Intra- and inter-operator reproducibility of automated cloud-based carotid lumen diameter ultrasound measurement

COVLIAS 3.0: cloud-based quantized hybrid UNet3+ deep learning for COVID-19 lesion detection in lung computed tomography

Background and novelty

When RT-PCR is ineffective in early diagnosis and understanding of COVID-19 severity, Computed Tomography (CT) scans are needed for COVID diagnosis, especially in patients having high ground-glass opacities, consolidations, and crazy paving. Radiologists find the manual method for lesion detection in CT very challenging and tedious. Previously solo deep learning (SDL) was tried but they had low to moderate-level performance. This study presents two new cloud-based quantized deep learning UNet3+ hybrid (HDL) models, which incorporated full-scale skip connections to enhance and improve the detections.

Methodology

Annotations from expert radiologists were used to train one SDL (UNet3+), and two HDL models, namely, VGG-UNet3+ and ResNet-UNet3+. For accuracy, 5-fold cross-validation protocols, training on 3,500 CT scans, and testing on unseen 500 CT scans were adopted in the cloud framework. Two kinds of loss functions were used: Dice Similarity (DS) and binary cross-entropy (BCE). Performance was evaluated using (i) Area error, (ii) DS, (iii) Jaccard Index, (iii) Bland-Altman, and (iv) Correlation plots.

Results

Among the two HDL models, ResNet-UNet3+ was superior to UNet3+ by 17 and 10% for Dice and BCE loss. The models were further compressed using quantization showing a percentage size reduction of 66.76, 36.64, and 46.23%, respectively, for UNet3+, VGG-UNet3+, and ResNet-UNet3+. Its stability and reliability were proved by statistical tests such as the Mann-Whitney, Paired t-Test, Wilcoxon test, and Friedman test all of which had a p < 0.001.

Conclusion

Full-scale skip connections of UNet3+ with VGG and ResNet in HDL framework proved the hypothesis showing powerful results improving the detection accuracy of COVID-19.

UltraAIGenomics: Artificial Intelligence-Based Cardiovascular Disease Risk Assessment by Fusion of Ultrasound-Based Radiomics and Genomics Features for Preventive, Personalized and Precision Medicine: A Narrative Review

Cardiovascular disease (CVD) diagnosis and treatment are challenging since symptoms appear late in the disease's progression. Despite clinical risk scores, cardiac event prediction is inadequate, and many at-risk patients are not adequately categorised by conventional risk factors alone. Integrating genomic-based biomarkers (GBBM), specifically those found in plasma and/or serum samples, along with novel non-invasive radiomic-based biomarkers (RBBM) such as plaque area and plaque burden can improve the overall specificity of CVD risk. This review proposes two hypotheses: (i) RBBM and GBBM biomarkers have a strong correlation and can be used to detect the severity of CVD and stroke precisely, and (ii) introduces a proposed artificial intelligence (AI)-based preventive, precision, and personalized ( aiP 3 ) CVD/Stroke risk model. The PRISMA search selected 246 studies for the CVD/Stroke risk. It showed that using the RBBM and GBBM biomarkers, deep learning (DL) modelscould be used for CVD/Stroke risk stratification in the aiP 3 framework. Furthermore, we present a concise overview of platelet function, complete blood count (CBC), and diagnostic methods. As part of the AI paradigm, we discuss explainability, pruning, bias, and benchmarking against previous studies and their potential impacts. The review proposes the integration of RBBM and GBBM, an innovative solution streamlined in the DL paradigm for predicting CVD/Stroke risk in the aiP 3 framework. The combination of RBBM and GBBM introduces a powerful CVD/Stroke risk assessment paradigm. aiP 3 model signifies a promising advancement in CVD/Stroke risk assessment.

Deep Learning Paradigm and Its Bias for Coronary Artery Wall Segmentation in Intravascular Ultrasound Scans: A Closer Look

BACKGROUND AND MOTIVATION

Coronary artery disease (CAD) has the highest mortality rate; therefore, its diagnosis is vital. Intravascular ultrasound (IVUS) is a high-resolution imaging solution that can image coronary arteries, but the diagnosis software via wall segmentation and quantification has been evolving. In this study, a deep learning (DL) paradigm was explored along with its bias.

