Cost-Sensitive Diagnosis and Learning Leveraging Public Health Data

TL;DR

This paper introduces a health dataset with assigned feature costs and compares recent cost-sensitive learning methods on medical classification tasks, highlighting the importance of considering feature acquisition costs in healthcare analytics.

Contribution

It provides a new health dataset with feature costs and evaluates state-of-the-art cost-sensitive learning methods on multiple medical diagnosis problems.

Findings

Reinforcement learning methods outperform sensitivity-based approaches in certain scenarios.

The dataset enables more realistic evaluation of cost-sensitive health analytics.

Cost-aware methods improve diagnostic accuracy while reducing data collection costs.

Abstract

Traditionally, machine learning algorithms rely on the assumption that all features of a given dataset are available for free. However, there are many concerns such as monetary data collection costs, patient discomfort in medical procedures, and privacy impacts of data collection that require careful consideration in any real-world health analytics system. An efficient solution would only acquire a subset of features based on the value it provides while considering acquisition costs. Moreover, datasets that provide feature costs are very limited, especially in healthcare. In this paper, we provide a health dataset as well as a method for assigning feature costs based on the total level of inconvenience asking for each feature entails. Furthermore, based on the suggested dataset, we provide a comparison of recent and state-of-the-art approaches to cost-sensitive feature acquisition and learning. Specifically, we analyze the performance of major sensitivity-based and reinforcement learning based methods in the literature on three different problems in the health domain, including diabetes, heart disease, and hypertension classification.