Explainability in Healthcare AI
Abstract: This tutorial extensively covers the definitions, nuances, challenges, and requirements for the design of interpretable and explainable machine learning models and systems in healthcare. We discuss many uses in which interpretable machine learning models are needed in healthcare and how they should be deployed. Additionally, we explore the landscape of recent advances to address the challenges model interpretability in healthcare and also describe how one would go about choosing the right interpretable machine learning algorithm for a given problem in healthcare.
Cite: Muhammad Aurangzeb Ahmad, Dr. Carly Eckert M.D., Ankur Teredesai Explainable Models for Healthcare AIKDD London, United Kingdom August 19-23, 2018
Automatic Detection of Excess Healthcare Spending and Cost Variation in ACOs
Abstract: There are more than nine hundred Accountable Care Organizations (ACOs) in the United States, both in the public and private sector, serving millions of patients across the country in a process to transition from fee-for-service to a value-based-care model for healthcare delivery in an effort to contain expenditures. Identifying fraud, waste, and abuse resulting in superfluous expenditures associated with care delivery is central to the success of ACOs and for making the cost of healthcare sustainable. In theory, such expendi- tures should be easily identifiable with large amounts of historical data. However, to the best of our knowledge there is no data mining framework that systematically addresses the problem of identifying unwarranted variation in expenditures on high dimensional claims data using unsupervised machine learning techniques. In this paper we propose methods to uncover unwarranted variation in health- care spending by automatically extracting reference groups of peer- providers from the data and then detecting high cost outliers within these groups. We demonstrate the utility of our proposed frame- work on datasets from a large ACO in the United States to success- fully identify unwarranted variation in therapeutic procedures even in low cost claims that had previously gone unnoticed.
Cite: Eric Liu, Muhammad Aurangzeb Ahmad, Carly Eckert, Anderson Nascimento, Martine De Cock, Karthik Padthe, Ankur Teredesai, Greg McKelvey Automatic Detection of Excess Healthcare Spending and Cost Variation in ACOs SDM 2018 Workshop on Data Mining for Medicine and Healthcare May 5, 2018 San Diego, USA
Death vs. Data Science: Predicting End of Life
Abstract: Death is an inevitable part of life and while it cannot be delayed indefinitely it is possible to predict with some certainty when the health of a person is going to deteriorate. In this paper, we predict risk of mortality for patients from two large hospital systems in the Pacific Northwest. Using medical claims and electronic medical records (EMR) data we greatly improve prediction for risk of mortality and ex- plore machine learning models with explanations for end of life predictions. The insights that are derived from the predic- tions can then be used to improve the quality of patient care towards the end of life.
Cite: Muhammad Aurangzeb Ahmad, Carly Eckert, Greg McKelvey, Kiyana Zolfagar, Anam Zahid, Ankur Teredesai. Death vs. Data Science: Predicting End of Life IAAI February 2-6, 2018
- Muhammad Aurangzeb Ahmad, Carly Eckert, Ankur Teredesai, Greg McKelvey Machine Learning Based Explanations to Support Risk of Readmission Prediction Models AIMed 2017 Laguna Niguel, CA December 11-14, 2017
- Muhammad Aurangzeb Ahmad, Carly Eckert Machine Learning in the Prediction of Organ Transplant Processes AIMed 2017 Laguna Niguel, CA December 11-14, 2017
- Matthew Eckert, Muhammad Aurangzeb Ahmad Deep Learning and Thermal Imaging Technology In the Detection of Surgical Site Infection AIMed 2017 Laguna Niguel, CA December 11-14, 2017
- Xinran Liu, Carly Eckert, Muhammad Aurangzeb Ahmad Development of Sepsis Prediction Algorithm for Clinical Use AIMed 2017 Laguna Niguel, CA December 11-14, 2017
- Xinran Liu, Greg McKelvey, Muhammad Aurangzeb Ahmad, Rosemary Grant, David Carlbom Using Machine learning to predict Hospital acquired Sepsis AMIA 2017 Annual Symposium Washington, DC November 4 – 8, 2017