METHODS

Using a PRISMA model, 145 best UNet-based and non-UNet-based methods for wall segmentation were selected and analyzed for their characteristics and scientific and clinical validation. This study computed the coronary wall thickness by estimating the inner and outer borders of the coronary artery IVUS cross-sectional scans. Further, the review explored the bias in the DL system for the first time when it comes to wall segmentation in IVUS scans. Three bias methods, namely (i) ranking, (ii) radial, and (iii) regional area, were applied and compared using a Venn diagram. Finally, the study presented explainable AI (XAI) paradigms in the DL framework.

FINDINGS AND CONCLUSIONS

UNet provides a powerful paradigm for the segmentation of coronary walls in IVUS scans due to its ability to extract automated features at different scales in encoders, reconstruct the segmented image using decoders, and embed the variants in skip connections. Most of the research was hampered by a lack of motivation for XAI and pruned AI (PAI) models. None of the UNet models met the criteria for bias-free design. For clinical assessment and settings, it is necessary to move from a paper-to-practice approach.

DermAI 1.0: A Robust, Generalized, and Novel Attention-Enabled Ensemble-Based Transfer Learning Paradigm for Multiclass Classification of Skin Lesion Images

Skin lesion classification plays a crucial role in dermatology, aiding in the early detection, diagnosis, and management of life-threatening malignant lesions. However, standalone transfer learning (TL) models failed to deliver optimal performance. In this study, we present an attention-enabled ensemble-based deep learning technique, a powerful, novel, and generalized method for extracting features for the classification of skin lesions. This technique holds significant promise in enhancing diagnostic accuracy by using seven pre-trained TL models for classification. Six ensemble-based DL (EBDL) models were created using stacking, softmax voting, and weighted average techniques. Furthermore, we investigated the attention mechanism as an effective paradigm and created seven attention-enabled transfer learning (aeTL) models before branching out to construct three attention-enabled ensemble-based DL (aeEBDL) models to create a reliable, adaptive, and generalized paradigm. The mean accuracy of the TL models is 95.30%, and the use of an ensemble-based paradigm increased it by 4.22%, to 99.52%. The aeTL models' performance was superior to the TL models in accuracy by 3.01%, and aeEBDL models outperformed aeTL models by 1.29%. Statistical tests show significant p-value and Kappa coefficient along with a 99.6% reliability index for the aeEBDL models. The approach is highly effective and generalized for the classification of skin lesions.

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.

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.

COVLIAS 1.0Lesion vs. MedSeg: An Artificial Intelligence Framework for Automated Lesion Segmentation in COVID-19 Lung Computed Tomography Scans

Background: COVID-19 is a disease with multiple variants, and is quickly spreading throughout the world. It is crucial to identify patients who are suspected of having COVID-19 early, because the vaccine is not readily available in certain parts of the world. Methodology: Lung computed tomography (CT) imaging can be used to diagnose COVID-19 as an alternative to the RT-PCR test in some cases. The occurrence of ground-glass opacities in the lung region is a characteristic of COVID-19 in chest CT scans, and these are daunting to locate and segment manually. The proposed study consists of a combination of solo deep learning (DL) and hybrid DL (HDL) models to tackle the lesion location and segmentation more quickly. One DL and four HDL models—namely, PSPNet, VGG-SegNet, ResNet-SegNet, VGG-UNet, and ResNet-UNet—were trained by an expert radiologist. The training scheme adopted a fivefold cross-validation strategy on a cohort of 3000 images selected from a set of 40 COVID-19-positive individuals. Results: The proposed variability study uses tracings from two trained radiologists as part of the validation. Five artificial intelligence (AI) models were benchmarked against MedSeg. The best AI model, ResNet-UNet, was superior to MedSeg by 9% and 15% for Dice and Jaccard, respectively, when compared against MD 1, and by 4% and 8%, respectively, when compared against MD 2. Statistical tests—namely, the Mann−Whitney test, paired t-test, and Wilcoxon test—demonstrated its stability and reliability, with p < 0.0001. The online system for each slice was <1 s. Conclusions: The AI models reliably located and segmented COVID-19 lesions in CT scans. The COVLIAS 1.0Lesion lesion locator passed the intervariability test.

