Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation

Image segmentation plays a central role in computer vision applications such as medical imaging, industrial inspection, and environmental monitoring. However, evaluating segmentation performance can be particularly challenging when ground truth is not clearly defined, as is often the case in tasks i...

Full description

Saved in:
Bibliographic Details
Main Authors: Andrés Larroza, Francisco Javier Pérez-Benito, Raquel Tendero, Juan Carlos Perez-Cortes, Marta Román, Rafael Llobet
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Journal of Imaging
Subjects:
Online Access:https://www.mdpi.com/2313-433X/11/5/170
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849710854104154112
author Andrés Larroza
Francisco Javier Pérez-Benito
Raquel Tendero
Juan Carlos Perez-Cortes
Marta Román
Rafael Llobet
author_facet Andrés Larroza
Francisco Javier Pérez-Benito
Raquel Tendero
Juan Carlos Perez-Cortes
Marta Román
Rafael Llobet
author_sort Andrés Larroza
collection DOAJ
description Image segmentation plays a central role in computer vision applications such as medical imaging, industrial inspection, and environmental monitoring. However, evaluating segmentation performance can be particularly challenging when ground truth is not clearly defined, as is often the case in tasks involving subjective interpretation. These challenges are amplified by inter- and intra-observer variability, which complicates the use of human annotations as a reliable reference. To address this, we propose a novel validation framework—referred to as the three-blind validation strategy—that enables rigorous assessment of segmentation models in contexts where subjectivity and label variability are significant. The core idea is to have a third independent expert, blind to the labeler identities, assess a shuffled set of segmentations produced by multiple human annotators and/or automated models. This allows for the unbiased evaluation of model performance and helps uncover patterns of disagreement that may indicate systematic issues with either human or machine annotations. The primary objective of this study is to introduce and demonstrate this validation strategy as a generalizable framework for robust model evaluation in subjective segmentation tasks. We illustrate its practical implementation in a mammography use case involving dense tissue segmentation while emphasizing its potential applicability to a broad range of segmentation scenarios.
format Article
id doaj-art-b93ee28b0fdb431da26d405b39e2f355
institution DOAJ
issn 2313-433X
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Journal of Imaging
spelling doaj-art-b93ee28b0fdb431da26d405b39e2f3552025-08-20T03:14:46ZengMDPI AGJournal of Imaging2313-433X2025-05-0111517010.3390/jimaging11050170Three-Blind Validation Strategy of Deep Learning Models for Image SegmentationAndrés Larroza0Francisco Javier Pérez-Benito1Raquel Tendero2Juan Carlos Perez-Cortes3Marta Román4Rafael Llobet5Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, s/n, 46022 València, SpainInstituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, s/n, 46022 València, SpainInstituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, s/n, 46022 València, SpainInstituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, s/n, 46022 València, SpainDepartment of Epidemiology and Evaluation, IMIM (Hospital del Mar Research Institute), Passeig Marítim 25-29, 08003 Barcelona, SpainInstituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, s/n, 46022 València, SpainImage segmentation plays a central role in computer vision applications such as medical imaging, industrial inspection, and environmental monitoring. However, evaluating segmentation performance can be particularly challenging when ground truth is not clearly defined, as is often the case in tasks involving subjective interpretation. These challenges are amplified by inter- and intra-observer variability, which complicates the use of human annotations as a reliable reference. To address this, we propose a novel validation framework—referred to as the three-blind validation strategy—that enables rigorous assessment of segmentation models in contexts where subjectivity and label variability are significant. The core idea is to have a third independent expert, blind to the labeler identities, assess a shuffled set of segmentations produced by multiple human annotators and/or automated models. This allows for the unbiased evaluation of model performance and helps uncover patterns of disagreement that may indicate systematic issues with either human or machine annotations. The primary objective of this study is to introduce and demonstrate this validation strategy as a generalizable framework for robust model evaluation in subjective segmentation tasks. We illustrate its practical implementation in a mammography use case involving dense tissue segmentation while emphasizing its potential applicability to a broad range of segmentation scenarios.https://www.mdpi.com/2313-433X/11/5/170deep learningimage segmentationmammography
spellingShingle Andrés Larroza
Francisco Javier Pérez-Benito
Raquel Tendero
Juan Carlos Perez-Cortes
Marta Román
Rafael Llobet
Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation
Journal of Imaging
deep learning
image segmentation
mammography
title Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation
title_full Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation
title_fullStr Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation
title_full_unstemmed Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation
title_short Three-Blind Validation Strategy of Deep Learning Models for Image Segmentation
title_sort three blind validation strategy of deep learning models for image segmentation
topic deep learning
image segmentation
mammography
url https://www.mdpi.com/2313-433X/11/5/170
work_keys_str_mv AT andreslarroza threeblindvalidationstrategyofdeeplearningmodelsforimagesegmentation
AT franciscojavierperezbenito threeblindvalidationstrategyofdeeplearningmodelsforimagesegmentation
AT raqueltendero threeblindvalidationstrategyofdeeplearningmodelsforimagesegmentation
AT juancarlosperezcortes threeblindvalidationstrategyofdeeplearningmodelsforimagesegmentation
AT martaroman threeblindvalidationstrategyofdeeplearningmodelsforimagesegmentation
AT rafaelllobet threeblindvalidationstrategyofdeeplearningmodelsforimagesegmentation