, respectively. The five EDL systems showed improvements in accuracy of 2.12%, 5.78%, 6.72%, 32.05%, and 2.40%, thus validating our hypothesis. All statistical tests proved positive for reliability and stability.

CONCLUSION

EDL showed superior performance to TL systems for both (a) unbalanced and unaugmented and (b) balanced and augmented datasets for both (i) seen and (ii) unseen paradigms, validating both our hypotheses.

Analytical models for external photon beam radiotherapy out-of-field dose calculation: a scoping review

A growing body of scientific evidence indicates that exposure to low dose ionizing radiation (< 2 Gy) is associated with a higher risk of developing radio-induced cancer. Additionally, it has been shown to have significant impacts on both innate and adaptive immune responses. As a result, the evaluation of the low doses inevitably delivered outside the treatment fields (out-of-field dose) in photon radiotherapy is a topic that is regaining interest at a pivotal moment in radiotherapy. In this work, we proposed a scoping review in order to identify evidence of strengths and limitations of available analytical models for out-of-field dose calculation in external photon beam radiotherapy for the purpose of implementation in clinical routine. Papers published between 1988 and 2022 proposing a novel analytical model that estimated at least one component of the out-of-field dose for photon external radiotherapy were included. Models focusing on electrons, protons and Monte-Carlo methods were excluded. The methodological quality and potential limitations of each model were analyzed to assess their generalizability. Twenty-one published papers were selected for analysis, of which 14 proposed multi-compartment models, demonstrating that research efforts are directed towards an increasingly detailed description of the underlying physical phenomena. Our synthesis revealed great inhomogeneities in practices, in particular in the acquisition of experimental data and the standardization of measurements, in the choice of metrics used for the evaluation of model performance and even in the definition of regions considered out-of-the-field, which makes quantitative comparisons impossible. We therefore propose to clarify some key concepts. The analytical methods do not seem to be easily suitable for massive use in clinical routine, due to the inevitable cumbersome nature of their implementation. Currently, there is no consensus on a mathematical formalism that comprehensively describes the out-of-field dose in external photon radiotherapy, partly due to the complex interactions between a large number of influencing factors. Out-of-field dose calculation models based on neural networks could be promising tools to overcome these limitations and thus favor a transfer to the clinic, but the lack of sufficiently large and heterogeneous data sets is the main obstacle.

Deep Learning Algorithms for Estimation of Demographic and Anthropometric Features from Electrocardiograms

The electrocardiogram (ECG) has been known to be affected by demographic and anthropometric factors. This study aimed to develop deep learning models to predict the subject's age, sex, ABO blood type, and body mass index (BMI) based on ECGs. This retrospective study included individuals aged 18 years or older who visited a tertiary referral center with ECGs acquired from October 2010 to February 2020. Using convolutional neural networks (CNNs) with three convolutional layers, five kernel sizes, and two pooling sizes, we developed both classification and regression models. We verified a classification model to be applicable for age (<40 years vs. ≥40 years), sex (male vs. female), BMI (<25 kg/m² vs. ≥25 kg/m²), and ABO blood type. A regression model was also developed and validated for age and BMI estimation. A total of 124,415 ECGs (1 ECG per subject) were included. The dataset was constructed by dividing the entire set of ECGs at a ratio of 4:3:3. In the classification task, the area under the receiver operating characteristic (AUROC), which represents a quantitative indicator of the judgment threshold, was used as the primary outcome. The mean absolute error (MAE), which represents the difference between the observed and estimated values, was used in the regression task. For age estimation, the CNN achieved an AUROC of 0.923 with an accuracy of 82.97%, and a MAE of 8.410. For sex estimation, the AUROC was 0.947 with an accuracy of 86.82%. For BMI estimation, the AUROC was 0.765 with an accuracy of 69.89%, and a MAE of 2.332. For ABO blood type estimation, the CNN showed an inferior performance, with a top-1 accuracy of 31.98%. For the ABO blood type estimation, the CNN showed an inferior performance, with a top-1 accuracy of 31.98% (95% CI, 31.98-31.98%). Our model could be adapted to estimate individuals' demographic and anthropometric features from their ECGs; this would enable the development of physiologic biomarkers that can better reflect their health status than chronological age.

Role of Ensemble Deep Learning for Brain Tumor Classification in Multiple Magnetic Resonance Imaging Sequence Data

The biopsy is a gold standard method for tumor grading. However, due to its invasive nature, it has sometimes proved fatal for brain tumor patients. As a result, a non-invasive computer-aided diagnosis (CAD) tool is required. Recently, many magnetic resonance imaging (MRI)-based CAD tools have been proposed for brain tumor grading. The MRI has several sequences, which can express tumor structure in different ways. However, a suitable MRI sequence for brain tumor classification is not yet known. The most common brain tumor is 'glioma', which is the most fatal form. Therefore, in the proposed study, to maximize the classification ability between low-grade versus high-grade glioma, three datasets were designed comprising three MRI sequences: T1-Weighted (T1W), T2-weighted (T2W), and fluid-attenuated inversion recovery (FLAIR). Further, five well-established convolutional neural networks, AlexNet, VGG16, ResNet18, GoogleNet, and ResNet50 were adopted for tumor classification. An ensemble algorithm was proposed using the majority vote of above five deep learning (DL) models to produce more consistent and improved results than any individual model. Five-fold cross validation (K5-CV) protocol was adopted for training and testing. For the proposed ensembled classifier with K5-CV, the highest test accuracies of 98.88 ± 0.63%, 97.98 ± 0.86%, and 94.75 ± 0.61% were achieved for FLAIR, T2W, and T1W-MRI data, respectively. FLAIR-MRI data was found to be most significant for brain tumor classification, where it showed a 4.17% and 0.91% improvement in accuracy against the T1W-MRI and T2W-MRI sequence data, respectively. The proposed ensembled algorithm (MajVot) showed significant improvements in the average accuracy of three datasets of 3.60%, 2.84%, 1.64%, 4.27%, and 1.14%, respectively, against AlexNet, VGG16, ResNet18, GoogleNet, and ResNet50.

Ischemic stroke prediction of patients with carotid atherosclerotic stenosis via multi-modality fused network

Carotid atherosclerotic stenosis of the carotid artery is an important cause of ischemic cerebrovascular disease. The aim of this study was to predict the presence or absence of clinical symptoms in unknown patients by studying the existence or lack of symptoms of patients with carotid atherosclerotic stenosis. First, a deep neural network prediction model based on brain MRI imaging data of patients with multiple modalities is constructed; it uses the multi-modality features extracted from the neural network as inputs and the incidence of diagnosis as output to train the model. Then, a machine learning-based classification algorithm is developed to utilize the clinical features for comparison and evaluation. The experimental results showed that the deep learning model using imaging data could better predict the clinical symptom classification of patients. As part of preventive medicine, this study could help patients with carotid atherosclerosis narrowing to prepare for stroke prevention based on the prediction results.

Application of Deep Learning in Histopathology Images of Breast Cancer: A Review

With the development of artificial intelligence technology and computer hardware functions, deep learning algorithms have become a powerful auxiliary tool for medical image analysis. This study was an attempt to use statistical methods to analyze studies related to the detection, segmentation, and classification of breast cancer in pathological images. After an analysis of 107 articles on the application of deep learning to pathological images of breast cancer, this study is divided into three directions based on the types of results they report: detection, segmentation, and classification. We introduced and analyzed models that performed well in these three directions and summarized the related work from recent years. Based on the results obtained, the significant ability of deep learning in the application of breast cancer pathological images can be recognized. Furthermore, in the classification and detection of pathological images of breast cancer, the accuracy of deep learning algorithms has surpassed that of pathologists in certain circumstances. Our study provides a comprehensive review of the development of breast cancer pathological imaging-related research and provides reliable recommendations for the structure of deep learning network models in different application scenarios.