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.

COVLIAS 1.0 vs. MedSeg: Artificial Intelligence-Based Comparative Study for Automated COVID-19 Computed Tomography Lung Segmentation in Italian and Croatian Cohorts

(1) Background: COVID-19 computed tomography (CT) lung segmentation is critical for COVID lung severity diagnosis. Earlier proposed approaches during 2020-2021 were semiautomated or automated but not accurate, user-friendly, and industry-standard benchmarked. The proposed study compared the COVID Lung Image Analysis System, COVLIAS 1.0 (GBTI, Inc., and AtheroPointTM, Roseville, CA, USA, referred to as COVLIAS), against MedSeg, a web-based Artificial Intelligence (AI) segmentation tool, where COVLIAS uses hybrid deep learning (HDL) models for CT lung segmentation. (2) Materials and Methods: The proposed study used 5000 ITALIAN COVID-19 positive CT lung images collected from 72 patients (experimental data) that confirmed the reverse transcription-polymerase chain reaction (RT-PCR) test. Two hybrid AI models from the COVLIAS system, namely, VGG-SegNet (HDL 1) and ResNet-SegNet (HDL 2), were used to segment the CT lungs. As part of the results, we compared both COVLIAS and MedSeg against two manual delineations (MD 1 and MD 2) using (i) Bland-Altman plots, (ii) Correlation coefficient (CC) plots, (iii) Receiver operating characteristic curve, and (iv) Figure of Merit and (v) visual overlays. A cohort of 500 CROATIA COVID-19 positive CT lung images (validation data) was used. A previously trained COVLIAS model was directly applied to the validation data (as part of Unseen-AI) to segment the CT lungs and compare them against MedSeg. (3) Result: For the experimental data, the four CCs between COVLIAS (HDL 1) vs. MD 1, COVLIAS (HDL 1) vs. MD 2, COVLIAS (HDL 2) vs. MD 1, and COVLIAS (HDL 2) vs. MD 2 were 0.96, 0.96, 0.96, and 0.96, respectively. The mean value of the COVLIAS system for the above four readings was 0.96. CC between MedSeg vs. MD 1 and MedSeg vs. MD 2 was 0.98 and 0.98, respectively. Both had a mean value of 0.98. On the validation data, the CC between COVLIAS (HDL 1) vs. MedSeg and COVLIAS (HDL 2) vs. MedSeg was 0.98 and 0.99, respectively. For the experimental data, the difference between the mean values for COVLIAS and MedSeg showed a difference of <2.5%, meeting the standard of equivalence. The average running times for COVLIAS and MedSeg on a single lung CT slice were ~4 s and ~10 s, respectively. (4) Conclusions: The performances of COVLIAS and MedSeg were similar. However, COVLIAS showed improved computing time over MedSeg.

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.

Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography

Background: For COVID-19 lung severity, segmentation of lungs on computed tomography (CT) is the first crucial step. Current deep learning (DL)-based Artificial Intelligence (AI) models have a bias in the training stage of segmentation because only one set of ground truth (GT) annotations are evaluated. We propose a robust and stable inter-variability analysis of CT lung segmentation in COVID-19 to avoid the effect of bias. Methodology: The proposed inter-variability study consists of two GT tracers for lung segmentation on chest CT. Three AI models, PSP Net, VGG-SegNet, and ResNet-SegNet, were trained using GT annotations. We hypothesized that if AI models are trained on the GT tracings from multiple experience levels, and if the AI performance on the test data between these AI models is within the 5% range, one can consider such an AI model robust and unbiased. The K5 protocol (training to testing: 80%:20%) was adapted. Ten kinds of metrics were used for performance evaluation. Results: The database consisted of 5000 CT chest images from 72 COVID-19-infected patients. By computing the coefficient of correlations (CC) between the output of the two AI models trained corresponding to the two GT tracers, computing their differences in their CC, and repeating the process for all three AI-models, we show the differences as 0%, 0.51%, and 2.04% (all < 5%), thereby validating the hypothesis. The performance was comparable; however, it had the following order: ResNet-SegNet > PSP Net > VGG-SegNet. Conclusions: The AI models were clinically robust and stable during the inter-variability analysis on the CT lung segmentation on COVID-19 patients.

