Deep learning approach for cardiovascular disease risk stratification and survival analysis on a Canadian cohort.

The quantification of carotid plaque has been routinely used to predict cardiovascular risk in cardiovascular disease (CVD) and coronary artery disease (CAD). To determine how well carotid plaque features predict the likelihood of CAD and cardiovascular (CV) events using deep learning (DL) and compare against the machine learning (ML) paradigm. The participants in this study consisted of 459 individuals who had undergone coronary angiography, contrast-enhanced ultrasonography, and focused carotid B-mode ultrasound. Each patient was tracked for thirty days. The measurements on these patients consisted of maximum plaque height (MPH), total plaque area (TPA), carotid intima-media thickness (cIMT), and intraplaque neovascularization (IPN). CAD risk and CV event stratification were performed by applying eight types of DL-based models. Univariate and multivariate analysis was also conducted to predict the most significant risk predictors. The DL's model effectiveness was evaluated by the area-under-the-curve measurement while the CV event prediction was evaluated using the Cox proportional hazard model (CPHM) and compared against the DL-based concordance index (c-index). IPN showed a substantial ability to predict CV events (p < 0.0001). The best DL system improved by 21% (0.929 vs. 0.762) over the best ML system. DL-based CV event prediction showed a ~ 17% increase in DL-based c-index compared to the CPHM (0.86 vs. 0.73). CAD and CV incidents were linked to IPN and carotid imaging characteristics. For survival analysis and CAD prediction, the DL-based system performs superior to ML-based models.

Recent advances and controversial issues in the optimal management of asymptomatic carotid stenosis.

OBJECTIVE: The optimal management of patients with asymptomatic carotid stenosis (AsxCS) is enduringly controversial. We updated our 2021 Expert Review and Position Statement, focusing on recent advances in the diagnosis and management of patients with AsxCS. METHODS: A systematic review of the literature was performed up to August 1, 2023, using PubMed/PubMed Central, EMBASE and Scopus. The following keywords were used in various combinations: "asymptomatic carotid stenosis," "carotid endarterectomy" (CEA), "carotid artery stenting" (CAS), and "transcarotid artery revascularization" (TCAR). Areas covered included (i) improvements in best medical treatment (BMT) for patients with AsxCS and declining stroke risk, (ii) technological advances in surgical/endovascular skills/techniques and outcomes, (iii) risk factors, clinical/imaging characteristics and risk prediction models for the identification of high-risk AsxCS patient subgroups, and (iv) the association between cognitive dysfunction and AsxCS. RESULTS: BMT is essential for all patients with AsxCS, regardless of whether they will eventually be offered CEA, CAS, or TCAR. Specific patient subgroups at high risk for stroke despite BMT should be considered for a carotid revascularization procedure. These patients include those with severe (≥80%) AsxCS, transcranial Doppler-detected microemboli, plaque echolucency on Duplex ultrasound examination, silent infarcts on brain computed tomography or magnetic resonance angiography scans, decreased cerebrovascular reserve, increased size of juxtaluminal hypoechoic area, AsxCS progression, carotid plaque ulceration, and intraplaque hemorrhage. Treatment of patients with AsxCS should be individualized, taking into consideration individual patient preferences and needs, clinical and imaging characteristics, and cultural, ethnic, and social factors. Solid evidence supporting or refuting an association between AsxCS and cognitive dysfunction is lacking. CONCLUSIONS: The optimal management of patients with AsxCS should include BMT for all individuals and a prophylactic carotid revascularization procedure (CEA, CAS, or TCAR) for some asymptomatic patient subgroups, additionally taking into consideration individual patient needs and preference, clinical and imaging characteristics, social and cultural factors, and the available stroke risk prediction models. Future studies should investigate the association between AsxCS with cognitive function and the role of carotid revascularization procedures in the progression or reversal of cognitive dysfunction.

An international, multispecialty, expert-based Delphi Consensus document on controversial issues in the management of patients with asymptomatic and symptomatic carotid stenosis.

