Cardiovascular/stroke risk prevention: A new machine learning framework integrating carotid ultrasound image-based phenotypes and its harmonics with conventional risk factors

Motivation Machine learning (ML)-based stroke risk stratification systems have typically focused on conventional risk factors (CRF) (AtheroRisk-conventional). Besides CRF, carotid ultrasound image phenotypes (CUSIP) have shown to be powerful phenotypes risk stratification. This is the first ML study of its kind that integrates CUSIP and CRF for risk stratification (AtheroRisk-integrated) and compares against AtheroRisk-conventional. Methods Two types of ML-based setups called (i) AtheroRisk-integrated and (ii) AtheroRisk-conventional were developed using random forest (RF) classifiers. AtheroRisk-conventional uses a feature set of 13 CRF such as age, gender, hemoglobin A1c, fasting blood sugar, low-density lipoprotein, and high-density lipoprotein (HDL) cholesterol, total cholesterol (TC), a ratio of TC and HDL, hypertension, smoking, family history, triglyceride, and ultrasound-based carotid plaque score. AtheroRisk-integrated system uses the feature set of 38 features with a combination of 13 CRF and 25 CUSIP features (6 types of current CUSIP, 6 types of 10-year CUSIP, 12 types of quadratic CUSIP (harmonics), and age-adjusted grayscale median). Logistic regression approach was used to select the significant features on which the RF classifier was trained. The performance of both ML systems was evaluated by area-under-the-curve (AUC) statistics computed using a leave-one-out cross-validation protocol. Results Left and right common carotid arteries of 202 Japanese patients were retrospectively examined to obtain 404 ultrasound scans. RF classifier showed higher improvement in AUC (~57%) for leave-one-out cross-validation protocol. Using RF classifier, AUC statistics for AtheroRisk-integrated system was higher (AUC = 0.99,p-value<0.001) compared to AtheroRisk-conventional (AUC = 0.63,p-value<0.001). Conclusion The AtheroRisk-integrated ML system outperforms the AtheroRisk-conventional ML system using RF classifier.

[1]  Filippo Molinari,et al.  Carotid IMT Variability (IMTV) and Its Validation in Symptomatic versus Asymptomatic Italian Population: Can This Be a Useful Index for Studying Symptomaticity? , 2012, Echocardiography.

[2]  U. Rajendra Acharya,et al.  Completely Automated Multiresolution Edge Snapper—A New Technique for an Accurate Carotid Ultrasound IMT Measurement: Clinical Validation and Benchmarking on a Multi-Institutional Database , 2012, IEEE Transactions on Image Processing.

[3]  Petros Sfikakis,et al.  Rheumatoid Arthritis: Atherosclerosis Imaging and Cardiovascular Risk Assessment Using Machine and Deep Learning–Based Tissue Characterization , 2019, Current Atherosclerosis Reports.

[4]  Jasjit S Suri,et al.  Data mining framework for fatty liver disease classification in ultrasound: A hybrid feature extraction paradigm. , 2012, Medical physics.

[5]  Ben Goldacre,et al.  Prediction of Cardiovascular Risk Using Framingham, ASSIGN and QRISK2: How Well Do They Predict Individual Rather than Population Risk? , 2014, PloS one.

[6]  Michiel L Bots,et al.  Carotid intima-media thickness as a surrogate marker for cardiovascular disease in intervention studies , 2006, Current medical research and opinion.

[7]  Ioannis A. Kakadiaris,et al.  Machine Learning Outperforms ACC/AHA CVD Risk Calculator in MESA , 2018, Journal of the American Heart Association.

[8]  S. Gabriel,et al.  Challenges of developing a cardiovascular risk calculator for patients with rheumatoid arthritis , 2017, PloS one.

[9]  Jasjit S Suri,et al.  A Survey on Coronary Atherosclerotic Plaque Tissue Characterization in Intravascular Optical Coherence Tomography , 2018, Current Atherosclerosis Reports.

[10]  PulmonaryRehabilitation 2013 ACC/AHA Guideline on the Assessment of CardiovascularRisk: A Report of the American College of Cardiology/American Heart Association TaskForce on Practice Guidelines , 2014 .

[11]  J. Suri,et al.  Computed tomography carotid wall plaque characterization using a combination of discrete wavelet transform and texture features: A pilot study , 2013, Proceedings of the Institution of Mechanical Engineers. Part H, Journal of engineering in medicine.

[12]  Jasjit S. Suri,et al.  A Special Report on Changing Trends in Preventive Stroke/Cardiovascular Risk Assessment Via B-Mode Ultrasonography , 2019, Current Atherosclerosis Reports.

[13]  P. Kolh,et al.  ESC/EAS Guidelines for the management of dyslipidaemias: the Task Force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and the European Atherosclerosis Society (EAS). , 2011, European heart journal.