Economics of Artificial Intelligence in Healthcare: Diagnosis vs. Treatment

Motivation: The price of medical treatment continues to rise due to (i) an increasing population; (ii) an aging human growth; (iii) disease prevalence; (iv) a rise in the frequency of patients that utilize health care services; and (v) increase in the price. Objective: Artificial Intelligence (AI) is already well-known for its superiority in various healthcare applications, including the segmentation of lesions in images, speech recognition, smartphone personal assistants, navigation, ride-sharing apps, and many more. Our study is based on two hypotheses: (i) AI offers more economic solutions compared to conventional methods; (ii) AI treatment offers stronger economics compared to AI diagnosis. This novel study aims to evaluate AI technology in the context of healthcare costs, namely in the areas of diagnosis and treatment, and then compare it to the traditional or non-AI-based approaches. Methodology: PRISMA was used to select the best 200 studies for AI in healthcare with a primary focus on cost reduction, especially towards diagnosis and treatment. We defined the diagnosis and treatment architectures, investigated their characteristics, and categorized the roles that AI plays in the diagnostic and therapeutic paradigms. We experimented with various combinations of different assumptions by integrating AI and then comparing it against conventional costs. Lastly, we dwell on three powerful future concepts of AI, namely, pruning, bias, explainability, and regulatory approvals of AI systems. Conclusions: The model shows tremendous cost savings using AI tools in diagnosis and treatment. The economics of AI can be improved by incorporating pruning, reduction in AI bias, explainability, and regulatory approvals.

Cardiovascular/Stroke Risk Stratification in Diabetic Foot Infection Patients Using Deep Learning-Based Artificial Intelligence: An Investigative Study

A diabetic foot infection (DFI) is among the most serious, incurable, and costly to treat conditions. The presence of a DFI renders machine learning (ML) systems extremely nonlinear, posing difficulties in CVD/stroke risk stratification. In addition, there is a limited number of well-explained ML paradigms due to comorbidity, sample size limits, and weak scientific and clinical validation methodologies. Deep neural networks (DNN) are potent machines for learning that generalize nonlinear situations. The objective of this article is to propose a novel investigation of deep learning (DL) solutions for predicting CVD/stroke risk in DFI patients. The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) search strategy was used for the selection of 207 studies. We hypothesize that a DFI is responsible for increased morbidity and mortality due to the worsening of atherosclerotic disease and affecting coronary artery disease (CAD). Since surrogate biomarkers for CAD, such as carotid artery disease, can be used for monitoring CVD, we can thus use a DL-based model, namely, Long Short-Term Memory (LSTM) and Recurrent Neural Networks (RNN) for CVD/stroke risk prediction in DFI patients, which combines covariates such as office and laboratory-based biomarkers, carotid ultrasound image phenotype (CUSIP) lesions, along with the DFI severity. We confirmed the viability of CVD/stroke risk stratification in the DFI patients. Strong designs were found in the research of the DL architectures for CVD/stroke risk stratification. Finally, we analyzed the AI bias and proposed strategies for the early diagnosis of CVD/stroke in DFI patients. Since DFI patients have an aggressive atherosclerotic disease, leading to prominent CVD/stroke risk, we, therefore, conclude that the DL paradigm is very effective for predicting the risk of CVD/stroke in DFI patients.

Deep learning artificial intelligence framework for multiclass coronary artery disease prediction using combination of conventional risk factors, carotid ultrasound, and intraplaque neovascularization

OBJECTIVE

Cardiovascular disease (CVD) is a major healthcare challenge and therefore early risk assessment is vital. Previous assessment techniques use either "conventional CVD risk calculators (CCVRC)" or machine learning (ML) paradigms. These techniques are ad-hoc, unreliable, not fully automated, and have variabilities. We, therefore, introduce AtheroEdge-MCDLAI (AE3.0DL) windows-based platform using multiclass Deep Learning (DL) system.

METHODS

Data was collected on 500 patients having both carotid ultrasound and corresponding coronary angiography scores (CAS), measured as stenosis in coronary arteries and considered as the gold standard. A total of 39 covariates were used, clubbed into three clusters, namely (i) Office-based: age, gender, body mass index, smoker, hypertension, systolic blood pressure, and diastolic blood pressure; (ii) Laboratory-based: Hyperlipidemia, hemoglobin A1c, and estimated glomerular filtration rate; and (iii) Carotid ultrasound image phenotypes: maximum plaque height, total plaque area, and intra-plaque neovascularization. Baseline characteristics for four classes (target labels) having significant (p < 0.0001) values were calculated using Chi-square and ANOVA. For handling the cohort's imbalance in the risk classes, AE3.0DL used the synthetic minority over-sampling technique (SMOTE). AE3.0DL used Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) DL models and the performance (accuracy and area-under-the-curve) was computed using 10-fold cross-validation (90% training, 10% testing) frameworks. AE3.0DL was validated and benchmarked.

RESULTS

The AE3.0DL using RNN and LSTM showed an accuracy and AUC (p < 0.0001) pairs as (95.00% and 0.98), and (95.34% and 0.99), respectively, and showed an improvement of 32.93% and 9.94% against CCVRC and ML, respectively. AE3.0DL runs in <1 s.

CONCLUSION

DL algorithms are a powerful paradigm for coronary artery disease (CAD) risk prediction and CVD risk stratification.

Influence of contrast and texture based image modifications on the performance and attention shift of U-Net models for brain tissue segmentation

Contrast and texture modifications applied during training or test-time have recently shown promising results to enhance the generalization performance of deep learning segmentation methods in medical image analysis. However, a deeper understanding of this phenomenon has not been investigated. In this study, we investigated this phenomenon using a controlled experimental setting, using datasets from the Human Connectome Project and a large set of simulated MR protocols, in order to mitigate data confounders and investigate possible explanations as to why model performance changes when applying different levels of contrast and texture-based modifications. Our experiments confirm previous findings regarding the improved performance of models subjected to contrast and texture modifications employed during training and/or testing time, but further show the interplay when these operations are combined, as well as the regimes of model improvement/worsening across scanning parameters. Furthermore, our findings demonstrate a spatial attention shift phenomenon of trained models, occurring for different levels of model performance, and varying in relation to the type of applied image modification.

Deep Learning Hybrid Techniques for Brain Tumor Segmentation

Medical images play an important role in medical diagnosis and treatment. Oncologists analyze images to determine the different characteristics of deadly diseases, plan the therapy, and observe the evolution of the disease. The objective of this paper is to propose a method for the detection of brain tumors. Brain tumors are identified from Magnetic Resonance (MR) images by performing suitable segmentation procedures. The latest technical literature concerning radiographic images of the brain shows that deep learning methods can be implemented to extract specific features of brain tumors, aiding clinical diagnosis. For this reason, most data scientists and AI researchers work on Machine Learning methods for designing automatic screening procedures. Indeed, an automated method would result in quicker segmentation findings, providing a robust output with respect to possible differences in data sources, mostly due to different procedures in data recording and storing, resulting in a more consistent identification of brain tumors. To improve the performance of the segmentation procedure, new architectures are proposed and tested in this paper. We propose deep neural networks for the detection of brain tumors, trained on the MRI scans of patients' brains. The proposed architectures are based on convolutional neural networks and inception modules for brain tumor segmentation. A comparison of these proposed architectures with the baseline reference ones shows very interesting results. MI-Unet showed a performance increase in comparison to baseline Unet architecture by 7.5% in dice score, 23.91% insensitivity, and 7.09% in specificity. Depth-wise separable MI-Unet showed a performance increase by 10.83% in dice score, 2.97% in sensitivity, and 12.72% in specificity as compared to the baseline Unet architecture. Hybrid Unet architecture achieved performance improvement of 9.71% in dice score, 3.56% in sensitivity, and 12.6% in specificity. Whereas the depth-wise separable hybrid Unet architecture outperformed the baseline architecture by 15.45% in dice score, 20.56% in sensitivity, and 12.22% in specificity.