OBJECTIVE: Despite the publication of various national/international guidelines, several questions concerning the management of patients with asymptomatic (AsxCS) and symptomatic (SxCS) carotid stenosis remain unanswered. The aim of this international, multi-specialty, expert-based Delphi Consensus document was to address these issues to help clinicians make decisions when guidelines are unclear. METHODS: Fourteen controversial topics were identified. A three-round Delphi Consensus process was performed including 61 experts. The aim of Round 1 was to investigate the differing views and opinions regarding these unresolved topics. In Round 2, clarifications were asked from each participant. In Round 3, the questionnaire was resent to all participants for their final vote. Consensus was reached when ≥75% of experts agreed on a specific response. RESULTS: Most experts agreed that: (1) the current periprocedural/in-hospital stroke/death thresholds for performing a carotid intervention should be lowered from 6% to 4% in patients with SxCS and from 3% to 2% in patients with AsxCS; (2) the time threshold for a patient being considered "recently symptomatic" should be reduced from the current definition of "6 months" to 3 months or less; (3) 80% to 99% AsxCS carries a higher risk of stroke compared with 60% to 79% AsxCS; (4) factors beyond the grade of stenosis and symptoms should be added to the indications for revascularization in AsxCS patients (eg, plaque features of vulnerability and silent infarctions on brain computed tomography scans); and (5) shunting should be used selectively, rather than always or never. Consensus could not be reached on the remaining topics due to conflicting, inadequate, or controversial evidence. CONCLUSIONS: The present international, multi-specialty expert-based Delphi Consensus document attempted to provide responses to several unanswered/unresolved issues. However, consensus could not be achieved on some topics, highlighting areas requiring future research.

Carotid Plaque-RADS: A Novel Stroke Risk Classification System.

BACKGROUND: Carotid artery atherosclerosis is highly prevalent in the general population and is a well-established risk factor for acute ischemic stroke. Although the morphological characteristics of vulnerable plaques are well recognized, there is a lack of consensus in reporting and interpreting carotid plaque features. OBJECTIVES: The aim of this paper is to establish a consistent and comprehensive approach for imaging and reporting carotid plaque by introducing the Plaque-RADS (Reporting and Data System) score. METHODS: A panel of experts recognized the necessity to develop a classification system for carotid plaque and its defining characteristics. Using a multimodality analysis approach, the Plaque-RADS categories were established through consensus, drawing on existing published reports. RESULTS: The authors present a universal classification that is applicable to both researchers and clinicians. The Plaque-RADS score offers a morphological assessment in addition to the prevailing quantitative parameter of "stenosis." The Plaque-RADS score spans from grade 1 (indicating complete absence of plaque) to grade 4 (representing complicated plaque). Accompanying visual examples are included to facilitate a clear understanding of the Plaque-RADS categories. CONCLUSIONS: Plaque-RADS is a standardized and reliable system of reporting carotid plaque composition and morphology via different imaging modalities, such as ultrasound, computed tomography, and magnetic resonance imaging. This scoring system has the potential to help in the precise identification of patients who may benefit from exclusive medical intervention and those who require alternative treatments, thereby enhancing patient care. A standardized lexicon and structured reporting promise to enhance communication between radiologists, referring clinicians, and scientists.

Reduction of lower-limb edema in patients with chronic venous disease by micronized purified flavonoid fraction: a systematic literature review and meta-analysis.

