Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Diet-ODIN: A Novel Framework for Opioid Misuse Detection with Interpretable Dietary Patterns

Zheyuan Zhang, Zehong Wang, Shifu Hou, Evan Hall, Landon Bachman, Vincent Galassi, Jasmine White, Nitesh V. Chawla, Chuxu Zhang, Yanfang Ye

TL;DR

This paper establishes a large-scale multifaceted dietary benchmark dataset related to opioid users at the first attempt and develops a novel framework to bridge heterogeneous graph (HG) and large language model (LLM) for the identification of users with opioid misuse and the interpretation of their associated dietary patterns.

Abstract

The opioid crisis has been one of the most critical society concerns in the United States. Although the medication assisted treatment (MAT) is recognized as the most effective treatment for opioid misuse and addiction, the various side effects can trigger opioid relapse. In addition to MAT, the dietary nutrition intervention has been demonstrated its importance in opioid misuse prevention and recovery. However, research on the alarming connections between dietary patterns and opioid misuse remain under-explored. In response to this gap, in this paper, we first establish a large-scale multifaceted dietary benchmark dataset related to opioid users at the first attempt and then develop a novel framework - i.e., namely Opioid Misuse Detection with Interpretable Dietary Patterns (Diet-ODIN) - to bridge heterogeneous graph (HG) and large language model (LLM) for the identification of users with opioid misuse and the interpretation of their associated dietary patterns. Specifically, in Diet-ODIN, we first construct an HG to comprehensively incorporate both dietary and health-related information, and then we devise a holistic graph learning framework with noise reduction to fully capitalize both users' individual dietary habits and shared dietary patterns for the detection of users with opioid misuse. To further delve into the intricate correlations between dietary patterns and opioid misuse, we exploit an LLM by utilizing the knowledge obtained from the graph learning model for interpretation. The extensive experimental results based on our established benchmark with quantitative and qualitative measures demonstrate the outstanding performance of Diet-ODIN in exploring the complex interplay between opioid misuse and dietary patterns, by comparison with state-of-the-art baseline methods.

Diet-ODIN: A Novel Framework for Opioid Misuse Detection with Interpretable Dietary Patterns

TL;DR

This paper establishes a large-scale multifaceted dietary benchmark dataset related to opioid users at the first attempt and develops a novel framework to bridge heterogeneous graph (HG) and large language model (LLM) for the identification of users with opioid misuse and the interpretation of their associated dietary patterns.

Abstract

The opioid crisis has been one of the most critical society concerns in the United States. Although the medication assisted treatment (MAT) is recognized as the most effective treatment for opioid misuse and addiction, the various side effects can trigger opioid relapse. In addition to MAT, the dietary nutrition intervention has been demonstrated its importance in opioid misuse prevention and recovery. However, research on the alarming connections between dietary patterns and opioid misuse remain under-explored. In response to this gap, in this paper, we first establish a large-scale multifaceted dietary benchmark dataset related to opioid users at the first attempt and then develop a novel framework - i.e., namely Opioid Misuse Detection with Interpretable Dietary Patterns (Diet-ODIN) - to bridge heterogeneous graph (HG) and large language model (LLM) for the identification of users with opioid misuse and the interpretation of their associated dietary patterns. Specifically, in Diet-ODIN, we first construct an HG to comprehensively incorporate both dietary and health-related information, and then we devise a holistic graph learning framework with noise reduction to fully capitalize both users' individual dietary habits and shared dietary patterns for the detection of users with opioid misuse. To further delve into the intricate correlations between dietary patterns and opioid misuse, we exploit an LLM by utilizing the knowledge obtained from the graph learning model for interpretation. The extensive experimental results based on our established benchmark with quantitative and qualitative measures demonstrate the outstanding performance of Diet-ODIN in exploring the complex interplay between opioid misuse and dietary patterns, by comparison with state-of-the-art baseline methods.
Paper Structure (27 sections, 14 equations, 7 figures, 5 tables)

This paper contains 27 sections, 14 equations, 7 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. NHANES Dietary Graph
  4. The Framework of Diet-ODIN
  5. NR-HGNN for Detecting Users with Opioid Misuse
  6. Shared Dietary Pattern Learning with Marco-level Aggregation
  7. Individual Dietary Habit Learning with Micro-level Aggregation
  8. Graph Refinement with Noise Reduction for the Detection
  9. Bridging NR-HGNN and LLM for Interpretation
  10. Experiments
  11. Experiment Setup
  12. Results
  13. Deep Dive into the Dietary Patterns
  14. Reasoning Analysis
  15. Qualitative Analysis
  16. ...and 12 more sections

Figures (7)

  • Figure 1: The illustration of (a) dietary role in opioid misuse prevention/recovery, and (b) goal of Diet-ODIN.
  • Figure 2: The schema of NHANES Dietary Graph.
  • Figure 3: The overview framework of Diet-ODIN, which consists of (a) a graph learning framework called NR-HGNN for detecting opioid users, and (b) an LLM-powered reasoning module for interpreting the most important dietary patterns.
  • Figure 4: The prompts (highlighted in blue and red) generated from the NR-HGNN for interpreting key dietary patterns that indicate opioid misuse in individuals.
  • Figure 5: T-SNE Visualization of opioid (blue) and non-opioid (orange) users embeddings.
  • ...and 2 more figures

Theorems & Definitions (3)

  • definition 1: Heterogeneous Graphs
  • definition 2: Meta-Path
  • definition 3: Meta-Path-based Neighborhoods