Saved in:
Bibliographic Details
Main Author: Lee, EunYoung
Format: Recurso digital
Language:
Published: Zenodo 2025
Online Access:https://doi.org/10.5281/zenodo.18013689
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • <p>Pain remains inadequately represented in both clinical practice and artificial intelligence <br>systems, largely due to its reduction to unidimensional numeric scales. This study proposes <br>the Pain Kernelas a computational–phenomenological measurement frameworkfor pain <br>representationthat integrates clinical phenomenology, narrative structure, and cognitive <br>encodingprocesses. Grounded in clinical observations and phenomenological analysis, the <br>Pain Kernel conceptualizes pain as an organized, interpretable kernel composed of distinct <br>domains rather than a single intensity value.<br>By reframing pain as a structured and encodable human experience, this framework <br>addresses critical limitations of current assessment tools and enables human-centered AI <br>approaches to pain modeling. The proposed structure supports interpretability, preserves <br>subjective meaning, and allows for future quantification without collapsing experiential <br>complexity. Importantly, the Pain Kernel provides a translational bridge between clinical <br>understanding and computational modeling, offering a foundation for explainable and <br>ethically aligned health AI systems. This work contributes a novel pathway for integrating <br>subjective human experience into AI-ready representations while maintaining clinical validity <br>and phenomenological depth.<br>Keywords<br>Pain Kernel<br>Pain Measurement<br>Pain Representation<br>Human-Centered AI<br>Cognitive Encoding<br>Clinical Phenomenology<br>Measurement Framework<br>Narrative Medicine<br>Explainable AI<br>Health AI<br>Cognitive Framework<br>Human AI Interface<br><br></p>