Deep learning-based quantitative estimation of lymphedema-induced fibrosis using three-dimensional computed tomography images

In lymphedema, proinflammatory cytokine-mediated progressive cascades always occur, leading to macroscopic fibrosis. However, no methods are practically available for measuring lymphedema-induced fibrosis before its deterioration. Technically, CT can visualize fibrosis in superficial and deep locations. For standardized measurement, verification of deep learning (DL)-based recognition was performed. A cross-sectional, observational cohort trial was conducted. After narrowing window width of the absorptive values in CT images, SegNet-based semantic segmentation model of every pixel into 5 classes (air, skin, muscle/water, fat, and fibrosis) was trained (65%), validated (15%), and tested (20%). Then, 4 indices were formulated and compared with the standardized circumference difference ratio (SCDR) and bioelectrical impedance (BEI) results. In total, 2138 CT images of 27 chronic unilateral lymphedema patients were analyzed. Regarding fibrosis segmentation, the mean boundary F1 score and accuracy were 0.868 and 0.776, respectively. Among 19 subindices of the 4 indices, 73.7% were correlated with the BEI (partial correlation coefficient: 0.420–0.875), and 13.2% were correlated with the SCDR (0.406–0.460). The mean subindex of Index 2 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\frac{{P_{Fibrosis\, in\, Affected} - P_{Fibrosis\, in\, Unaffected} }}{{P_{Limb\, in\, Unaffected} }}} \right)$$\end{document}PFibrosisinAffected-PFibrosisinUnaffectedPLimbinUnaffected presented the highest correlation. DL has potential applications in CT image-based lymphedema-induced fibrosis recognition. The subtraction-type formula might be the most promising estimation method.

Brain Tumor Characterization Using Radiogenomics in Artificial Intelligence Framework

Brain tumor characterization (BTC) is the process of knowing the underlying cause of brain tumors and their characteristics through various approaches such as tumor segmentation, classification, detection, and risk analysis. The substantial brain tumor characterization includes the identification of the molecular signature of various useful genomes whose alteration causes the brain tumor. The radiomics approach uses the radiological image for disease characterization by extracting quantitative radiomics features in the artificial intelligence (AI) environment. However, when considering a higher level of disease characteristics such as genetic information and mutation status, the combined study of "radiomics and genomics" has been considered under the umbrella of "radiogenomics". Furthermore, AI in a radiogenomics' environment offers benefits/advantages such as the finalized outcome of personalized treatment and individualized medicine. The proposed study summarizes the brain tumor's characterization in the prospect of an emerging field of research, i.e., radiomics and radiogenomics in an AI environment, with the help of statistical observation and risk-of-bias (RoB) analysis. The PRISMA search approach was used to find 121 relevant studies for the proposed review using IEEE, Google Scholar, PubMed, MDPI, and Scopus. Our findings indicate that both radiomics and radiogenomics have been successfully applied aggressively to several oncology applications with numerous advantages. Furthermore, under the AI paradigm, both the conventional and deep radiomics features have made an impact on the favorable outcomes of the radiogenomics approach of BTC. Furthermore, risk-of-bias (RoB) analysis offers a better understanding of the architectures with stronger benefits of AI by providing the bias involved in them.

Multicenter Study on COVID-19 Lung Computed Tomography Segmentation with varying Glass Ground Opacities using Unseen Deep Learning Artificial Intelligence Paradigms: COVLIAS 1.0 Validation

Variations in COVID-19 lesions such as glass ground opacities (GGO), consolidations, and crazy paving can compromise the ability of solo-deep learning (SDL) or hybrid-deep learning (HDL) artificial intelligence (AI) models in predicting automated COVID-19 lung segmentation in Computed Tomography (CT) from unseen data leading to poor clinical manifestations. As the first study of its kind, “COVLIAS 1.0-Unseen” proves two hypotheses, (i) contrast adjustment is vital for AI, and (ii) HDL is superior to SDL. In a multicenter study, 10,000 CT slices were collected from 72 Italian (ITA) patients with low-GGO, and 80 Croatian (CRO) patients with high-GGO. Hounsfield Units (HU) were automatically adjusted to train the AI models and predict from test data, leading to four combinations—two Unseen sets: (i) train-CRO:test-ITA, (ii) train-ITA:test-CRO, and two Seen sets: (iii) train-CRO:test-CRO, (iv) train-ITA:test-ITA. COVILAS used three SDL models: PSPNet, SegNet, UNet and six HDL models: VGG-PSPNet, VGG-SegNet, VGG-UNet, ResNet-PSPNet, ResNet-SegNet, and ResNet-UNet. Two trained, blinded senior radiologists conducted ground truth annotations. Five types of performance metrics were used to validate COVLIAS 1.0-Unseen which was further benchmarked against MedSeg, an open-source web-based system. After HU adjustment for DS and JI, HDL (Unseen AI) > SDL (Unseen AI) by 4% and 5%, respectively. For CC, HDL (Unseen AI) > SDL (Unseen AI) by 6%. The COVLIAS-MedSeg difference was < 5%, meeting regulatory guidelines.Unseen AI was successfully demonstrated using automated HU adjustment. HDL was found to be superior to SDL.

Vascular Implications of COVID-19: Role of Radiological Imaging, Artificial Intelligence, and Tissue Characterization: A Special Report

The SARS-CoV-2 virus has caused a pandemic, infecting nearly 80 million people worldwide, with mortality exceeding six million. The average survival span is just 14 days from the time the symptoms become aggressive. The present study delineates the deep-driven vascular damage in the pulmonary, renal, coronary, and carotid vessels due to SARS-CoV-2. This special report addresses an important gap in the literature in understanding (i) the pathophysiology of vascular damage and the role of medical imaging in the visualization of the damage caused by SARS-CoV-2, and (ii) further understanding the severity of COVID-19 using artificial intelligence (AI)-based tissue characterization (TC). PRISMA was used to select 296 studies for AI-based TC. Radiological imaging techniques such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound were selected for imaging of the vasculature infected by COVID-19. Four kinds of hypotheses are presented for showing the vascular damage in radiological images due to COVID-19. Three kinds of AI models, namely, machine learning, deep learning, and transfer learning, are used for TC. Further, the study presents recommendations for improving AI-based architectures for vascular studies. We conclude that the process of vascular damage due to COVID-19 has similarities across vessel types, even though it results in multi-organ dysfunction. Although the mortality rate is ~2% of those infected, the long-term effect of COVID-19 needs monitoring to avoid deaths. AI seems to be penetrating the health care industry at warp speed, and we expect to see an emerging role in patient care, reduce the mortality and morbidity rate.

Applications of Artificial Intelligence Based on Medical Imaging in Glioma: Current State and Future Challenges

Glioma is one of the most fatal primary brain tumors, and it is well-known for its difficulty in diagnosis and management. Medical imaging techniques such as magnetic resonance imaging (MRI), positron emission tomography (PET), and spectral imaging can efficiently aid physicians in diagnosing, treating, and evaluating patients with gliomas. With the increasing clinical records and digital images, the application of artificial intelligence (AI) based on medical imaging has reduced the burden on physicians treating gliomas even further. This review will classify AI technologies and procedures used in medical imaging analysis. Additionally, we will discuss the applications of AI in glioma, including tumor segmentation and classification, prediction of genetic markers, and prediction of treatment response and prognosis, using MRI, PET, and spectral imaging. Despite the benefits of AI in clinical applications, several issues such as data management, incomprehension, safety, clinical efficacy evaluation, and ethical or legal considerations, remain to be solved. In the future, doctors and researchers should collaborate to solve these issues, with a particular emphasis on interdisciplinary teamwork.

Deep Learning-Based Segmentation of Post-Mortem Human’s Olfactory Bulb Structures in X-ray Phase-Contrast Tomography

The human olfactory bulb (OB) has a laminar structure. The segregation of cell populations in the OB image poses a significant challenge because of indistinct boundaries of the layers. Standard 3D visualization tools usually have a low resolution and cannot provide the high accuracy required for morphometric analysis. X-ray phase contrast tomography (XPCT) offers sufficient resolution and contrast to identify single cells in large volumes of the brain. The numerous microanatomical structures detectable in XPCT image of the OB, however, greatly complicate the manual delineation of OB neuronal cell layers. To address the challenging problem of fully automated segmentation of XPCT images of human OB morphological layers, we propose a new pipeline for tomographic data processing. Convolutional neural networks (CNN) were used to segment XPCT image of native unstained human OB. Virtual segmentation of the whole OB and an accurate delineation of each layer in a healthy non-demented OB is mandatory as the first step for assessing OB morphological changes in smell impairment research. In this framework, we proposed an effective tool that could help to shed light on OB layer-specific degeneration in patients with olfactory disorder.