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.

Carotid Ultrasound Boundary Study (CUBS): An Open Multicenter Analysis of Computerized Intima-Media Thickness Measurement Systems and Their Clinical Impact

Common carotid intima-media thickness (CIMT) is a commonly used marker for atherosclerosis and is often computed in carotid ultrasound images. An analysis of different computerized techniques for CIMT measurement and their clinical impacts on the same patient data set is lacking. Here we compared and assessed five computerized CIMT algorithms against three expert analysts' manual measurements on a data set of 1088 patients from two centers. Inter- and intra-observer variability was assessed, and the computerized CIMT values were compared with those manually obtained. The CIMT measurements were used to assess the correlation with clinical parameters, cardiovascular event prediction through a generalized linear model and the Kaplan-Meier hazard ratio. CIMT measurements obtained with a skilled analyst's segmentation and the computerized segmentation were comparable in statistical analyses, suggesting they can be used interchangeably for CIMT quantification and clinical outcome investigation. To facilitate future studies, the entire data set used is made publicly available for the community at http://dx.doi.org/10.17632/fpv535fss7.1.

A Review on Joint Carotid Intima-Media Thickness and Plaque Area Measurement in Ultrasound for Cardiovascular/Stroke Risk Monitoring: Artificial Intelligence Framework

Cardiovascular diseases (CVDs) are the top ten leading causes of death worldwide. Atherosclerosis disease in the arteries is the main cause of the CVD, leading to myocardial infarction and stroke. The two primary image-based phenotypes used for monitoring the atherosclerosis burden is carotid intima-media thickness (cIMT) and plaque area (PA). Earlier segmentation and measurement methods were based on ad hoc conventional and semi-automated digital imaging solutions, which are unreliable, tedious, slow, and not robust. This study reviews the modern and automated methods such as artificial intelligence (AI)-based. Machine learning (ML) and deep learning (DL) can provide automated techniques in the detection and measurement of cIMT and PA from carotid vascular images. Both ML and DL techniques are examples of supervised learning, i.e., learn from "ground truth" images and transformation of test images that are not part of the training. This review summarizes (1) the evolution and impact of the fast-changing AI technology on cIMT/PA measurement, (2) the mathematical representations of ML/DL methods, and (3) segmentation approaches for cIMT/PA regions in carotid scans based for (a) region-of-interest detection and (b) lumen-intima and media-adventitia interface detection using ML/DL frameworks. AI-based methods for cIMT/PA segmentation have emerged for CVD/stroke risk monitoring and may expand to the recommended parameters for atherosclerosis assessment by carotid ultrasound.

Integration of cardiovascular risk assessment with COVID-19 using artificial intelligence

Artificial Intelligence (AI), in general, refers to the machines (or computers) that mimic "cognitive" functions that we associate with our mind, such as "learning" and "solving problem". New biomarkers derived from medical imaging are being discovered and are then fused with non-imaging biomarkers (such as office, laboratory, physiological, genetic, epidemiological, and clinical-based biomarkers) in a big data framework, to develop AI systems. These systems can support risk prediction and monitoring. This perspective narrative shows the powerful methods of AI for tracking cardiovascular risks. We conclude that AI could potentially become an integral part of the COVID-19 disease management system. Countries, large and small, should join hands with the WHO in building biobanks for scientists around the world to build AI-based platforms for tracking the cardiovascular risk assessment during COVID-19 times and long-term follow-up of the survivors.

