AI models predicting Risk of Cardio Vascular Diseases - The Limitations, Challenges and Necessity for Regulatory Framework
Abstract
Artificial intelligence (AI) algorithms have changed the landscape of Cardio Vascular Disease (CVD) risk assessment and demonstrated a better performance mainly due to its ability to handle the input nonlinear variations. Further, it has the flexibility to add risk factors derived from medical imaging modalities using Computer Vision (CV). Most commonly used algorithms in CVD risk predications were classification and regression tress (CART).
Though most of the developed models have shown good accuracy but have not considered risks factors or dependent variables related to specific population which plays an integral role in predicting the risk of CVDs. This Include gender specific clinical risk factors (hormonal changes, bone density etc.), metrological, chronological data, exposure to environmental pollutants, race, genotype, hereditary, dietary intake, physical inactivity, psychological stress etc. Secondly the existing models have not included the weighing and grading of the risks, as all factors won’t contribute equally to the Cardiac Risk. Importantly predictive models can be readily used within the populations in which they were developed but practically they often give a less than satisfactory performance, when applied to another population because of the Inter genetic variations especially in CVDs.
India accounts for one-fifth of these deaths worldwide especially in younger population. The results of Global Burden of Disease study state age-standardized CVD death rate of 272 per 100000 populations in India, which is much higher than that of global average of 225. CVDs strike Indians a decade earlier than the western population. For Indians, particular causes of concern in CVD are early age of onset, rapid progression and high mortality rate. Indians are known to have the highest coronary artery disease (CAD) rates, and the conventional risk factors fail to explain this increased risk.
In Indian context, aggressive screening tests should begin at an early age and will be beneficial for early detection and treatment to reduce the mortality. Hence there is necessity to develop upgraded AI models specific to a subset of population (Indian, Caucasoid, Dravidian etc.) inclusive of the risk factors in that specific population. Secondly allotting weighing, grading of risk factors in the model will provide accurate cardiac risk prediction compared to other approaches.
The regulatory and policy landscape for AI is an emerging issue in jurisdictions globally, including in the European Union and in supra-national bodies like the IEEE, OECD and others. Since 2016, a wave of AI ethics guidelines has been published in order to maintain social control over the technology.
Downloads
References
doi: 10.1016/S0140-6736(20)30925-9.
2. Mensah GA, Roth GA, Fuster V. The global burden of cardiovascular diseases and risk factors: 2020 and beyond. J Am Coll Cardiol. 2019;74:2529–32.
3. Mensah GA, Wei GS, Sorlie PD, et al. Decline in cardiovascular mortality. Circ Res 2017;120:366–80.
4. Murray CJL, Aravkin AY, Zheng P, et al. Global burden of 87 risk factors in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396:1223–49.
5. GBD Compare [Internet]. vizhub; 2022 [cited 2022 Mar 25]. Available from:
https://vizhub. healthdata.org/gbd-compare
6. Global Health Data Exchange [Internet]. ghdx.healthdata; 2022 [cited 2022 Mar 23]. Available from:
http://ghdx.healthdata.org/
7. Tolonen H, Mähönen M, Asplund K, et al. Do trends in population levels of blood pressure and other cardiovascular risk factors explain trends in stroke event rates? Stroke. 2002; 33:2367–75.
8. GBD 2017 Risk Factor Collaborators. Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks for 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018; 392:1923–94.
9. Husain MJ, Datta BK, Kostova D, et al. Access to cardiovascular disease and hypertension medicines in developing countries: an analysis of essential medicine lists, price, availability, and affordability. J Am Heart Assoc. 2020;9:e015302.
10. Roth GA, Forouzanfar MH, Moran AE, et al. Demographic and epidemiologic drivers of global J A C C Vol. No.-2020 cardiovascular mortality. N Engl J Med. 2015;372:1333–41.
11. United Nations, Department of Economic and Social Affairs, Population Division. World Population Ageing 2019: Highlights. ST/ESA/SER.A/430 [Internet]. United Nations; 2019 [cited 2022 Mar 21]. Available from:
https://www.un.org/en/development/desa/population/publications/pdf/ ageing/WorldPopulationAgeing2019-Highlights. pdf.
