Explainable AI in Healthcare

5 Pages Posted: 25 Apr 2019

See all articles by Sujata Khedkar

Sujata Khedkar

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT)

Vignesh Subramanian

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT)

Gayatri Shinde

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT)

Priyanka Gandhi

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT)

Date Written: April 8, 2019

Abstract

The traditional cardiology prediction model includes factors like age, total cholesterol, HDL cholesterol, smoking, blood pressure, and diabetes.

But the machine algorithms turned up a wider array of factors in their models, including: COPD, severe mental illness, prescription of oral corticosteroids, triglyceride levels, atrial fibrillations, chronic kidney disease, and rheumatoid arthritis. Allowing machines to learn risk factors in a huge number of patients makes for better predictions of heart attacks. Machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others. However, the use of machine learning algorithms has been limited due to the lack of interpretability of more complex models, reducing the trust of users in the model. The solution to the limitation of the ‘black-box’ nature of machine-learning algorithms, in particular neural networks, which are difficult to interpret, the inherent complexity in how the risk factor variables are interacting and their independent effects on the outcome is achieved through explainable AI.

Keywords: Heart Failure, Predictive Modeling, Deep Learning, RNN, LIME, Explainability, interpretability, attention

Suggested Citation

Khedkar, Sujata and Subramanian, Vignesh and Shinde, Gayatri and Gandhi, Priyanka, Explainable AI in Healthcare (April 8, 2019). 2nd International Conference on Advances in Science & Technology (ICAST) 2019 on 8th, 9th April 2019 by K J Somaiya Institute of Engineering & Information Technology, Mumbai, India, Available at SSRN: https://ssrn.com/abstract=3367686 or http://dx.doi.org/10.2139/ssrn.3367686

Sujata Khedkar

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT) ( email )

India

Vignesh Subramanian (Contact Author)

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT) ( email )

India

Gayatri Shinde

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT) ( email )

India

Priyanka Gandhi

University of Mumbai - Vivekanand Education Society's Institute of Technology (VESIT) ( email )

India

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
1,147
Abstract Views
2,956
Rank
34,575
PlumX Metrics