Work

Which Uncertainty Quantification Method in Deep Learning Models for Knee Osteoarthritis Diagnosis?

Year: 2025

Type: conference-paper

Abstract: This article focuses on uncertainty quantification (UQ) in deep learning methods for knee osteoarthritis (OA) diagnosis. Knee OA is one of the leading causes of physical disability worldwide, with a high personal and socio-economic burden. However, the detection of knee OA relies on visual assessment, which is highly subjective. Recently, deep learning has led to remarkable improvements in the detection of knee OA. However, the deep learning methods proposed in the current literature seem to be more concerned with model performance than with UQ.In this paper, we explore several UQ methods that aim to complement existing deep learning approaches. In particular, we consider conformal prediction, Monte Carlo dropout and evidential deep learning. The advantages and disadvantages of each method are highlighted in the context of the particular application. The aim is to provide a simple means of expressing uncertainty while allowing greater model transparency for clinical users.

DOI: 10.1109/ssp64130.2025.11073440

Authors: Mame Diarra Fall, Elhadji C. Faye

Institutions: Centre National de la Recherche Scientifique, Institut Denis Poisson

Language: English

Topic: Artificial Intelligence in Healthcare and Education

Subfield: Health Informatics

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: green

Funder: Agence Nationale de la Recherche

Awards: AI.iO Artificial Intelligence in Orléans: Learning from heterogeneous data and expert knowledge. Applications in geological and environmental sciences, BAyesian nonparametrics, Complex models and Kernels, Uncertainty quantification and deeP methods, Machine learnIng and Multimodal imaging for knee OSteoArthritis prediction