Deep Learning-Based Automated Diagnosis for Coronary Artery Disease Using SPECT-MPI Images

(1) Background: Single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) is a long-established estimation methodology for medical diagnosis using image classification illustrating conditions in coronary artery disease. For these procedures, convolutional neural networks have proven to be very beneficial in achieving near-optimal accuracy for the automatic classification of SPECT images. (2) Methods: This research addresses the supervised learning-based ideal observer image classification utilizing an RGB-CNN model in heart images to diagnose CAD. For comparison purposes, we employ VGG-16 and DenseNet-121 pre-trained networks that are indulged in an image dataset representing stress and rest mode heart states acquired by SPECT. In experimentally evaluating the method, we explore a wide repertoire of deep learning network setups in conjunction with various robust evaluation and exploitation metrics. Additionally, to overcome the image dataset cardinality restrictions, we take advantage of the data augmentation technique expanding the set into an adequate number. Further evaluation of the model was performed via 10-fold cross-validation to ensure our model's reliability. (3) Results: The proposed RGB-CNN model achieved an accuracy of 91.86%, while VGG-16 and DenseNet-121 reached 88.54% and 86.11%, respectively. (4) Conclusions: The abovementioned experiments verify that the newly developed deep learning models may be of great assistance in nuclear medicine and clinical decision-making.

Eight pruning deep learning models for low storage and high-speed COVID-19 computed tomography lung segmentation and heatmap-based lesion localization: A multicenter study using COVLIAS 2.0

BACKGROUND

COVLIAS 1.0: an automated lung segmentation was designed for COVID-19 diagnosis. It has issues related to storage space and speed. This study shows that COVLIAS 2.0 uses pruned AI (PAI) networks for improving both storage and speed, wiliest high performance on lung segmentation and lesion localization.

METHOD

ology: The proposed study uses multicenter ∼9,000 CT slices from two different nations, namely, CroMed from Croatia (80 patients, experimental data), and NovMed from Italy (72 patients, validation data). We hypothesize that by using pruning and evolutionary optimization algorithms, the size of the AI models can be reduced significantly, ensuring optimal performance. Eight different pruning techniques (i) differential evolution (DE), (ii) genetic algorithm (GA), (iii) particle swarm optimization algorithm (PSO), and (iv) whale optimization algorithm (WO) in two deep learning frameworks (i) Fully connected network (FCN) and (ii) SegNet were designed. COVLIAS 2.0 was validated using "Unseen NovMed" and benchmarked against MedSeg. Statistical tests for stability and reliability were also conducted.

RESULTS

Pruning algorithms (i) FCN-DE, (ii) FCN-GA, (iii) FCN-PSO, and (iv) FCN-WO showed improvement in storage by 92.4%, 95.3%, 98.7%, and 99.8% respectively when compared against solo FCN, and (v) SegNet-DE, (vi) SegNet-GA, (vii) SegNet-PSO, and (viii) SegNet-WO showed improvement by 97.1%, 97.9%, 98.8%, and 99.2% respectively when compared against solo SegNet. AUC > 0.94 (p < 0.0001) on CroMed and > 0.86 (p < 0.0001) on NovMed data set for all eight EA model. PAI <0.25 s per image. DenseNet-121-based Grad-CAM heatmaps showed validation on glass ground opacity lesions.

CONCLUSIONS

Eight PAI networks that were successfully validated are five times faster, storage efficient, and could be used in clinical settings.

From YouTube to the brain: Transfer learning can improve brain-imaging predictions with deep learning

Deep learning has recently achieved best-in-class performance in several fields, including biomedical domains such as X-ray images. Yet, data scarcity poses a strict limit on training successful deep learning systems in many, if not most, biomedical applications, including those involving brain images. In this study, we translate state-of-the-art transfer learning techniques for single-subject prediction of simpler (sex and age) and more complex phenotypes (number of people in household, household income, fluid intelligence and smoking behavior). We fine-tuned 2D and 3D ResNet-18 convolutional neural networks for target phenotype predictions from brain images of ∼40,000 UK Biobank participants, after pretraining on YouTube videos from the Kinetics dataset and natural images from the ImageNet dataset. Transfer learning was effective on several phenotypes, especially sex and age classification. Additionally, transfer learning in particular outperformed deep learning models trained from scratch especially on smaller sample sizes. The out-of-sample performance using transfer learning from previously learned knowledge based on real-world images and videos could unlock the potential in many areas of imaging neuroscience where deep learning solutions are currently infeasible.

Role of Artificial Intelligence in Radiogenomics for Cancers in the Era of Precision Medicine

Simple Summary

Recently, radiogenomics has played a significant role and offered a new understanding of cancer’s biology and behavior in response to standard therapy. It also provides a more precise prognosis, investigation, and analysis of the patient’s cancer. Over the years, Artificial Intelligence (AI) has provided a significant strength in radiogenomics. In this paper, we offer computational and oncological prospects of the role of AI in radiogenomics, as well as its offers, achievements, opportunities, and limitations in the current clinical practices.

Abstract

Radiogenomics, a combination of “Radiomics” and “Genomics,” using Artificial Intelligence (AI) has recently emerged as the state-of-the-art science in precision medicine, especially in oncology care. Radiogenomics syndicates large-scale quantifiable data extracted from radiological medical images enveloped with personalized genomic phenotypes. It fabricates a prediction model through various AI methods to stratify the risk of patients, monitor therapeutic approaches, and assess clinical outcomes. It has recently shown tremendous achievements in prognosis, treatment planning, survival prediction, heterogeneity analysis, reoccurrence, and progression-free survival for human cancer study. Although AI has shown immense performance in oncology care in various clinical aspects, it has several challenges and limitations. The proposed review provides an overview of radiogenomics with the viewpoints on the role of AI in terms of its promises for computational as well as oncological aspects and offers achievements and opportunities in the era of precision medicine. The review also presents various recommendations to diminish these obstacles.

Medical deep learning-A systematic meta-review

Deep learning has remarkably impacted several different scientific disciplines over the last few years. For example, in image processing and analysis, deep learning algorithms were able to outperform other cutting-edge methods. Additionally, deep learning has delivered state-of-the-art results in tasks like autonomous driving, outclassing previous attempts. There are even instances where deep learning outperformed humans, for example with object recognition and gaming. Deep learning is also showing vast potential in the medical domain. With the collection of large quantities of patient records and data, and a trend towards personalized treatments, there is a great need for automated and reliable processing and analysis of health information. Patient data is not only collected in clinical centers, like hospitals and private practices, but also by mobile healthcare apps or online websites. The abundance of collected patient data and the recent growth in the deep learning field has resulted in a large increase in research efforts. In Q2/2020, the search engine PubMed returned already over 11,000 results for the search term 'deep learning', and around 90% of these publications are from the last three years. However, even though PubMed represents the largest search engine in the medical field, it does not cover all medical-related publications. Hence, a complete overview of the field of 'medical deep learning' is almost impossible to obtain and acquiring a full overview of medical sub-fields is becoming increasingly more difficult. Nevertheless, several review and survey articles about medical deep learning have been published within the last few years. They focus, in general, on specific medical scenarios, like the analysis of medical images containing specific pathologies. With these surveys as a foundation, the aim of this article is to provide the first high-level, systematic meta-review of medical deep learning surveys.

Cardiovascular/Stroke Risk Assessment in Patients with Erectile Dysfunction-A Role of Carotid Wall Arterial Imaging and Plaque Tissue Characterization Using Artificial Intelligence Paradigm: A Narrative Review

PURPOSE

The role of erectile dysfunction (ED) has recently shown an association with the risk of stroke and coronary heart disease (CHD) via the atherosclerotic pathway. Cardiovascular disease (CVD)/stroke risk has been widely understood with the help of carotid artery disease (CTAD), a surrogate biomarker for CHD. The proposed study emphasizes artificial intelligence-based frameworks such as machine learning (ML) and deep learning (DL) that can accurately predict the severity of CVD/stroke risk using carotid wall arterial imaging in ED patients.