Does the Carotid Bulb Offer a Better 10-Year CVD/Stroke Risk Assessment Compared to the Common Carotid Artery? A 1516 Ultrasound Scan Study

The objectives of this study are to (1) examine the "10-year cardiovascular risk" in the common carotid artery (CCA) versus carotid bulb using an integrated calculator called "AtheroEdge Composite Risk Score 2.0" (AECRS2.0) and (2) evaluate the performance of AECRS2.0 against "conventional cardiovascular risk calculators." These objectives are met by measuring (1) image-based phenotypes and AECRS2.0 score computation and (2) performance evaluation of AECRS2.0 against 12 conventional cardiovascular risk calculators. The Asian-Indian cohort (n = 379) with type 2 diabetes mellitus (T2DM), chronic kidney disease (CKD), or hypertension were retrospectively analyzed by acquiring the 1516 carotid ultrasound scans (mean age: 55 ± 10.1 years, 67% males, ∼92% with T2DM, ∼83% with CKD [stage 1-5], and 87.5% with hypertension [stage 1-2]). The carotid bulb showed a higher 10-year cardiovascular risk compared to the CCA by 18% (P < .0001). Patients with T2DM and/or CKD also followed a similar trend. The carotid bulb demonstrated a superior risk assessment compared to CCA in patients with T2DM and/or CKD by showing: (1) ∼13% better than CCA (0.93 vs 0.82, P = .0001) and (2) ∼29% better compared with 12 types of risk conventional calculators (0.93 vs 0.72, P = .06).

Ultrasound-based stroke/cardiovascular risk stratification using Framingham Risk Score and ASCVD Risk Score based on "Integrated Vascular Age" instead of "Chronological Age": a multi-ethnic study of Asian Indian, Caucasian, and Japanese cohorts

Background

Vascular age (VA) has recently emerged for CVD risk assessment and can either be computed using conventional risk factors (CRF) or by using carotid intima-media thickness (cIMT) derived from carotid ultrasound (CUS). This study investigates a novel method of integrating both CRF and cIMT for estimating VA [so-called integrated VA (IVA)]. Further, the study analyzes and compares CVD/stroke risk using the Framingham Risk Score (FRS)-based risk calculator when adapting IVA against VA.

Methods

The system follows a four-step process: (I) VA using cIMT based using linear-regression (LR) model and its coefficients; (II) VA prediction using ten CRF using a multivariate linear regression (MLR)-based model with gender adjustment; (III) coefficients from the LR-based model and MLR-based model are combined using a linear model to predict the final IVA; (IV) the final step consists of FRS-based risk stratification with IVA as inputs and benchmarked against FRS using conventional method of CA. Area-under-the-curve (AUC) is computed using IVA and benchmarked against CA while taking the response variable as a standardized combination of cIMT and glycated hemoglobin.

Results

The study recruited 648 patients, 202 were Japanese, 314 were Asian Indian, and 132 were Caucasians. Both left and right common carotid arteries (CCA) of all the population were scanned, thus a total of 1,287 ultrasound scans. The 10-year FRS using IVA reported higher AUC (AUC =0.78) compared with 10-year FRS using CA (AUC =0.66) by ~18%.

Conclusions

IVA is an efficient biomarker for risk stratifications for patients in routine practice.

Two-stage artificial intelligence model for jointly measurement of atherosclerotic wall thickness and plaque burden in carotid ultrasound: A screening tool for cardiovascular/stroke risk assessment

MOTIVATION

The early screening of cardiovascular diseases (CVD) can lead to effective treatment. Thus, accurate and reliable atherosclerotic carotid wall detection and plaque measurements are crucial. Current measurement methods are time-consuming and do not utilize the power of knowledge-based paradigms such as artificial intelligence (AI). We present an AI-based methodology for the joint automated detection and measurement of wall thickness and carotid plaque (CP) in the form of carotid intima-media thickness (cIMT) and total plaque area (TPA), a class of AtheroEdge™ system (AtheroPoint™, CA, USA).

METHOD

The novel system consists of two stages, and each stage comprises an independent deep learning (DL) model. In Stage I, the first DL model segregates the common carotid artery (CCA) patches from ultrasound (US) images into the rectangular wall and non-wall patches. The characterized wall patches are integrated to form the region of interest (ROI), which is then fed into Stage II. In Stage II, the second DL model segments the far wall region. Lumen-intima (LI) and media-adventitial (MA) boundaries are then extracted from the wall region, which is then used for cIMT and PA measurement.