INTRODUCTION: The prevalence of lower limb edema is high among patients with chronic venous disease (CVD). Several clinical studies with various designs have assessed the effect of micronized purified flavonoid fraction (MPFF) on edema. The aim of this work was to provide a comprehensive and accurate evaluation of the reduction in ankle and calf circumference as an indicator of lower limb edema reduction in patients with CVD treated with MPFF by combining studies that use different designs in a single group meta-analysis. EVIDENCE ACQUISITION: We conducted a systematic literature review in April 2022 based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) criteria to identify prospective studies investigating the effect of oral MPFF treatment 1000 mg/day on ankle and calf circumference in patients with CVD. Studies with population including at least one patient with an ulcer were excluded. All prospective studies irrespectively of design (i.e., interventional and non-interventional studies, randomized controlled trials (RCTs), non-randomized studies, studies without a control or reference treatment) were eligible. The Medline, Embase and Cochrane databases were searched. Endpoints were ankle and calf circumference measurements and their overall mean change from baseline estimated with random-effects meta-analysis methods. The evaluation criterion feeling of swelling was also analyzed as a standardized mean change (SMC) with 95% confidence intervals after combination of quantitative scales. EVIDENCE SYNTHESIS: Among 861 articles identified, eight studies (five RCTs including one placebo-controlled, three non-comparative studies) met the criteria. The overall population consisted of 1635 patients, predominantly female (89% ranging from 64% to 94%) with a mean age of 47 years ranging from 41 to 48 years. Mean reduction in ankle circumference was 6.0 mm (95%CI: 3.6 to 8.4; P<0.001) and 7.0 mm (95%CI: 0.9 to 13.1; P=0.024) after two and at least six months of treatment respectively. The results were similar when considering the study type RCTs and non-RCTs. Mean reduction in calf circumference was 5.7 mm (95%CI: 2.8 to 8.6; P<0.001) and 6.7 mm (95%CI: 5.2 to 8.1; P<0.001), at two months and at the last post-baseline evaluation respectively. Heterogeneity among studies was statistically significant (degree of consistency I2=93.5%; P<0.001 and I2=81.1%, P<0.01 for ankle and calf circumference, respectively). In the three studies reporting the effect on feeling of swelling a significant standardized mean change (SMC) reduction of 2.2 (95%CI: 0.2 to 4.2; P=0.028) on a quantitative scale was observed after two months of treatment with MPFF. CONCLUSIONS: MPFF appeared to be effective in reducing ankle and calf circumference as well as feeling of swelling irrespective of study design. The circumference reduction is present at short and long term, suggesting that benefit occurs early and is maintained overtime. Despite the observed heterogeneity among included studies, this meta-analysis supports the significant therapeutic efficacy of MPFF in reducing lower-limb edema in patients with CVD. The complete video presentation of the work is available online at www.minervamedica.it (Supplementary Digital Material 1: Supplementary Video 1, 5 min, 192 MB).

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.

A Pharmaceutical Paradigm for Cardiovascular Composite Risk Assessment Using Novel Radiogenomics Risk Predictors in Precision Explainable Artificial Intelligence Framework: Clinical Trial Tool.

BACKGROUND: Cardiovascular disease (CVD) is challenging to diagnose and treat since symptoms appear late during the progression of atherosclerosis. Conventional risk factors alone are not always sufficient to properly categorize at-risk patients, and clinical risk scores are inadequate in predicting cardiac events. Integrating genomic-based biomarkers (GBBM) found in plasma/serum samples with novel non-invasive radiomics-based biomarkers (RBBM) such as plaque area, plaque burden, and maximum plaque height can improve composite CVD risk prediction in the pharmaceutical paradigm. These biomarkers consider several pathways involved in the pathophysiology of atherosclerosis disease leading to CVD. OBJECTIVE: This review proposes two hypotheses: (i) The composite biomarkers are strongly correlated and can be used to detect the severity of CVD/Stroke precisely, and (ii) an explainable artificial intelligence (XAI)-based composite risk CVD/Stroke model with survival analysis using deep learning (DL) can predict in preventive, precision, and personalized (aiP3) framework benefiting the pharmaceutical paradigm. METHOD: The PRISMA search technique resulted in 214 studies assessing composite biomarkers using radiogenomics for CVD/Stroke. The study presents a XAI model using AtheroEdgeTM 4.0 to determine the risk of CVD/Stroke in the pharmaceutical framework using the radiogenomics biomarkers. CONCLUSIONS: Our observations suggest that the composite CVD risk biomarkers using radiogenomics provide a new dimension to CVD/Stroke risk assessment. The proposed review suggests a unique, unbiased, and XAI model based on AtheroEdgeTM 4.0 that can predict the composite risk of CVD/Stroke using radiogenomics in the pharmaceutical paradigm.