METHODS

Using the PRISMA model, 231 of the best studies were selected. The proposed study mainly consists of two components: (i) the pathophysiology of ED and its link with coronary artery disease (COAD) and CHD in the ED framework and (ii) the ultrasonic-image morphological changes in the carotid arterial walls by quantifying the wall parameters and the characterization of the wall tissue by adapting the ML/DL-based methods, both for the prediction of the severity of CVD risk. The proposed study analyzes the hypothesis that ML/DL can lead to an accurate and early diagnosis of the CVD/stroke risk in ED patients. Our finding suggests that the routine ED patient practice can be amended for ML/DL-based CVD/stroke risk assessment using carotid wall arterial imaging leading to fast, reliable, and accurate CVD/stroke risk stratification.

SUMMARY

We conclude that ML and DL methods are very powerful tools for the characterization of CVD/stroke in patients with varying ED conditions. We anticipate a rapid growth of these tools for early and better CVD/stroke risk management in ED patients.

Cardiovascular Risk Stratification in Diabetic Retinopathy via Atherosclerotic Pathway in COVID-19/non-COVID-19 Frameworks using Artificial Intelligence Paradigm: A Narrative Review

Diabetes is one of the main causes of the rising cases of blindness in adults. This microvascular complication of diabetes is termed diabetic retinopathy (DR) and is associated with an expanding risk of cardiovascular events in diabetes patients. DR, in its various forms, is seen to be a powerful indicator of atherosclerosis. Further, the macrovascular complication of diabetes leads to coronary artery disease (CAD). Thus, the timely identification of cardiovascular disease (CVD) complications in DR patients is of utmost importance. Since CAD risk assessment is expensive for low-income countries, it is important to look for surrogate biomarkers for risk stratification of CVD in DR patients. Due to the common genetic makeup between the coronary and carotid arteries, low-cost, high-resolution imaging such as carotid B-mode ultrasound (US) can be used for arterial tissue characterization and risk stratification in DR patients. The advent of artificial intelligence (AI) techniques has facilitated the handling of large cohorts in a big data framework to identify atherosclerotic plaque features in arterial ultrasound. This enables timely CVD risk assessment and risk stratification of patients with DR. Thus, this review focuses on understanding the pathophysiology of DR, retinal and CAD imaging, the role of surrogate markers for CVD, and finally, the CVD risk stratification of DR patients. The review shows a step-by-step cyclic activity of how diabetes and atherosclerotic disease cause DR, leading to the worsening of CVD. We propose a solution to how AI can help in the identification of CVD risk. Lastly, we analyze the role of DR/CVD in the COVID-19 framework.

Application of SPECT and PET / CT with computer-aided diagnosis in bone metastasis of prostate cancer: a review

Bone metastasis has a significant influence on the prognosis of prostate cancer(PCa) patients. In this review, we discussed the current application of PCa bone metastasis diagnosis with single-photon emission computed tomography (SPECT) and positron emission tomography/computed tomography (PET/CT) computer-aided diagnosis(CAD) systems. A literature search identified articles concentrated on PCa bone metastasis and PET/CT or SPECT CAD systems using the PubMed database. We summarized the previous studies focused on CAD systems and manual quantitative markers calculation, and the coincidence rate was acceptable. We also analyzed the quantification methods, advantages, and disadvantages of CAD systems. CAD systems can detect abnormal lesions of PCa patients’ 99mTc-MDP-SPECT, 18F-FDG-PET/CT, 18F-NaF-PET/CT, and 68 Ga-PSMA PET/CT images automated or semi-automated. CAD systems can also calculate the quantitative markers, which can quantify PCa patients’ whole-body bone metastasis tumor burden accurately and quickly and give a standardized and objective result. SPECT and PET/CT CAD systems are potential tools to monitor and quantify bone metastasis lesions of PCa patients simply and accurately, the future clinical application of CAD systems in diagnosing PCa bone metastasis lesions is necessary and feasible.

Head Pitch Angular Velocity Discriminates (Sub-)Acute Neck Pain Patients and Controls Assessed with the DidRen Laser Test

Understanding neck pain is an important societal issue. Kinematic data from sensors may help to gain insight into the pathophysiological mechanisms associated with neck pain through a quantitative sensorimotor assessment of one patient. The objective of this study was to evaluate the potential usefulness of artificial intelligence with several machine learning (ML) algorithms in assessing neck sensorimotor performance. Angular velocity and acceleration measured by an inertial sensor placed on the forehead during the DidRen laser test in thirty-eight acute and subacute non-specific neck pain (ANSP) patients were compared to forty-two healthy control participants (HCP). Seven supervised ML algorithms were chosen for the predictions. The most informative kinematic features were computed using Sequential Feature Selection methods. The best performing algorithm is the Linear Support Vector Machine with an accuracy of 82% and Area Under Curve of 84%. The best discriminative kinematic feature between ANSP patients and HCP is the first quartile of head pitch angular velocity. This study has shown that supervised ML algorithms could be used to classify ANSP patients and identify discriminatory kinematic features potentially useful for clinicians in the assessment and monitoring of the neck sensorimotor performance in ANSP patients.

An artificial intelligence framework and its bias for brain tumor segmentation: A narrative review

BACKGROUND

Artificial intelligence (AI) has become a prominent technique for medical diagnosis and represents an essential role in detecting brain tumors. Although AI-based models are widely used in brain lesion segmentation (BLS), understanding their effectiveness is challenging due to their complexity and diversity. Several reviews on brain tumor segmentation are available, but none of them describe a link between the threats due to risk-of-bias (RoB) in AI and its architectures. In our review, we focused on linking RoB and different AI-based architectural Cluster in popular DL framework. Further, due to variance in these designs and input data types in medical imaging, it is necessary to present a narrative review considering all facets of BLS.

APPROACH

The proposed study uses a PRISMA strategy based on 75 relevant studies found by searching PubMed, Scopus, and Google Scholar. Based on the architectural evolution, DL studies were subsequently categorized into four classes: convolutional neural network (CNN)-based, encoder-decoder (ED)-based, transfer learning (TL)-based, and hybrid DL (HDL)-based architectures. These studies were then analyzed considering 32 AI attributes, with clusters including AI architecture, imaging modalities, hyper-parameters, performance evaluation metrics, and clinical evaluation. Then, after these studies were scored for all attributes, a composite score was computed, normalized, and ranked. Thereafter, a bias cutoff (AP(ai)Bias 1.0, AtheroPoint, Roseville, CA, USA) was established to detect low-, moderate- and high-bias studies.

CONCLUSION

The four classes of architectures, from best-to worst-performing, are TL > ED > CNN > HDL. ED-based models had the lowest AI bias for BLS. This study presents a set of three primary and six secondary recommendations for lowering the RoB.

Cardiovascular/Stroke Risk Stratification in Parkinson's Disease Patients Using Atherosclerosis Pathway and Artificial Intelligence Paradigm: A Systematic Review

Parkinson’s disease (PD) is a severe, incurable, and costly condition leading to heart failure. The link between PD and cardiovascular disease (CVD) is not available, leading to controversies and poor prognosis. Artificial Intelligence (AI) has already shown promise for CVD/stroke risk stratification. However, due to a lack of sample size, comorbidity, insufficient validation, clinical examination, and a lack of big data configuration, there have been no well-explained bias-free AI investigations to establish the CVD/Stroke risk stratification in the PD framework. The study has two objectives: (i) to establish a solid link between PD and CVD/stroke; and (ii) to use the AI paradigm to examine a well-defined CVD/stroke risk stratification in the PD framework. The PRISMA search strategy selected 223 studies for CVD/stroke risk, of which 54 and 44 studies were related to the link between PD-CVD, and PD-stroke, respectively, 59 studies for joint PD-CVD-Stroke framework, and 66 studies were only for the early PD diagnosis without CVD/stroke link. Sequential biological links were used for establishing the hypothesis. For AI design, PD risk factors as covariates along with CVD/stroke as the gold standard were used for predicting the CVD/stroke risk. The most fundamental cause of CVD/stroke damage due to PD is cardiac autonomic dysfunction due to neurodegeneration that leads to heart failure and its edema, and this validated our hypothesis. Finally, we present the novel AI solutions for CVD/stroke risk prediction in the PD framework. The study also recommends strategies for removing the bias in AI for CVD/stroke risk prediction using the PD framework.