12. United Nations, Department of Economic and Social Affairs, Population Division. World Population Prospects 2019: Highlights. ST/ESA/SER.A/ 423 [Internet]. United Nations; 2019 [cited 2022 Mar 21]. Available from: https://population.un.org/ wpp/Publications/Files/WPP2019_Highlights.pdf.
13. Reynolds I, Page RL, Boxer RS. Cardiovascular health and healthy aging. In: Coll PP, editor. Healthy Aging: A Complete Guide to Clinical Management. Cham, Switzerland: Springer International Publishing. 2019:31-51.
14. Fuster V. Global burden of cardiovascular disease: time to implement feasible strategies and to monitor results. J Am Coll Cardiol. 2014;64:520–2.
15. Lee DS, Austin PC, Rouleau JL, Liu PP, Naimark D, Tu JV. Predicting mortality among patients hospitalized for heart failure: derivation and validation of a clinical model. Journal of the American Medical Association 2003;290:2581-87.
16. Tu JV, Jaglal SB, Naylor CD et al. Multicenter validation of a risk index for mortality, intensive care unit stay, and overall hospital length of stay after cardiac surgery. Circulation. 1995;91:677–684.
17. Lee KL, Woodlief LH, Topol EJ et al. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction. Circulation. 1995;91:1659-68.
18. Sullivan LM, Massaro JM, D’Agostino RB. Presentation of multivariate data for clinical use: the Framingham study risk score functions. Statistics in Medicine. 2004; 23:1631-60.
19. Iezzoni LI. Risk Adjustment for Measuring Healthcare Outcomes (2nd edn). Health Administration Press: Chicago, IL; 1997.
20. Geelhoed M, Boerrigter AO, Camfield P, Geerts AT, Arts W, Smith B, Camfield C. The accuracy of outcome prediction models for childhood-onset epilepsy. Epilepsia. 2005;46:1526–1532.
21. Nishida N, Tanaka M, Hayashi N, Nagata H, Takeshita T, Nakayama K, Morimoto K, Shizukuishi S. Determination of smoking and obesity as periodontitis risks using the classification and regression tree method. Journal of Periodontology. 2005;76:923-928.
22. Dessein PH, Joffe BI, Veller MG, Stevens BA, Tobias M, Reddi K, Stanwix AE. Traditional and non-traditional cardiovascular risk factors are associated with atherosclerosis in rheumatoid arthritis. Journal of Rheumatology. 2005;32:435-442.
23. Kuchibhatla M, Fillenbaum GG. Alternative statistical approaches to identifying dementia in a community-dwelling sample. Aging and Mental Health. 2003;7:383-389.
24. Avila PC, Segal MR, Wong HH, Boushey HA, Fahy JV. Predictors of late asthmatic response. Logistic regression and classification tree analyses. American Journal of Respiratory and Critical Care Medicine. 2000;161:2092-2095.
25. Bhavnani SM, Drake JA, Forrest A, Deinhart JA, Jones RN, Biedenbach DJ, Ballow CH. A nationwise, multicenter, case–control study comparing risk factors, treatment, and outcome for vancomycin-resistant and -susceptible enterococcal bacteremia. Diagnostic Microbiology and Infectious Disease. 2000;36:145-158.
26. Wietlisbach V, Vader JP, Porchet F, Costanza MC, Burnand B. Statistical approaches in the development of clinical practice guidelines from expert panels: the case of laminectomy in sciatica patients. Medical Care. 1999; 37:785–797.
27. Roehrborn CG, Malice M, Cook TJ, Girman CJ. Clinical predictors of spontaneous acute urinary retention in men with LUTS and clinical BPH: a comprehensive analysis of the pooled placebo groups of several large clinical trials. Urology. 2001;58:210–216.
28. Knuiman MW, Vu HT, Segal MR. An empirical comparison of multivariable methods for estimating risk of death from coronary heart disease. Journal of Cardiovascular Risk. 1997;4:127-134.
29. Litvan I, Campbell G, Mangone CA, Verny M, McKee A, Chaudhuri KR, Jellinger K, Pearce RK, D’Olhaberriague L. Which clinical features differentiate progressive supranuclear palsy (Steele–Richardson– Olszewski syndrome) from related disorders? A clinicopathological study. Brain 1997;120:65-74.