Cardiovascular disease/stroke risk stratification in deep learning framework: a review.

The global mortality rate is known to be the highest due to cardiovascular disease (CVD). Thus, preventive, and early CVD risk identification in a non-invasive manner is vital as healthcare cost is increasing day by day. Conventional methods for risk prediction of CVD lack robustness due to the non-linear relationship between risk factors and cardiovascular events in multi-ethnic cohorts. Few recently proposed machine learning-based risk stratification reviews without deep learning (DL) integration. The proposed study focuses on CVD risk stratification by the use of techniques mainly solo deep learning (SDL) and hybrid deep learning (HDL). Using a PRISMA model, 286 DL-based CVD studies were selected and analyzed. The databases included were Science Direct, IEEE Xplore, PubMed, and Google Scholar. This review is focused on different SDL and HDL architectures, their characteristics, applications, scientific and clinical validation, along with plaque tissue characterization for CVD/stroke risk stratification. Since signal processing methods are also crucial, the study further briefly presented Electrocardiogram (ECG)-based solutions. Finally, the study presented the risk due to bias in AI systems. The risk of bias tools used were (I) ranking method (RBS), (II) region-based map (RBM), (III) radial bias area (RBA), (IV) prediction model risk of bias assessment tool (PROBAST), and (V) risk of bias in non-randomized studies-of interventions (ROBINS-I). The surrogate carotid ultrasound image was mostly used in the UNet-based DL framework for arterial wall segmentation. Ground truth (GT) selection is vital for reducing the risk of bias (RoB) for CVD risk stratification. It was observed that the convolutional neural network (CNN) algorithms were widely used since the feature extraction process was automated. The ensemble-based DL techniques for risk stratification in CVD are likely to supersede the SDL and HDL paradigms. Due to the reliability, high accuracy, and faster execution on dedicated hardware, these DL methods for CVD risk assessment are powerful and promising. The risk of bias in DL methods can be best reduced by considering multicentre data collection and clinical evaluation.

Why do guidelines recommend screening for abdominal aortic aneurysms, but not for asymptomatic carotid stenosis? A plea for a randomized controlled trial.

BACKGROUND: Current guidelines do not recommend screening for asymptomatic carotid artery stenosis (AsxCS). The rationale behind this recommendation is that detection of AsxCS may lead to an unnecessary carotid intervention. In contrast, screening for abdominal aortic aneurysms is strongly recommended. METHODS: A critical analysis of the literature was performed to evaluate the implications of detecting AsxCS. RESULTS: Patients with AsxCS are at high risk for future stroke, myocardial infarction and vascular death. Population-wide screening for AsxCS should not be recommended. Additionally, screening of high-risk individuals for AsxCS with the purpose of identifying candidates for a carotid intervention is inappropriate. Instead, selective screening for AsxCS should be considered and should be viewed as an opportunity to identify individuals at high risk for atherosclerotic cardiovascular disease and future cardiovascular events for the timely initiation of intensive medical therapy and risk factor modification. CONCLUSIONS: Although mass screening should not be recommended, there are several arguments suggesting that selective screening for AsxCS should be considered. The rationale supporting such selective screening is to optimize risk factor control and to initiate intensive medical therapy for prevention of future cardiovascular events, rather than to identify candidates for an intervention.

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.

Attention-Based UNet Deep Learning Model for Plaque Segmentation in Carotid Ultrasound for Stroke Risk Stratification: An Artificial Intelligence Paradigm.