RESULTS

Using the database of 250 carotid scans, the cIMT error using the AI model is 0.0935±0.0637 mm, which is lower than those of all previous methods. The PA error is found to be 2.7939±2.3702 mm². The system's correlation coefficient (CC) between AI and ground truth (GT) values for cIMT is 0.99 (p < 0.0001), which is higher compared with the CC of 0.96 (p < 0.0001) shown by the earlier DL method. The CC for PA between AI and GT values is 0.89 (p < 0.0001).

CONCLUSION

A novel AI-based strategy was applied to carotid US images for the joint detection of carotid wall thickness (cWT) and plaque area (PA), followed by cIMT and PA measurement. This AI-based strategy shows improved performance using the patch technique compared with previous methods using full carotid scans.

Morphological Carotid Plaque Area Is Associated With Glomerular Filtration Rate: A Study of South Asian Indian Patients With Diabetes and Chronic Kidney Disease

We evaluated the association between automatically measured carotid total plaque area (TPA) and the estimated glomerular filtration rate (eGFR), a biomarker of chronic kidney disease (CKD). Automated average carotid intima-media thickness (cIMTave) and TPA measurements in carotid ultrasound (CUS) were performed using AtheroEdge (AtheroPoint). Pearson correlation coefficient (CC) was then computed between the TPA and eGFR for (1) males versus females, (2) diabetic versus nondiabetic patients, and (3) between the left and right carotid artery. Overall, 339 South Asian Indian patients with either type 2 diabetes mellitus (T2DM) or CKD, or hypertension (stage 1 or stage 2) were retrospectively analyzed by acquiring cIMTave and TPA measurements of their left and right common carotid arteries (CCA; total CUS: 678, mean age: 54.2 ± 9.8 years; 75.2% males; 93.5% with T2DM). The CC between TPA and eGFR for different scenarios were (1) for males and females -0.25 (P < .001) and -0.35 (P < .001), respectively; (2) for T2DM and non-T2DM -0.26 (P < .001) and -0.49 (P = .02), respectively, and (3) for left and right CCA -0.25 (P < .001) and -0.23 (P < .001), respectively. Automated TPA is an equally reliable biomarker compared with cIMTave for patients with CKD (with or without T2DM) with subclinical atherosclerosis.

Global perspective on carotid intima-media thickness and plaque: should the current measurement guidelines be revisited?

Carotid intima-media thickness (cIMT) and carotid plaque (CP) currently act as risk predictors for CVD/Stroke risk assessment. Over 2000 articles have been published that cover either use cIMT/CP or alterations of cIMT/CP and additional image-based phenotypes to associate cIMT related markers with CVD/Stroke risk. These articles have shown variable results, which likely reflect a lack of standardization in the tools for measurement, risk stratification, and risk assessment. Guidelines for cIMT/CP measurement are influenced by major factors like the atherosclerosis disease itself, conventional risk factors, 10-year measurement tools, types of CVD/Stroke risk calculators, incomplete validation of measurement tools, and the fast pace of computer technology advancements. This review discusses the following major points: 1) the American Society of Echocardiography and Mannheim guidelines for cIMT/CP measurements; 2) forces that influence the guidelines; and 3) calculators for risk stratification and assessment under the influence of advanced intelligence methods. The review also presents the knowledge-based learning strategies such as machine and deep learning which may play a future role in CVD/stroke risk assessment. We conclude that both machine learning and non-machine learning strategies will flourish for current and 10-year CVD/Stroke risk prediction as long as they integrate image-based phenotypes with conventional risk factors.

Ultrasound-based carotid stenosis measurement and risk stratification in diabetic cohort: a deep learning paradigm

BACKGROUND

Stroke is in the top three leading causes of death worldwide. Non-invasive monitoring of stroke can be accomplished via stenosis measurements. The current conventional image-based methods for these measurements are not accurate and reliable. They do not incorporate shape and intelligent learning component in their design.

