Active Learning Strategies on a Real-World Thyroid Ultrasound Dataset - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Active Learning Strategies on a Real-World Thyroid Ultrasound Dataset

Abstract

Machine learning applications in ultrasound imaging are limited by access to ground-truth expert annotations, especially in specialized applications such as thyroid nodule evaluation. Active learning strategies seek to alleviate this concern by making more effective use of expert annotations; however, many proposed techniques do not adapt well to small-scale (i.e. a few hundred images) datasets. In this work, we test active learning strategies including an uncertainty-weighted selection approach with supervised and semi-supervised learning to evaluate the effectiveness of these tools for the prediction of nodule presence on a clinical ultrasound dataset. The results on this as well as two other medical image datasets suggest that even successful active learning strategies have limited clinical significance in terms of reducing annotation burden.

Domains

Medical Imaging
Fichier principal
Vignette du fichier
manuscript.pdf (911.68 Ko) Télécharger le fichier
supplementary_material.pdf (367.3 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Licence : CC BY - Attribution
Origin : Files produced by the author(s)
Licence : CC BY - Attribution

Dates and versions

hal-04209622 , version 1 (18-09-2023)

Licence

Attribution

Identifiers

  • HAL Id : hal-04209622 , version 1

Cite

Hari Sreedhar, Guillaume P R Lajoinie, Charles Raffaelli, Hervé Delingette. Active Learning Strategies on a Real-World Thyroid Ultrasound Dataset. DALI 2023 - Data Augmentation, Labelling, and Imperfections / MICCAI Workshop 2023, Oct 2023, Vancouver, Canada. ⟨hal-04209622⟩
53 View
54 Download

Share

Gmail Facebook X LinkedIn More