Unified Screening for Multiple Diseases

Yiğit Narter, Alihan Hüyük, Mihaela van der Schaar, Cem Tekin

Abstract: Current screening programs that focus on improving patient health while minimizing screening costs are tailored for individual diseases. Designing unified screening programs for multiple diseases requires carefully balancing competing disease risks, which is an open problem. In this work, we address this problem by casting unified screening as a referral problem, in which we choose to activate a subset of screening policies for individual diseases by accounting for competing risks that influence patient outcomes. We introduce a novel optimization framework that incorporates disease risks, budget constraints, and diagnostic error limits and characterize the structural properties of the optimal referral policy. For the unified screening of two diseases, we show that the optimal activation threshold for the screening of one disease depends on the risk of the other, resulting in decision boundaries with distinct risk-dependent profiles. We compare our unified model with independent screening programs that apply isolated activation thresholds for screening of each disease. Our approach optimizes screening decisions collectively, improving overall survival outcomes, particularly for patients with high disease risks.

Lay Summary:

Most medical screening programs are designed for just one disease at a time. But in real life, people can face risks for several diseases at once, and choosing how and when to screen for each one can be tricky, especially when there are limited resources like time, money, and testing capacity.

Our study explores a better way to handle this: instead of looking at each disease separately, we treat screening as a joint decision-making problem. We built a computer model that helps decide which disease screenings should be done together and when, based on a patient’s specific risk for each disease. The model takes into account how diseases can affect each other’s outcomes, and aims to improve survival while staying within realistic limits for cost and accuracy.

We found that this unified approach leads to better health outcomes, especially for high-risk patients, compared to treating each disease in isolation.

Link To Code: https://github.com/ynarter/UniScreen

Primary Area: Applications->Health / Medicine

Keywords: Unified screening, healthcare, competing risks, constrained optimization, threshold policy

Submission Number: 15730