METHODS

In this study, we propose a deep learning (DL)-based methodology for accurate measurement of stenosis in common carotid artery (CCA) ultrasound (US) scans using a class of AtheroEdge system from AtheroPoint, USA. Three radiologists manually traced the lumen-intima (LI) for the near and the far walls, respectively, which served as a gold standard (GS) for training the DL-based model. Three DL-based systems were developed based on three types of GS.

RESULTS

IRB approved (Toho University, Japan) 407 US scans from 204 patients were collected. The risk was characterized into three classes: low, moderate, and high-risk. The area-under-curve (AUC) corresponding to three DL systems using receiver operating characteristic (ROC) analysis computed were: 0.90, 0.94 and 0.86, respectively.

CONCLUSIONS

Novel DL-based strategy showed reliable, accurate and stable stenosis severity index (SSI) measurements.

A Special Report on Changing Trends in Preventive Stroke/Cardiovascular Risk Assessment Via B-Mode Ultrasonography

PURPOSE OF REVIEW

Cardiovascular disease (CVD) and stroke risk assessment have been largely based on the success of traditional statistically derived risk calculators such as Pooled Cohort Risk Score or Framingham Risk Score. However, over the last decade, automated computational paradigms such as machine learning (ML) and deep learning (DL) techniques have penetrated into a variety of medical domains including CVD/stroke risk assessment. This review is mainly focused on the changing trends in CVD/stroke risk assessment and its stratification from statistical-based models to ML-based paradigms using non-invasive carotid ultrasonography.

RECENT FINDINGS

In this review, ML-based strategies are categorized into two types: non-image (or conventional ML-based) and image-based (or integrated ML-based). The success of conventional (non-image-based) ML-based algorithms lies in the different data-driven patterns or features which are used to train the ML systems. Typically these features are the patients' demographics, serum biomarkers, and multiple clinical parameters. The integrated (image-based) ML-based algorithms integrate the features derived from the ultrasound scans of the arterial walls (such as morphological measurements) with conventional risk factors in ML frameworks. Even though the review covers ML-based system designs for carotid and coronary ultrasonography, the main focus of the review is on CVD/stroke risk scores based on carotid ultrasound. There are two key conclusions from this review: (i) fusion of image-based features with conventional cardiovascular risk factors can lead to more accurate CVD/stroke risk stratification; (ii) the ability to handle multiple sources of information in big data framework using artificial intelligence-based paradigms (such as ML and DL) is likely to be the future in preventive CVD/stroke risk assessment.

Effect of carotid image-based phenotypes on cardiovascular risk calculator: AECRS1.0

Today, the 10-year cardiovascular risk largely relies on conventional cardiovascular risk factors (CCVRFs) and suffers from the effect of atherosclerotic wall changes. In this study, we present a novel risk calculator AtheroEdge Composite Risk Score (AECRS1.0), designed by fusing CCVRF with ultrasound image-based phenotypes. Ten-year risk was computed using the Framingham Risk Score (FRS), United Kingdom Prospective Diabetes Study 56 (UKPDS56), UKPDS60, Reynolds Risk Score (RRS), and pooled composite risk (PCR) score. AECRS1.0 was computed by measuring the 10-year five carotid phenotypes such as IMT (ave., max., min.), IMT variability, and total plaque area (TPA) by fusing eight CCVRFs and then compositing them. AECRS1.0 was then benchmarked against the five conventional cardiovascular risk calculators by computing the receiver operating characteristics (ROC) and area under curve (AUC) values with a 95% CI. Two hundred four IRB-approved Japanese patients' left/right common carotid arteries (407 ultrasound scans) were collected with a mean age of 69 ± 11 years. The calculators gave the following AUC: FRS, 0.615; UKPDS56, 0.576; UKPDS60, 0.580; RRS, 0.590; PCRS, 0.613; and AECRS1.0, 0.990. When fusing CCVRF, TPA reported the highest AUC of 0.81. The patients were risk-stratified into low, moderate, and high risk using the standardized thresholds. The AECRS1.0 demonstrated the best performance on a Japanese diabetes cohort when compared with five conventional calculators. Graphical abstract AECRS1.0: Carotid ultrasound image phenotype-based 10-year cardiovascular risk calculator. The figure provides brief overview of the proposed carotid image phenotype-based 10-year cardiovascular risk calculator called AECRS1.0. AECRS1.0 was also benchmarked against five conventional cardiovascular risk calculators (Framingham Risk Score (FRS), United Kingdom Prospective Diabetes Study 56 (UKPDS56), UKPDS60, Reynolds Risk Score (RRS), and pooled composite risk (PCR) score).