30. Pilote L, Miller DP, Califf RM, Rao JS, Weaver WD, Topol EJ. Determinants of the use of coronary angiography and revascularization after thrombolysis for acute myocardial infarction. New England Journal of Medicine. 1996;335:1198–1205.
31. Bhattacharyya S, Siegel ER, Petersen GM, Chari ST, Suva LJ, Haun RS. Diagnosis of pancreatic cancer using serum proteomic profiling. Neoplasia. 2004;6:674–686.
32. Taylor WJ, Marchesoni A, Arreghini M, Sokoll K, Helliwell PS. A comparison of the performance characteristics of classification criteria for the diagnosis of psoriatic arthritis. Seminars in Arthritis and Rheumatism. 2004;34:575–584.
33. Seligman DA, Pullinger AG. Improved interaction models of temporomandibular joint anatomic relationships in asymptomatic subjects and patients with disc displacement with or without reduction. Journal of Orofacial Pain. 2004;18:192–202.
34. Stalans LJ, Yarnold PR, Seng M, Olson DE, Repp M. Identifying three types of violent offenders and predicting violent recidivism while on probation: a classification tree analysis. Law and Human Behavior. 2004;28:253–271.
35. Arena VC, Sussman NB, Mazumdar S, Yu S, Macina OT. The utility of structure-activity relationship (SAR) models for prediction and covariate selection in developmental toxicity: comparative analysis of logistic regression and decision three models. Sar and Qsar in Environmental Research. 2004;15:1–18.
36. Schwarzer G, Nagata T, Mattern D, Schmelzeisen R, Schumacher M. Comparison of fuzzy inference, logistic regression, and classification trees (CART). Prediction of cervical lymph node metastasis in carcinoma of the tongue. Methods of Information in Medicine 2003; 42:572–577.
37. Thwaites GE, Chau TT, Stepniewska K, Phu NH, Chuong LV, Sinh DX, et al. Diagnosis of adult tuberculous meningitis by use of clinical and laboratory features. Lancet. 2002;360:1287-92.
38. Pullinger AG, Seligman DA, John MT, Harkins S. Multifactorial modelling of temporomandibular anatomic and orthopaedic relationships in normal versus undifferentiated disk displacement joints. Journal of Prosthetic Dentistry. 2002;87:289-97.
39. Pullinger AG, Seligman DA. Multifactorial analysis of differences in temporomandibular joint hard tissue anatomic relationships between disk displacement with and without reduction in women. Journal of Prosthetic Dentistry. 2001;86:407-19.
40. Seligman DA, Pullinger AG. Analysis of occlusal variables, dental attrition, and age for distinguishing healthy controls from female patients with intracapsular temporomandibular disorders. Journal of Prosthetic Dentistry. 2000;83:76-82.
41. Woolas RP, Conaway MR, Xu F, Jacobs IJ, Yu Y, Daly L, Davies AP, O’Briant K, Berchuck A, Soper JT et al. Combinations of multiple serum markers are superior to individual assays for discriminating malignant from benign pelvic masses. Gynaecologic Oncology. 1995; 59:111-16.
42. Li D, German D, Lulla S, Thomas RG, Wilson SR. Prospective study of hospitalization for asthma. A preliminary risk factor model. American Journal of Respiratory and Critical Care Medicine. 1995;151:647-55.
43. Hasford J, Ansari H, Lehmann K. CART and logistic regression analyses of risk factors for first dose hypotension by an ACE-inhibitor. Therapie. 1993; 48:479-82.
44. Stewart PW, Stamm JW. Classification tree prediction models for dental caries from clinical, microbiological, and interview data. Journal of Dental Research. 1991; 70:1239-51.
45. Sauerbrei W, Madjar H, Prompeler HJ. Differentiation of benign and malignant breast tumors by logistic regression and a classification tree using Doppler flow signals. Methods of Information in Medicine. 1998; 37:226-34.
46. Rosenfeld B, Lewis C. Assessing violence risk in stalking cases: a regression tree approach. Law and Human Behavior. 2005;29:343-57.
47. Garzotto M, Beer TM, Hudson RG, Peters L, Hsieh YC, Barrera E, Klein T, Mori M. Improved detection of prostate cancer using classification and regression tree analysis. Journal of Clinical Oncology. 2005;23:4322-29.