A Powerful Paradigm for Cardiovascular Risk Stratification Using Multiclass, Multi-Label, and Ensemble-Based Machine Learning Paradigms: A Narrative Review

Background and Motivation: Cardiovascular disease (CVD) causes the highest mortality globally. With escalating healthcare costs, early non-invasive CVD risk assessment is vital. Conventional methods have shown poor performance compared to more recent and fast-evolving Artificial Intelligence (AI) methods. The proposed study reviews the three most recent paradigms for CVD risk assessment, namely multiclass, multi-label, and ensemble-based methods in (i) office-based and (ii) stress-test laboratories. Methods: A total of 265 CVD-based studies were selected using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) model. Due to its popularity and recent development, the study analyzed the above three paradigms using machine learning (ML) frameworks. We review comprehensively these three methods using attributes, such as architecture, applications, pro-and-cons, scientific validation, clinical evaluation, and AI risk-of-bias (RoB) in the CVD framework. These ML techniques were then extended under mobile and cloud-based infrastructure. Findings: Most popular biomarkers used were office-based, laboratory-based, image-based phenotypes, and medication usage. Surrogate carotid scanning for coronary artery risk prediction had shown promising results. Ground truth (GT) selection for AI-based training along with scientific and clinical validation is very important for CVD stratification to avoid RoB. It was observed that the most popular classification paradigm is multiclass followed by the ensemble, and multi-label. The use of deep learning techniques in CVD risk stratification is in a very early stage of development. Mobile and cloud-based AI technologies are more likely to be the future. Conclusions: AI-based methods for CVD risk assessment are most promising and successful. Choice of GT is most vital in AI-based models to prevent the RoB. The amalgamation of image-based strategies with conventional risk factors provides the highest stability when using the three CVD paradigms in non-cloud and cloud-based frameworks.

Four Types of Multiclass Frameworks for Pneumonia Classification and Its Validation in X-ray Scans Using Seven Types of Deep Learning Artificial Intelligence Models

Background and Motivation: The novel coronavirus causing COVID-19 is exceptionally contagious, highly mutative, decimating human health and life, as well as the global economy, by consistent evolution of new pernicious variants and outbreaks. The reverse transcriptase polymerase chain reaction currently used for diagnosis has major limitations. Furthermore, the multiclass lung classification X-ray systems having viral, bacterial, and tubercular classes—including COVID-19—are not reliable. Thus, there is a need for a robust, fast, cost-effective, and easily available diagnostic method. Method: Artificial intelligence (AI) has been shown to revolutionize all walks of life, particularly medical imaging. This study proposes a deep learning AI-based automatic multiclass detection and classification of pneumonia from chest X-ray images that are readily available and highly cost-effective. The study has designed and applied seven highly efficient pre-trained convolutional neural networks—namely, VGG16, VGG19, DenseNet201, Xception, InceptionV3, NasnetMobile, and ResNet152—for classification of up to five classes of pneumonia. Results: The database consisted of 18,603 scans with two, three, and five classes. The best results were using DenseNet201, VGG16, and VGG16, respectively having accuracies of 99.84%, 96.7%, 92.67%; sensitivity of 99.84%, 96.63%, 92.70%; specificity of 99.84, 96.63%, 92.41%; and AUC of 1.0, 0.97, 0.92 (p < 0.0001 for all), respectively. Our system outperformed existing methods by 1.2% for the five-class model. The online system takes <1 s while demonstrating reliability and stability. Conclusions: Deep learning AI is a powerful paradigm for multiclass pneumonia classification.

Bias Investigation in Artificial Intelligence Systems for Early Detection of Parkinson's Disease: A Narrative Review

Background and Motivation: Diagnosis of Parkinson's disease (PD) is often based on medical attention and clinical signs. It is subjective and does not have a good prognosis. Artificial Intelligence (AI) has played a promising role in the diagnosis of PD. However, it introduces bias due to lack of sample size, poor validation, clinical evaluation, and lack of big data configuration. The purpose of this study is to compute the risk of bias (RoB) automatically.

METHOD

The PRISMA search strategy was adopted to select the best 39 AI studies out of 85 PD studies closely associated with early diagnosis PD. The studies were used to compute 30 AI attributes (based on 6 AI clusters), using AP(ai)Bias 1.0 (AtheroPointTM, Roseville, CA, USA), and the mean aggregate score was computed. The studies were ranked and two cutoffs (Moderate-Low (ML) and High-Moderate (MH)) were determined to segregate the studies into three bins: low-, moderate-, and high-bias.

RESULT

The ML and HM cutoffs were 3.50 and 2.33, respectively, which constituted 7, 13, and 6 for low-, moderate-, and high-bias studies. The best and worst architectures were "deep learning with sketches as outcomes" and "machine learning with Electroencephalography," respectively. We recommend (i) the usage of power analysis in big data framework, (ii) that it must undergo scientific validation using unseen AI models, and (iii) that it should be taken towards clinical evaluation for reliability and stability tests.

CONCLUSION

The AI is a vital component for the diagnosis of early PD and the recommendations must be followed to lower the RoB.

A machine learning framework for risk prediction of multi-label cardiovascular events based on focused carotid plaque B-Mode ultrasound: A Canadian study

MOTIVATION

Machine learning (ML) algorithms can provide better cardiovascular event (CVE) prediction. However, ML algorithms are mostly explored for predicting a single CVE at a time. The objective of this study is to design and develop an ML-based system to predict multi-label CVEs, such as (i) coronary artery disease, (ii) acute coronary syndrome, and (iii) a composite CVE-a class of AtheroEdge 3.0 (ML) system.

METHODS

Focused carotid B-mode ultrasound and coronary angiography are performed on a group of 459 participants consisting of three cardiovascular labels. Initially, 23 risk predictors comprising (i) patients' demographics, (ii) clinical blood-biomarkers, and (iii) carotid ultrasound image-based phenotypes are collected. Six types of classification techniques comprising (a) four problem transformation methods (PTM) and (b) two algorithm adaptation methods (AAM) are used for multi-label CVE prediction. The performance of the proposed system is evaluated for accuracy, sensitivity, specificity, F1-score, and area-under-the-curve (AUC) using 10-fold cross-validation. The proposed system is also verified using another database of 522 participants.

RESULTS

For the primary database, PTM demonstrated a better multi-label CVE prediction than AAM (mean accuracy: 80.89% vs. 62.83%, mean AUC: 0.89 vs. 0.63), validating our hypothesis. The PTM-based binary relevance (BR) technique provided optimal performance in multi-label CVE prediction. The overall multi-label classification accuracy, sensitivity, specificity, F1-score, and AUC using BR are 81.2 ± 3.01%, 76.5 ± 8.8%, 83.8 ± 3.8%, 75.37 ± 5.8%, and 0.89 ± 0.02 (p < 0.0001), respectively. When used on the second Canadian database with seven cardiovascular events (acute coronary syndrome, myocardial infarction, angina, stroke, transient ischemic attack, heart failure, and death), the proposed system showed an accuracy of 96.36 ± 0.87% (AUC: 0.61 ± 0.06, p < 0.0001).

CONCLUSION

ML-based multi-label classification algorithms, such as binary relevance, yielded the best predictions for three cardiovascular endpoints.

Human activity recognition in artificial intelligence framework: a narrative review

Human activity recognition (HAR) has multifaceted applications due to its worldly usage of acquisition devices such as smartphones, video cameras, and its ability to capture human activity data. While electronic devices and their applications are steadily growing, the advances in Artificial intelligence (AI) have revolutionized the ability to extract deep hidden information for accurate detection and its interpretation. This yields a better understanding of rapidly growing acquisition devices, AI, and applications, the three pillars of HAR under one roof. There are many review articles published on the general characteristics of HAR, a few have compared all the HAR devices at the same time, and few have explored the impact of evolving AI architecture. In our proposed review, a detailed narration on the three pillars of HAR is presented covering the period from 2011 to 2021. Further, the review presents the recommendations for an improved HAR design, its reliability, and stability. Five major findings were: (1) HAR constitutes three major pillars such as devices, AI and applications; (2) HAR has dominated the healthcare industry; (3) Hybrid AI models are in their infancy stage and needs considerable work for providing the stable and reliable design. Further, these trained models need solid prediction, high accuracy, generalization, and finally, meeting the objectives of the applications without bias; (4) little work was observed in abnormality detection during actions; and (5) almost no work has been done in forecasting actions. We conclude that: (a) HAR industry will evolve in terms of the three pillars of electronic devices, applications and the type of AI. (b) AI will provide a powerful impetus to the HAR industry in future.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10462-021-10116-x.