Rheumatoid Arthritis: Atherosclerosis Imaging and Cardiovascular Risk Assessment Using Machine and Deep Learning-Based Tissue Characterization

PURPOSE OF THE REVIEW

Rheumatoid arthritis (RA) is a chronic, autoimmune disease which may result in a higher risk of cardiovascular (CV) events and stroke. Tissue characterization and risk stratification of patients with rheumatoid arthritis are a challenging problem. Risk stratification of RA patients using traditional risk factor-based calculators either underestimates or overestimates the CV risk. Advancements in medical imaging have facilitated early and accurate CV risk stratification compared to conventional cardiovascular risk calculators.

RECENT FINDING

In recent years, a link between carotid atherosclerosis and rheumatoid arthritis has been widely discussed by multiple studies. Imaging the carotid artery using 2-D ultrasound is a noninvasive, economic, and efficient imaging approach that provides an atherosclerotic plaque tissue-specific image. Such images can help to morphologically characterize the plaque type and accurately measure vital phenotypes such as media wall thickness and wall variability. Intelligence-based paradigms such as machine learning- and deep learning-based techniques not only automate the risk characterization process but also provide an accurate CV risk stratification for better management of RA patients. This review provides a brief understanding of the pathogenesis of RA and its association with carotid atherosclerosis imaged using the B-mode ultrasound technique. Lacunas in traditional risk scores and the role of machine learning-based tissue characterization algorithms are discussed and could facilitate cardiovascular risk assessment in RA patients. The key takeaway points from this review are the following: (i) inflammation is a common link between RA and atherosclerotic plaque buildup, (ii) carotid ultrasound is a better choice to characterize the atherosclerotic plaque tissues in RA patients, and (iii) intelligence-based paradigms are useful for accurate tissue characterization and risk stratification of RA patients.

Nonlinear model for the carotid artery disease 10-year risk prediction by fusing conventional cardiovascular factors to carotid ultrasound image phenotypes: A Japanese diabetes cohort study

MOTIVATION

This study presents a novel nonlinear model which can predict 10-year carotid ultrasound image-based phenotypes by fusing nine traditional cardiovascular risk factors (ethnicity, gender, age, artery type, body mass index, hemoglobin A1c, hypertension, low-density lipoprotein, and smoking) with five types of carotid automated image phenotypes (three types of carotid intima-media thickness (IMT), wall variability, and total plaque area).

METHODOLOGY

Two-step process was adapted: First, five baseline carotid image-based phenotypes were automatically measured using AtheroEdge^™ (AtheroPoint^™ , CA, USA) system by two operators (novice and experienced) and an expert. Second, based on the annual progression rates of cIMT due to nine traditional cardiovascular risk factors, a novel nonlinear model was adapted for 10-year predictions of carotid phenotypes.

RESULTS

Institute review board (IRB) approved 204 Japanese patients' left/right common carotid artery (407 ultrasound scans) was collected with a mean age of 69 ± 11 years. Age and hemoglobin were reported to have a high influence on the 10-year carotid phenotypes. Mean correlation coefficient (CC) between 10-year carotid image-based phenotype and age was improved by 39.35% in males and 25.38% in females. The area under the curves for the 10-year measurements of five phenotypes IMT_ave10yr , IMT_max10yr , IMT_min10yr , IMTV_10yr , and TPA_10yr were 0.96, 0.94, 0.90, 1.0, and 1.0. Inter-operator variability between two operators showed significant CC (P < 0.0001).

CONCLUSIONS

A nonlinear model was developed and validated by fusing nine conventional CV risk factors with current carotid image-based phenotypes for predicting the 10-year carotid ultrasound image-based phenotypes which may be used risk assessment.