When PCOS Meets Eating Disorders: An Explainable AI Approach to Detecting the Hidden Triple Burden

Apoorv Prasad; Susan McRoy

cs.CLApr 15, 2026

When PCOS Meets Eating Disorders: An Explainable AI Approach to Detecting the Hidden Triple Burden

Authors: Apoorv Prasad, Susan McRoy

Organizations: Computer Science, College of Engineering & Applied Science, University of Wisconsin - Milwaukee, Milwaukee, USA

Abstract

Women with polycystic ovary syndrome (PCOS) face substantially elevated risks of body image distress, disordered eating, and metabolic challenges, yet existing natural language processing approaches for detecting these conditions lack transparency and cannot identify co-occurring presentations. We developed small, open-source language models to automatically detect this triple burden in social media posts with grounded explainability. We collected 1,000 PCOS-related posts from six subreddits, with two trained annotators labeling posts using guidelines operationalizing Lee et al. (2017) clinical framework. Three models (Gemma-2-2B, Qwen3-1.7B, DeepSeek-R1-Distill-Qwen-1.5B) were fine-tuned using Low-Rank Adaptation to generate structured explanations with textual evidence. The best model achieved 75.3 percent exact match accuracy on 150 held-out posts, with robust comorbidity detection and strong explainability. Performance declined with diagnostic complexity, indicating their best use is for screening rather than autonomous diagnosis.

Clinical NLP

Clinical natural language processing (NLP) develops methods to extract, interpret, and generate information from healthcare text and conversations.

See all 142 papers

Investigation of Polycystic Ovary Syndrome (PCOS) Diagnosis Using Machine Learning Approaches
Jul 18, 2026Al Zadid Sultan Bin Habib, Md Asif Bin Syed, Md. Ekramul Islam +1Medical Diagnosis Clinical Prediction
Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV 26506, USA · Department of Industrial and Management Systems Engineering, West Virginia University, Morgantown, WV 26506, USA · Department of Computer Science & Engineering, Stamford University Bangladesh, Dhaka-1217, Bangladesh +1
Simulating Eating Disorder Patients with LLMs: Evaluating Psychological Persona Stability in Multi-Turn Conversations
May 12, 2026Jennifer Haase, Jana Gonnermann-Müller, See Heng Yim +3Persona Consistency Personality Modeling in Language Models
Weizenbaum Institute, Berlin, Germany · HU Berlin, Berlin, Germany · Zuse Institute Berlin, Berlin, Germany +1
Food Noise & False Safety: A Systematic Evaluation of How LLMs Fail to Adapt to Eating Disorder Queries with Clinician Feedback
Jun 1, 2026Giulia Pucci, Emily Hemendinger, Ruizhe Li +3Mental Health LLM Safety
University of Aberdeen · University of Colorado Anschutz · Heriot-Watt University +1

Jul 18, 2026J/K move · Enter open · S save

Investigation of Polycystic Ovary Syndrome (PCOS) Diagnosis Using Machine Learning Approaches

Al Zadid Sultan Bin Habib, Md Asif Bin Syed, Md. Ekramul Islam, Tanpia Tasnim

Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV 26506, USA · Department of Industrial and Management Systems Engineering, West Virginia University, Morgantown, WV 26506, USA · Department of Computer Science & Engineering, Stamford University Bangladesh, Dhaka-1217, Bangladesh+1

Medical Diagnosis Clinical Prediction

Open paper AskPDF arXiv

Polycystic Ovarian Syndrome (PCOS) is a widespread hormone problem for women of childbearing age. Women with PCOS may not ovulate; they might have high levels of androgens and have many small cysts on the ovaries. It can cause missed or irregular menstrual periods, excess hair growth, acne, infertility, and weight gain. Machine Learning (ML) can effectively diagnose this disease at an earlier stage as tons of medical data are available now. Traditional approaches to detect PCOS encompass a combination of clinical evaluation, medical history assessment, physical examination, and laboratory tests. These approaches aim to identify the characteristic symptoms and hormonal imbalances associated with PCOS. Physical examination requires good resources and costs time and money. In recent times, data-driven techniques have substantially advanced disease prediction within the medical field. We aim to utilize ML approaches, incorporating unique feature selection algorithms, to predict PCOS. This paper introduces a data-driven approach to PCOS diagnosis, combining Feature Engineering and ML. Several feature selection approaches have been considered to select sets of features for training the ML model, including CatBoost, Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LGBM), AdaBoost, Random Forest (RF). Results demonstrate that AdaBoost, with ten features selected by RF Feature Importance and Highest Correlation (HC), provides the highest test accuracy.

Abstract

Clinical NLP

Explore similar work

Investigation of Polycystic Ovary Syndrome (PCOS) Diagnosis Using Machine Learning Approaches

Simulating Eating Disorder Patients with LLMs: Evaluating Psychological Persona Stability in Multi-Turn Conversations

Food Noise & False Safety: A Systematic Evaluation of How LLMs Fail to Adapt to Eating Disorder Queries with Clinician Feedback