Stroke and cardiovascular diseases (CVD) significantly affect the world population. The early detection of such events may prevent the burden of death and costly surgery. Conventional methods are neither automated nor clinically accurate. Artificial Intelligence-based methods of automatically detecting and predicting the severity of CVD and stroke in their early stages are of prime importance. This study proposes an attention-channel-based UNet deep learning (DL) model that identifies the carotid plaques in the internal carotid artery (ICA) and common carotid artery (CCA) images. Our experiments consist of 970 ICA images from the UK, 379 CCA images from diabetic Japanese patients, and 300 CCA images from post-menopausal women from Hong Kong. We combined both CCA images to form an integrated database of 679 images. A rotation transformation technique was applied to 679 CCA images, doubling the database for the experiments. The cross-validation K5 (80% training: 20% testing) protocol was applied for accuracy determination. The results of the Attention-UNet model are benchmarked against UNet, UNet++, and UNet3P models. Visual plaque segmentation showed improvement in the Attention-UNet results compared to the other three models. The correlation coefficient (CC) value for Attention-UNet is 0.96, compared to 0.93, 0.96, and 0.92 for UNet, UNet++, and UNet3P models. Similarly, the AUC value for Attention-UNet is 0.97, compared to 0.964, 0.966, and 0.965 for other models. Conclusively, the Attention-UNet model is beneficial in segmenting very bright and fuzzy plaque images that are hard to diagnose using other methods. Further, we present a multi-ethnic, multi-center, racial bias-free study of stroke risk assessment.

Performance of SCORE2 and SCORE2-OP risk algorithms in a Cypriot cohort.

BACKGROUND: SCORE2 and SCORE2-OP algorithms and associated online calculators provide a new and easy method of estimating the 10-year cardiovascular risk in apparently healthy Europeans. The aim of the study was to determine the performance of these algorithms in terms of discrimination and calibration in the cohort of the Cyprus Epidemiological Study on Atherosclerosis (CESA), not only for the 10-year risk for myocardial infarction (MI), stroke and cardiovascular death, but also for all types of atherosclerotic cardiovascular events (ASCVE). METHODS: SCORE2 and SCORE2-OP for low-risk regions were calculated in a non-diabetic subset of CESA consisting of 908 people (mean age±SD: 57.8±10.5; range 40-89; 58.8% female) using baseline risk factors. Mean follow-up was 13.2±3.7 years (range 1-17) with 89 primary endpoints (MI, stroke and cardiovascular death) and 136 secondary endpoints (primary endpoints, angina, cardiac failure, coronary revascularization, transient ischemic attack, claudication and critical limb ischemia). RESULTS: The C-statistic for the prediction of the primary endpoint for all ages was 0.76 (95% CI 0.70 to 0.81) and the observed 10-year event rate was similar to the predicted one. However, the observed 10-year rate for secondary events was similar to the estimated one only when the algorithm for high-risk regions was used. CONCLUSIONS: SCORE2 and SCORE2-OP moderate risk algorithms perform well in the Cypriot population for predicting the 10-year risk for MI, stroke and fatal cardiovascular disease. However, an estimate of the 10-year risk for all ASCVD events is best calculated from the high-risk algorithm.

New Perspectives on Micronised Purified Flavonoid Fraction in Chronic Venous Disease: From Microvalves to Clinical Effectiveness.

The importance of chronic venous disease (CVD), as a cause of reduced quality of life and increased costs to healthcare systems, is expected to rise in parallel with population aging and the increasing prevalence of obesity. Venoactive drugs (VADs) are frequently used to treat the symptoms and signs of CVD. The most commonly used and widely studied VAD, micronised purified flavonoid fraction (MPFF), is effective at all stages of CVD, and has been shown to significantly reduce leg pain, leg heaviness and swelling, as well as ankle oedema and functional discomfort, in clinical trials. Recently, experiments employing animal models of CVD have demonstrated that MPFF has anti-inflammatory and venotonic effects at the microvalve level, and a pilot clinical study in patients with CVD has provided support for these findings. Collectively, these results suggest that early initiation of MPFF treatment may have the potential to favourably alter the clinical course of the disease, although further clinical data are required to confirm these findings. International guidelines on CVD management strongly recommend MPFF to reduce symptoms and improve quality of life. Studies are now needed to investigate the impact of long-term treatment on disease progression.

