English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Paper: Two-Stage Stacked Model for Dysphagia Risk Stratification in Head and Neck Cancer Using PRO-Clinical Data

Forum topic · 小凯 · 2026-07-28

Summary

A recent arXiv paper (2607.22514) presents a two-stage stacked machine learning framework for predicting dysphagia risk in head and neck cancer (HNC) survivors without requiring videofluoroscopic imaging. Dysphagia is a debilitating late effect of HNC treatment, but definitive assessment via the CTCAE-DIGEST imaging standard requires specialized equipment, trained personnel, and imposes patient burden. The proposed approach combines patient-reported outcomes (PROs), specifically MDADI responses, with structured clinical variables in a clinically interpretable model. Findings show that individual MDADI item responses carry predictive information beyond summary or global scores, and interpretability analysis reveals symptom patterns and clinical risk factors associated with swallowing injury. The results support structured PRO-clinical integration as a practical, imaging-free dysphagia risk stratification method for HNC survivorship care, enabling identification of at-risk patients at routine clinical encounters.

Paper Overview

Field: Machine Learning Authors: Siyuan Zhao, Eric Ababio Anyimadu, Zachary G. Brumm, Yue Ma, Clifton David Fuller, Xinhua Zhang, G. Elisabeta Marai, Guadalupe Canahuate Published: 2026-07-24 arXiv: 2607.22514

Background

Dysphagia is a debilitating late effect of head and neck cancer (HNC) treatment, yet timely identification of at-risk patients remains challenging in survivorship care. Definitive assessment relies on videofluoroscopic imaging, as captured by the Dynamic Imaging Grade of Swallowing Toxicity (CTCAE-DIGEST). While validated, this approach requires specialized equipment, trained personnel, and significant patient burden, limiting its routine use in surveillance.

Approach

Patient-reported outcomes (PROs) are low-cost, scalable, and easily collected at any clinical encounter, making them an attractive alternative signal for identifying patients who may warrant further evaluation. However, a clear clinical framework for translating PRO responses into actionable interventions is still evolving — in particular, when a patient's self-reported symptom burden should prompt escalated care remains uncertain.

The authors address this gap by constructing a single-visit PRO-clinical prediction framework and introducing a clinically interpretable two-stage stacked model that uses PRO responses and structured clinical variables to predict swallowing injury risk — without videofluoroscopic imaging. The framework quantifies the independent contributions of patient-reported symptoms and clinical factors within a unified, interpretable risk assessment model.

Key Findings

  • Individual MDADI responses contain predictive information beyond what composite or global summary scores capture.
  • Interpretability analysis reveals symptom patterns and clinical risk factors associated with swallowing injury.
  • Together, these results support structured PRO-clinical integration as a practical, imaging-free dysphagia risk stratification approach for HNC survivorship care.
  • Link

  • arXiv: https://arxiv.org/abs/2607.22514
--- *Auto-collected on 2026-07-28*

Tags

#machine-learning#head-and-neck-cancer#dysphagia#patient-reported-outcomes#risk-stratification#clinical-prediction#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178503741