48. El-Solh AA, Sikka P, Ramadan F. Outcome of older patients with severe pneumonia predicted by recursive partitioning. Journal of the American Geriatrics Society. 2001;49:1614-21.
49. Tsien CL, Fraser HS, Long WJ, Kennedy RL. Using classification tree and logistic regression methods to diagnose myocardial infarction. Medinfo. 1998;9:493-97.
50. Selker HP, Griffith JL, Patil S, Long WJ, D’Agostino RB. A comparison of performance of mathematical predictive methods for medical diagnosis: identifying acute cardiac ischemia among emergency department patients. Journal of Investigative Medicine. 1995; 43:468-76.
51. James KE, White RF, Kraemer HC. Repeated split sample validation to assess logistic regression and recursive partitioning: an application to the prediction of cognitive impairment. Statistics in Medicine. 2005; 24:3019-35.
52. Long WJ, Griffith JL, Selker HP, D’Agostino RB. A comparison of logistic regression to decision-tree induction in a medical domain. Computers and Biomedical Research. 1993;26:74–97.
53. Nelson LM, Bloch DA, Longstreth Jr WT, Shi H. Recursive partitioning for the identification of disease risk subgroups: a case–control study of subarachnoid haemorrhage. Journal of Clinical Epidemiology. 1998;51:199–209.
54. Lemon SC, Roy J, Clark MA, Friedmann PD, Rakowski W. Classification and regression tree analysis in public health: methodological review and comparison with logistic regression. Annals of Behavioral Medicine. 2003;26:172–181.
55. Ennis M, Hinton G, Naylor D, Revow M, Tibshirani R. A comparison of statistical learning methods on the GUSTO database. Statistics in Medicine. 1998;17:2501-08.
56. Harrell Jr FE. Regression Modeling Strategies. Springer: New York; 2001.
57. Van JC, Cessie LS. Predictive value of statistical models. Statistics in Medicine. 1990;8:1303-25.
58. Tu JV, Donovan LR, Lee DS, Austin PC, Ko DT, Wang JT, Newman AM. Quality of Cardiac Care in Ontario. Institute for Clinical Evaluative Sciences: Toronto, Ont.; 2004.
59. Bernd WW, Jan CW, Carolin G. Artificial Intelligence and the Public Sector-Applications and Challenges. International Journal of Public Administration. 2018; 42(7):596 615.
60. Miriam CB. "Towards Intelligent Regulation of Artificial Intelligence". European Journal of Risk Regulation. 2019;10(1):41-59.
61. Alessandro M, Maria SE. An evidence-based methodology for human rights impact assessment (HRIA) in the development of AI data-intensive systems. Computer Law & Security Review.2021;41:105561.
62. Vladimir PK, Evgenia RY, Evgeniy KA, Dina RK. Revisiting the Place of Artificial Intelligence in Society and the State. Artificial Intelligence: Anthropogenic Nature vs. Social Origin, Cham: Springer International Publishing. 2020;359-64.
63. Bernd WW, Jan WC, Benjamin JS. The Dark Sides of Artificial Intelligence: An Integrated AI Governance Framework for Public Administration. International Journal of Public Administration. 2020;43(9):818-29.
64. Ajay K, Vidur M, Kevin S, Ajit K, Saurabh J. Concepts in U.S. Food and Drug Administration Regulation of Artificial Intelligence for Medical Imaging. American Journal of Roentgenology. 2019;213(4):886-88.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The International Journal of Drug Regulatory affairs require a formal written transfer of copyright from the author(s) for each article published. We therefore ask you to complete and return this form, retaining a copy for your records. Your cooperation is essential and appreciated. Any delay will result in a delay in publication.
I/we have read and agree with the terms and conditions stated Page 2 of this agreement and I/we hereby confirm the transfer of all copyrights in and relating to the above-named manuscript, in all forms and media, now or hereafter known, to the International Journal of Drug Regulatory affairs, effective from the date stated below. I/we acknowledge that the IJDRA is relying on this agreement in publishing the above-named manuscript. However, this agreement will be null and void if the manuscript is not published in the IJDRA.
Download link for COPYRIGHT FORM