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.

Deep Learning Paradigm for Cardiovascular Disease/Stroke Risk Stratification in Parkinson's Disease Affected by COVID-19: A Narrative Review.

Background and Motivation: Parkinson's disease (PD) is one of the most serious, non-curable, and expensive to treat. Recently, machine learning (ML) has shown to be able to predict cardiovascular/stroke risk in PD patients. The presence of COVID-19 causes the ML systems to become severely non-linear and poses challenges in cardiovascular/stroke risk stratification. Further, due to comorbidity, sample size constraints, and poor scientific and clinical validation techniques, there have been no well-explained ML paradigms. Deep neural networks are powerful learning machines that generalize non-linear conditions. This study presents a novel investigation of deep learning (DL) solutions for CVD/stroke risk prediction in PD patients affected by the COVID-19 framework. Method: The PRISMA search strategy was used for the selection of 292 studies closely associated with the effect of PD on CVD risk in the COVID-19 framework. We study the hypothesis that PD in the presence of COVID-19 can cause more harm to the heart and brain than in non-COVID-19 conditions. COVID-19 lung damage severity can be used as a covariate during DL training model designs. We, therefore, propose a DL model for the estimation of, (i) COVID-19 lesions in computed tomography (CT) scans and (ii) combining the covariates of PD, COVID-19 lesions, office and laboratory arterial atherosclerotic image-based biomarkers, and medicine usage for the PD patients for the design of DL point-based models for CVD/stroke risk stratification. Results: We validated the feasibility of CVD/stroke risk stratification in PD patients in the presence of a COVID-19 environment and this was also verified. DL architectures like long short-term memory (LSTM), and recurrent neural network (RNN) were studied for CVD/stroke risk stratification showing powerful designs. Lastly, we examined the artificial intelligence bias and provided recommendations for early detection of CVD/stroke in PD patients in the presence of COVID-19. Conclusion: The DL is a very powerful tool for predicting CVD/stroke risk in PD patients affected by COVID-19.

COVLIAS 2.0-cXAI: Cloud-Based Explainable Deep Learning System for COVID-19 Lesion Localization in Computed Tomography Scans.

Background: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based explainable AI, the "COVLIAS 2.0-cXAI" system using four kinds of class activation maps (CAM) models. Methodology: Our cohort consisted of ~6000 CT slices from two sources (Croatia, 80 COVID-19 patients and Italy, 15 control patients). COVLIAS 2.0-cXAI design consisted of three stages: (i) automated lung segmentation using hybrid deep learning ResNet-UNet model by automatic adjustment of Hounsfield units, hyperparameter optimization, and parallel and distributed training, (ii) classification using three kinds of DenseNet (DN) models (DN-121, DN-169, DN-201), and (iii) validation using four kinds of CAM visualization techniques: gradient-weighted class activation mapping (Grad-CAM), Grad-CAM++, score-weighted CAM (Score-CAM), and FasterScore-CAM. The COVLIAS 2.0-cXAI was validated by three trained senior radiologists for its stability and reliability. The Friedman test was also performed on the scores of the three radiologists. Results: The ResNet-UNet segmentation model resulted in dice similarity of 0.96, Jaccard index of 0.93, a correlation coefficient of 0.99, with a figure-of-merit of 95.99%, while the classifier accuracies for the three DN nets (DN-121, DN-169, and DN-201) were 98%, 98%, and 99% with a loss of ~0.003, ~0.0025, and ~0.002 using 50 epochs, respectively. The mean AUC for all three DN models was 0.99 (p < 0.0001). The COVLIAS 2.0-cXAI showed 80% scans for mean alignment index (MAI) between heatmaps and gold standard, a score of four out of five, establishing the system for clinical settings. Conclusions: The COVLIAS 2.0-cXAI successfully showed a cloud-based explainable AI system for lesion localization in lung CT scans.