DP-RuL: Differentially-Private Rule Learning for Clinical Decision Support Systems

TL;DR

This paper introduces DP-RuL, a novel framework for learning privacy-preserving rule-based clinical decision support systems using local differential privacy, enabling effective rule discovery while protecting sensitive patient data.

Contribution

It develops a new distributed rule learning method with LDP, integrating Monte-Carlo Tree Search and adaptive privacy budgeting for clinical decision support.

Findings

High coverage of rules learned at low privacy budgets

Effective rule discovery in three clinical datasets

Maintains clinical utility with strong privacy guarantees

Abstract

Serious privacy concerns arise with the use of patient data in rule-based clinical decision support systems (CDSS). The goal of a privacy-preserving CDSS is to learn a population ruleset from individual clients' local rulesets, while protecting the potentially sensitive information contained in the rulesets. We present the first work focused on this problem and develop a framework for learning population rulesets with local differential privacy (LDP), suitable for use within a distributed CDSS and other distributed settings. Our rule discovery protocol uses a Monte-Carlo Tree Search (MCTS) method integrated with LDP to search a rule grammar in a structured way and find rule structures clients are likely to have. Randomized response queries are sent to clients to determine promising paths to search within the rule grammar. In addition, we introduce an adaptive budget allocation method which dynamically determines how much privacy loss budget to use at each query, resulting in better privacy-utility trade-offs. We evaluate our approach using three clinical datasets and find that we are able to learn population rulesets with high coverage (breadth of rules) and clinical utility even at low privacy loss budgets.