AI MSK clinical applications: spine imaging

Recent investigations have focused on the clinical application of artificial intelligence (AI) for tasks specifically addressing the musculoskeletal imaging routine. Several AI applications have been dedicated to optimizing the radiology value chain in spine imaging, independent from modality or specific application. This review aims to summarize the status quo and future perspective regarding utilization of AI for spine imaging. First, the basics of AI concepts are clarified. Second, the different tasks and use cases for AI applications in spine imaging are discussed and illustrated by examples. Finally, the authors of this review present their personal perception of AI in daily imaging and discuss future chances and challenges that come along with AI-based solutions.

A hybrid deep learning paradigm for carotid plaque tissue characterization and its validation in multicenter cohorts using a supercomputer framework

BACKGROUND

Early and automated detection of carotid plaques prevents strokes, which are the second leading cause of death worldwide according to the World Health Organization. Artificial intelligence (AI) offers automated solutions for plaque tissue characterization. Recently, solo deep learning (SDL) models have been used, but they do not take advantage of the tandem connectivity offered by AI's hybrid nature. Therefore, this study explores the use of hybrid deep learning (HDL) models in a multicenter framework, making this study the first of its kind.

METHODS

We hypothesize that HDL techniques perform better than SDL and transfer learning (TL) techniques. We propose two kinds of HDL frameworks: (i) the fusion of two SDLs (Inception with ResNet) or (ii) 10 other kinds of tandem models that fuse SDL with ML. The system Atheromatic™ 2.0_HDL (AtheroPoint, CA, USA) was designed on an augmentation framework and three kinds of loss functions (cross-entropy, hinge, and mean-square-error) during training to determine the best optimization paradigm. These 11 combined HDL models were then benchmarked against one SDL model and five types of TL models; thus, this study considers a total of 17 AI models.

RESULTS

Among the 17 AI models, the best performing HDL system was that comprising CNN and decision tree (DT), as its accuracy and area-under-the-curve were 99.78 ± 1.05% and 0.99 (p<0.0001), respectively. These values are 6.4% and 3.2% better than those recorded for the SDL and TL models, respectively. We validated the performance of the HDL models with diagnostics odds ratio (DOR) and Cohen and Kappa statistics; here, HDL outperformed DL and TL by 23% and 7%, respectively. The online system ran in <2 s.

CONCLUSION

HDL is a fast, reliable, and effective tool for characterizing the carotid plaque for early stroke risk stratification.

Unseen Artificial Intelligence-Deep Learning Paradigm for Segmentation of Low Atherosclerotic Plaque in Carotid Ultrasound: A Multicenter Cardiovascular Study

Background: The early detection of carotid wall plaque is recommended in the prevention of cardiovascular disease (CVD) in moderate-risk patients. Previous techniques for B-mode carotid atherosclerotic wall plaque segmentation used artificial intelligence (AI) methods on monoethnic databases, where training and testing are from the “same” ethnic group (“Seen AI”). Therefore, the versatility of the system is questionable. This is the first study of its kind that uses the “Unseen AI” paradigm where training and testing are from “different” ethnic groups. We hypothesized that deep learning (DL) models should perform in 10% proximity between “Unseen AI” and “Seen AI”. Methodology: Two cohorts from multi-ethnic groups (330 Japanese and 300 Hong Kong (HK)) were used for the validation of our hypothesis. We used a four-layered UNet architecture for the segmentation of the atherosclerotic wall with low plaque. “Unseen AI” (training: Japanese, testing: HK or vice versa) and “Seen AI” experiments (single ethnicity or mixed ethnicity) were performed. Evaluation was conducted by measuring the wall plaque area. Statistical tests were conducted for its stability and reliability. Results: When using the UNet DL architecture, the “Unseen AI” pair one (Training: 330 Japanese and Testing: 300 HK), the mean accuracy, dice-similarity, and correlation-coefficient were 98.55, 78.38, and 0.80 (p < 0.0001), respectively, while for “Unseen AI” pair two (Training: 300 HK and Testing: 330 Japanese), these were 98.67, 82.49, and 0.87 (p < 0.0001), respectively. Using “Seen AI”, the same parameters were 99.01, 86.89 and 0.92 (p < 0.0001), respectively. Conclusion: We demonstrated that “Unseen AI” was in close proximity (<10%) to “Seen AI”, validating our DL model for low atherosclerotic wall plaque segmentation. The online system runs < 1 s.

Trends in the application of deep learning networks in medical image analysis: Evolution between 2012 and 2020

PURPOSE

To evaluate the general rules and future trajectories of deep learning (DL) networks in medical image analysis through bibliometric and hot spot analysis of original articles published between 2012 and 2020.

METHODS

Original articles related to DL and medical imaging were retrieved from the PubMed database. For the analysis, data regarding radiological subspecialties; imaging techniques; DL networks; sample size; study purposes, setting, origins and design; statistical analysis; funding sources; authors; and first authors' affiliation was manually extracted from each article. The Bibliographic Item Co-Occurrence Matrix Builder and VOSviewer were used to identify the research topics of the included articles and illustrate the future trajectories of studies.

RESULTS

The study included 2685 original articles. The number of publications on DL and medical imaging has increased substantially since 2017, accounting for 97.2% of all included articles. We evaluated the rules of the application of 47 DL networks to eight radiological tasks on 11 human organ sites. Neuroradiology, thorax, and abdomen were frequent research subjects, while thyroid was under-represented. Segmentation and classification tasks were the primary purposes. U-Net, ResNet, and VGG were the most frequently used Convolutional neural network-derived networks. GAN-derived networks were widely developed and applied in 2020, and transfer learning was highlighted in the COVID-19 studies. Brain, prostate, and diabetic retinopathy-related studies were mature research topics in the field. Breast- and lung-related studies were in a stage of rapid development.

CONCLUSIONS

This study evaluates the general rules and future trajectories of DL network application in medical image analyses and provides guidance for future studies.

Automated deep learning-based paradigm for high-risk plaque detection in B-mode common carotid ultrasound scans: an asymptomatic Japanese cohort study

BACKGROUND

The death due to stroke is caused by embolism of the arteries which is due to the rupture of the atherosclerotic lesions in carotid arteries. The lesion formation is over time, and thus, early screening is recommended for asymptomatic and moderate-risk patients. The previous techniques adopted conventional methods or semi-automated and, more recently, machine learning solutions. A handful of studies have emerged based on solo deep learning (SDL) models such as UNet architecture.

METHODS

The proposed research is the first to adopt hybrid deep learning (HDL) artificial intelligence models such as SegNet-UNet. This model is benchmarked against UNet and advanced conventional models using scale-space such as AtheroEdge 2.0 (AtheroPoint, CA, USA). All our resultant statistics of the three systems were in the order of UNet, SegNet-UNet, and AtheroEdge 2.0.

RESULTS

Using the database of 379 ultrasound scans from a Japanese cohort of 190 patients having moderate risk and implementing the cross-validation deep learning framework, our system performance using area-under-the-curve (AUC) for UNet, SegNet-UNet, and AtheroEdge 2.0 were 0.93, 0.94, and 0.95 (p<0.001), respectively. The coefficient of correlation between the three systems and ground truth (GT) were: 0.82, 0.89, and 0.85 (p<0.001 for all three), respectively. The mean absolute area error for the three systems against manual GT was 4.07±4.70 mm2, 3.11±3.92 mm2, 3.72±4.76 mm2, respectively, proving the superior performance SegNet-UNet against UNet and AtheroEdge 2.0, respectively. Statistical tests were also conducted for their reliability and stability.

