Exploring Algorithmic Explainability: Generating Explainable AI Insights for Personalized Clinical Decision Support Focused on Cannabis Intoxication in Young Adults

TL;DR

The paper tackles the lack of transparency in algorithmic decisions guiding cannabis-related clinical interventions by integrating passive sensing with four Explainable AI approaches: SHAP, SkopeRules, decision trees, and counterfactual explanations, analyzed at the level of individual participants. It demonstrates a per-person modeling pipeline using $5$-minute sensor windows to predict cannabis intoxication, revealing how sleep, heart rate, location, noise, and mobility contexts shape intoxication predictions. The key contributions include case-based demonstrations of interpretable explanations, rules, and counterfactuals that clinicians can use to tailor interventions, along with a discussion of practical CDSS implications and ethical considerations. Overall, the work advances personalized, transparent AI in substance-use care, offering actionable insights for clinicians and public health researchers and outlining directions for broader validation and usability testing.

Abstract

This study explores the possibility of facilitating algorithmic decision-making by combining interpretable artificial intelligence (XAI) techniques with sensor data, with the aim of providing researchers and clinicians with personalized analyses of cannabis intoxication behavior. SHAP analyzes the importance and quantifies the impact of specific factors such as environmental noise or heart rate, enabling clinicians to pinpoint influential behaviors and environmental conditions. SkopeRules simplify the understanding of cannabis use for a specific activity or environmental use. Decision trees provide a clear visualization of how factors interact to influence cannabis consumption. Counterfactual models help identify key changes in behaviors or conditions that may alter cannabis use outcomes, to guide effective individualized intervention strategies. This multidimensional analytical approach not only unveils changes in behavioral and physiological states after cannabis use, such as frequent fluctuations in activity states, nontraditional sleep patterns, and specific use habits at different times and places, but also highlights the significance of individual differences in responses to cannabis use. These insights carry profound implications for clinicians seeking to gain a deeper understanding of the diverse needs of their patients and for tailoring precisely targeted intervention strategies. Furthermore, our findings highlight the pivotal role that XAI technologies could play in enhancing the transparency and interpretability of Clinical Decision Support Systems (CDSS), with a particular focus on substance misuse treatment. This research significantly contributes to ongoing initiatives aimed at advancing clinical practices that aim to prevent and reduce cannabis-related harms to health, positioning XAI as a supportive tool for clinicians and researchers alike.

Figures (5)

• Figure 1: Person-Specific XAI Pipeline
• Figure 2: SHAP Individual Waterfall Graph (P24)
• Figure 3: SHAP Individual Waterfall Graph (P63)
• Figure 4: Intrinsically Interpretable Models (P24)
• Figure 5: Intrinsically Interpretable Models (P63)