Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network

<b>Background/Objectives:</b> Clinical imaging techniques, such as magnetic resonance imaging and computed tomography, play a vital role in supporting clinicians by aiding disease diagnosis and facilitating the planning of appropriate interventions. This is particularly important in mali...

Full description

Saved in:
Bibliographic Details
Main Authors: Maria Chiara Brunese, Aldo Rocca, Antonella Santone, Mario Cesarelli, Luca Brunese, Francesco Mercaldo
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/7/878
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849730659511173120
author Maria Chiara Brunese
Aldo Rocca
Antonella Santone
Mario Cesarelli
Luca Brunese
Francesco Mercaldo
author_facet Maria Chiara Brunese
Aldo Rocca
Antonella Santone
Mario Cesarelli
Luca Brunese
Francesco Mercaldo
author_sort Maria Chiara Brunese
collection DOAJ
description <b>Background/Objectives:</b> Clinical imaging techniques, such as magnetic resonance imaging and computed tomography, play a vital role in supporting clinicians by aiding disease diagnosis and facilitating the planning of appropriate interventions. This is particularly important in malignant conditions like hepatocellular carcinoma, where accurate image segmentation, delineating the liver and tumor, is a critical initial step in optimizing diagnosis, staging, and treatment planning, including interventions like transplantation, surgical resection, radiotherapy, portal vein embolization, and other procedures. Therefore, effective segmentation methods can significantly influence both diagnostic accuracy and treatment outcomes. <b>Method:</b> In this paper, we propose a deep learning-based approach aimed at accurately segmenting the liver in medical images, thus addressing a critical need in hepatic disease diagnosis and treatment planning. We consider a U-Net architecture with residual connections to capture fine-grained anatomical details. We also take into account the prediction explainability, aiming to highlight, in the image under analysis, the areas that are symptomatic for a certain segmentation. In detail, by exploiting the U-Net architecture, two different models are trained with two annotated datasets of computed tomography medical images, resulting in four different experiments. <b>Results:</b> We consider two different datasets to improve robustness and generalization across diverse patient populations and imaging conditions. Experimental results demonstrate that the proposed method obtains interesting performances, with an accuracy ranging from 0.81 to 0.93. <b>Conclusions:</b> We thus show that the proposed method can provide a reliable and efficient solution for automated liver segmentation, promising significant advancements in clinical workflows and precision medicine.
format Article
id doaj-art-8b1e2cfb0634472eadbd302993342949
institution DOAJ
issn 2075-4418
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-8b1e2cfb0634472eadbd3029933429492025-08-20T03:08:47ZengMDPI AGDiagnostics2075-44182025-03-0115787810.3390/diagnostics15070878Explainable and Robust Deep Learning for Liver Segmentation Through U-Net NetworkMaria Chiara Brunese0Aldo Rocca1Antonella Santone2Mario Cesarelli3Luca Brunese4Francesco Mercaldo5Department of Medicine and Health Sciences “Vincenzo Tiberio”, University of Molise, 86100 Campobasso, ItalyDepartment of Medicine and Health Sciences “Vincenzo Tiberio”, University of Molise, 86100 Campobasso, ItalyDepartment of Medicine and Health Sciences “Vincenzo Tiberio”, University of Molise, 86100 Campobasso, ItalyDepartment of Engineering, University of Sannio, 82100 Benevento, ItalyDepartment of Medicine and Health Sciences “Vincenzo Tiberio”, University of Molise, 86100 Campobasso, ItalyDepartment of Medicine and Health Sciences “Vincenzo Tiberio”, University of Molise, 86100 Campobasso, Italy<b>Background/Objectives:</b> Clinical imaging techniques, such as magnetic resonance imaging and computed tomography, play a vital role in supporting clinicians by aiding disease diagnosis and facilitating the planning of appropriate interventions. This is particularly important in malignant conditions like hepatocellular carcinoma, where accurate image segmentation, delineating the liver and tumor, is a critical initial step in optimizing diagnosis, staging, and treatment planning, including interventions like transplantation, surgical resection, radiotherapy, portal vein embolization, and other procedures. Therefore, effective segmentation methods can significantly influence both diagnostic accuracy and treatment outcomes. <b>Method:</b> In this paper, we propose a deep learning-based approach aimed at accurately segmenting the liver in medical images, thus addressing a critical need in hepatic disease diagnosis and treatment planning. We consider a U-Net architecture with residual connections to capture fine-grained anatomical details. We also take into account the prediction explainability, aiming to highlight, in the image under analysis, the areas that are symptomatic for a certain segmentation. In detail, by exploiting the U-Net architecture, two different models are trained with two annotated datasets of computed tomography medical images, resulting in four different experiments. <b>Results:</b> We consider two different datasets to improve robustness and generalization across diverse patient populations and imaging conditions. Experimental results demonstrate that the proposed method obtains interesting performances, with an accuracy ranging from 0.81 to 0.93. <b>Conclusions:</b> We thus show that the proposed method can provide a reliable and efficient solution for automated liver segmentation, promising significant advancements in clinical workflows and precision medicine.https://www.mdpi.com/2075-4418/15/7/878liverU-Netsegmentationbiomedical imagesbioimagesdeep learning
spellingShingle Maria Chiara Brunese
Aldo Rocca
Antonella Santone
Mario Cesarelli
Luca Brunese
Francesco Mercaldo
Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network
Diagnostics
liver
U-Net
segmentation
biomedical images
bioimages
deep learning
title Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network
title_full Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network
title_fullStr Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network
title_full_unstemmed Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network
title_short Explainable and Robust Deep Learning for Liver Segmentation Through U-Net Network
title_sort explainable and robust deep learning for liver segmentation through u net network
topic liver
U-Net
segmentation
biomedical images
bioimages
deep learning
url https://www.mdpi.com/2075-4418/15/7/878
work_keys_str_mv AT mariachiarabrunese explainableandrobustdeeplearningforliversegmentationthroughunetnetwork
AT aldorocca explainableandrobustdeeplearningforliversegmentationthroughunetnetwork
AT antonellasantone explainableandrobustdeeplearningforliversegmentationthroughunetnetwork
AT mariocesarelli explainableandrobustdeeplearningforliversegmentationthroughunetnetwork
AT lucabrunese explainableandrobustdeeplearningforliversegmentationthroughunetnetwork
AT francescomercaldo explainableandrobustdeeplearningforliversegmentationthroughunetnetwork