[14]  Aditya M. Sharma,et al.  Integration of eGFR biomarker in image-based CV/Stroke risk calculator: a south Asian-Indian diabetes cohort with moderate chronic kidney disease. , 2020, International angiology : a journal of the International Union of Angiology.

[15]  Elena Tremoli,et al.  Carotid intima-media thickness by B-mode ultrasound as surrogate of coronary atherosclerosis: correlation with quantitative coronary angiography and coronary intravascular ultrasound findings. , 2007, European heart journal.

[16]  J. Suri,et al.  Atherosclerotic risk stratification strategy for carotid arteries using texture-based features. , 2012, Ultrasound in medicine & biology.

[17]  A. Bhansali,et al.  Agreement between Framingham Risk Score and United Kingdom Prospective Diabetes Study Risk Engine in Identifying High Coronary Heart Disease Risk in North Indian Population , 2015, Diabetes & metabolism journal.

[18]  J. Kai,et al.  Can machine-learning improve cardiovascular risk prediction using routine clinical data? , 2017, PloS one.

[19]  Filippo Molinari,et al.  Comparison between manual and automated analysis for the quantification of carotid wall by using sonography. A validation study with CT. , 2012, European journal of radiology.

[20]  J. Robson,et al.  Lipid modification: cardiovascular risk assessment and the modification of blood lipids for the primary and secondary prevention of cardiovascular disease , 2007, Heart.

[21]  Daniel H Solomon,et al.  Comparison of cardiovascular risk algorithms in patients with vs without rheumatoid arthritis and the role of C-reactive protein in predicting cardiovascular outcomes in rheumatoid arthritis , 2017, Rheumatology.

[22]  K. Chien,et al.  Comparing the Consistency and Performance of Various Coronary Heart Disease Prediction Models for Primary Prevention Using a National Representative Cohort in Taiwan. , 2018, Circulation journal : official journal of the Japanese Circulation Society.

[23]  U Rajendra Acharya,et al.  Automated carotid IMT measurement and its validation in low contrast ultrasound database of 885 patient Indian population epidemiological study: results of AtheroEdge™ Software. , 2012, International angiology : a journal of the International Union of Angiology.

[24]  Jasjit S. Suri,et al.  Atherosclerosis disease management , 2011 .

[25]  J. Shaw,et al.  Guidelines for the management of absolute cardiovascular disease risk , 2012 .

[26]  Ian Graham,et al.  ESC/EAS Guidelines for the management of dyslipidaemias: the Task Force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and the European Atherosclerosis Society (EAS). , 2011, Atherosclerosis.

[27]  Jennifer G. Robinson,et al.  2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines , 2014, Circulation.

[28]  Jasjit S Suri,et al.  A low-cost machine learning-based cardiovascular/stroke risk assessment system: integration of conventional factors with image phenotypes. , 2019, Cardiovascular diagnosis and therapy.

[29]  James F Sallis,et al.  AHA Guidelines for Primary Prevention of Cardiovascular Disease and Stroke: 2002 Update: Consensus Panel Guide to Comprehensive Risk Reduction for Adult Patients Without Coronary or Other Atherosclerotic Vascular Diseases. American Heart Association Science Advisory and Coordinating Committee. , 2002, Circulation.

[30]  G. M. Allan,et al.  Comparison of cardiovascular disease risk calculators , 2014, Current opinion in lipidology.

[31]  Aditya M. Sharma,et al.  Low-cost preventive screening using carotid ultrasound in patients with diabetes. , 2020, Frontiers in bioscience.

[32]  Filippo Molinari,et al.  Intima-media thickness: setting a standard for a completely automated method of ultrasound measurement , 2010, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control.

[33]  Package of Essential Noncommunicable (PEN) Disease Interventions for Primary Health Care in Low-Resource Settings , 2010 .

[34]  Jasjit S Suri,et al.  Echolucency-based phenotype in carotid atherosclerosis disease for risk stratification of diabetes patients. , 2018, Diabetes research and clinical practice.

[35]  Jasjit S. Suri,et al.  Performance evaluation of 10-year ultrasound image-based stroke/cardiovascular (CV) risk calculator by comparing against ten conventional CV risk calculators: A diabetic study , 2019, Comput. Biol. Medicine.

[36]  Jasjit S Suri,et al.  Cardiovascular/stroke risk predictive calculators: a comparison between statistical and machine learning models. , 2020, Cardiovascular diagnosis and therapy.

[37]  Jasjit S. Suri,et al.  Effect of carotid image-based phenotypes on cardiovascular risk calculator: AECRS1.0 , 2019, Medical & Biological Engineering & Computing.