CONCLUSIONS

The proposed study demonstrates a fast, accurate, and reliable solution for early detection and quantification of plaque lesions in common carotid artery ultrasound scans. The system runs on a test US image in < 1 second, proving overall performance to be clinically reliable.

Ten Fast Transfer Learning Models for Carotid Ultrasound Plaque Tissue Characterization in Augmentation Framework Embedded with Heatmaps for Stroke Risk Stratification

Background and Purpose: Only 1-2% of the internal carotid artery asymptomatic plaques are unstable as a result of >80% stenosis. Thus, unnecessary efforts can be saved if these plaques can be characterized and classified into symptomatic and asymptomatic using non-invasive B-mode ultrasound. Earlier plaque tissue characterization (PTC) methods were machine learning (ML)-based, which used hand-crafted features that yielded lower accuracy and unreliability. The proposed study shows the role of transfer learning (TL)-based deep learning models for PTC. Methods: As pertained weights were used in the supercomputer framework, we hypothesize that transfer learning (TL) provides improved performance compared with deep learning. We applied 11 kinds of artificial intelligence (AI) models, 10 of them were augmented and optimized using TL approaches-a class of Atheromatic™ 2.0 TL (AtheroPoint™, Roseville, CA, USA) that consisted of (i-ii) Visual Geometric Group-16, 19 (VGG16, 19); (iii) Inception V3 (IV3); (iv-v) DenseNet121, 169; (vi) XceptionNet; (vii) ResNet50; (viii) MobileNet; (ix) AlexNet; (x) SqueezeNet; and one DL-based (xi) SuriNet-derived from UNet. We benchmark 11 AI models against our earlier deep convolutional neural network (DCNN) model. Results: The best performing TL was MobileNet, with accuracy and area-under-the-curve (AUC) pairs of 96.10 ± 3% and 0.961 (p < 0.0001), respectively. In DL, DCNN was comparable to SuriNet, with an accuracy of 95.66% and 92.7 ± 5.66%, and an AUC of 0.956 (p < 0.0001) and 0.927 (p < 0.0001), respectively. We validated the performance of the AI architectures with established biomarkers such as greyscale median (GSM), fractal dimension (FD), higher-order spectra (HOS), and visual heatmaps. We benchmarked against previously developed Atheromatic™ 1.0 ML and showed an improvement of 12.9%. Conclusions: TL is a powerful AI tool for PTC into symptomatic and asymptomatic plaques.

Artificial intelligence-based hybrid deep learning models for image classification: The first narrative review

BACKGROUND

Artificial intelligence (AI) has served humanity in many applications since its inception. Currently, it dominates the imaging field-in particular, image classification. The task of image classification became much easier with machine learning (ML) and subsequently got automated and more accurate by using deep learning (DL). By default, DL consists of a single architecture and is termed solo deep learning (SDL). When two or more DL architectures are fused, the result is termed a hybrid deep learning (HDL) model. The use of HDL models is becoming popular in several applications, but no review of these uses has been designed thus far. Therefore, this study provides the first narrative HDL review by considering all facets of image classification using AI.

APPROACH

Our review employs a PRISMA search strategy using Google Scholar, PubMed, IEEE, and Elsevier Science Direct, through which 127 relevant HDL studies were considered. Based on the computer vision evolution, HDLs were subsequently classified into three categories (spatial, temporal, and spatial-temporal). Each study was then analyzed based on several attributes, including continent, publisher, hybridization of two DL or ML, architecture layout, application type, data set type, dataset size, feature extraction methodology, connecting classifier, performance evaluation metrics, and risk-of-bias.

CONCLUSION

The HDL models have shown stable and superior performance by taking the best aspects of two or more solo DL or fusion of DL with ML models. Our findings indicate that HDL is being applied aggressively to several medical and non-medical applications. Furthermore, risk-of-bias is highly debatable for DL and HDL models.

COVLIAS 1.0: Lung Segmentation in COVID-19 Computed Tomography Scans Using Hybrid Deep Learning Artificial Intelligence Models

BACKGROUND

COVID-19 lung segmentation using Computed Tomography (CT) scans is important for the diagnosis of lung severity. The process of automated lung segmentation is challenging due to (a) CT radiation dosage and (b) ground-glass opacities caused by COVID-19. The lung segmentation methodologies proposed in 2020 were semi- or automated but not reliable, accurate, and user-friendly. The proposed study presents a COVID Lung Image Analysis System (COVLIAS 1.0, AtheroPoint™, Roseville, CA, USA) consisting of hybrid deep learning (HDL) models for lung segmentation.

METHODOLOGY

The COVLIAS 1.0 consists of three methods based on solo deep learning (SDL) or hybrid deep learning (HDL). SegNet is proposed in the SDL category while VGG-SegNet and ResNet-SegNet are designed under the HDL paradigm. The three proposed AI approaches were benchmarked against the National Institute of Health (NIH)-based conventional segmentation model using fuzzy-connectedness. A cross-validation protocol with a 40:60 ratio between training and testing was designed, with 10% validation data. The ground truth (GT) was manually traced by a radiologist trained personnel. For performance evaluation, nine different criteria were selected to perform the evaluation of SDL or HDL lung segmentation regions and lungs long axis against GT.

RESULTS

Using the database of 5000 chest CT images (from 72 patients), COVLIAS 1.0 yielded AUC of ~0.96, ~0.97, ~0.98, and ~0.96 (p-value < 0.001), respectively within 5% range of GT area, for SegNet, VGG-SegNet, ResNet-SegNet, and NIH. The mean Figure of Merit using four models (left and right lung) was above 94%. On benchmarking against the National Institute of Health (NIH) segmentation method, the proposed model demonstrated a 58% and 44% improvement in ResNet-SegNet, 52% and 36% improvement in VGG-SegNet for lung area, and lung long axis, respectively. The PE statistics performance was in the following order: ResNet-SegNet > VGG-SegNet > NIH > SegNet. The HDL runs in <1 s on test data per image.

CONCLUSIONS

The COVLIAS 1.0 system can be applied in real-time for radiology-based clinical settings.

Hybrid deep learning segmentation models for atherosclerotic plaque in internal carotid artery B-mode ultrasound

The automated and accurate carotid plaque segmentation in B-mode ultrasound (US) is an essential part of stroke risk stratification. Previous segmented methods used AtheroEdge™ 2.0 (AtheroPoint™, Roseville, CA) for the common carotid artery (CCA). This study focuses on automated plaque segmentation in the internal carotid artery (ICA) using solo deep learning (SDL) and hybrid deep learning (HDL) models. The methodology consists of a novel design of 10 types of SDL/HDL models (AtheroEdge™ 3.0 systems (AtheroPoint™, Roseville, CA) with a depth of four layers each. Five of the models use cross-entropy (CE)-loss, and the other five models use Dice similarity coefficient (DSC)-loss functions derived from UNet, UNet+, SegNet, SegNet-UNet, and SegNet-UNet+. The K10 protocol (Train:Test:90%:10%) was applied for all 10 models for training and predicting (segmenting) the plaque region, which was then quantified to compute the plaque area in mm². Further, the data augmentation effect was analyzed. The database consisted of 970 ICA B-mode US scans taken from 99 moderate to high-risk patients. Using the difference area threshold of 10 mm² between ground truth (GT) and artificial intelligence (AI), the area under the curve (AUC) values were 0.91, 0.911, 0.908, 0.905, and 0.898, all with a p-value of <0.001 (for CE-loss models) and 0.883, 0.889, 0.905, 0.889, and 0.907, all with a p-value of <0.001 (for DSC-loss models). The correlations between the AI-based plaque area and GT plaque area were 0.98, 0.96, 0.97, 0.98, and 0.97, all with a p-value of <0.001 (for CE-loss models) and 0.98, 0.98, 0.97, 0.98, and 0.98 (for DSC-loss models). Overall, the online system performs plaque segmentation in less than 1 s. We validate our hypothesis that HDL and SDL models demonstrate comparable performance. SegNet-UNet was the best-performing hybrid architecture.

Deep-Learning-Based Color Doppler Ultrasound Image Feature in the Diagnosis of Elderly Patients with Chronic Heart Failure Complicated with Sarcopenia

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