RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability

<b>Background/Objectives:</b> To simplify the decision-making process in radiomics by employing RadiomiX, an algorithm designed to automatically identify the best model combination and validate them across multiple environments was developed, thus enhancing the reliability of results. &l...

Full description

Saved in:
Bibliographic Details
Main Authors: Harel Kotler, Luca Bergamin, Fabio Aiolli, Elena Scagliori, Angela Grassi, Giulia Pasello, Alessandra Ferro, Francesca Caumo, Gisella Gennaro
Format: Article
Language:English
Published: MDPI AG 2025-08-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/15/1968
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849239751396163584
author Harel Kotler
Luca Bergamin
Fabio Aiolli
Elena Scagliori
Angela Grassi
Giulia Pasello
Alessandra Ferro
Francesca Caumo
Gisella Gennaro
author_facet Harel Kotler
Luca Bergamin
Fabio Aiolli
Elena Scagliori
Angela Grassi
Giulia Pasello
Alessandra Ferro
Francesca Caumo
Gisella Gennaro
author_sort Harel Kotler
collection DOAJ
description <b>Background/Objectives:</b> To simplify the decision-making process in radiomics by employing RadiomiX, an algorithm designed to automatically identify the best model combination and validate them across multiple environments was developed, thus enhancing the reliability of results. <b>Methods</b>: RadiomiX systematically tests classifier and feature selection method combinations known to be suitable for radiomic datasets to determine the best-performing configuration across multiple train–test splits and K-fold cross-validation. The framework was validated on four public retrospective radiomics datasets including lung nodules, metastatic breast cancer, and hepatic encephalopathy using CT, PET/CT, and MRI modalities. Model performance was assessed using the area under the receiver-operating-characteristic curve (AUC) and accuracy metrics. <b>Results:</b> RadiomiX achieved superior performance across four datasets: LLN (AUC = 0.850 and accuracy = 0.785), SLN (AUC = 0.845 and accuracy = 0.754), MBC (AUC = 0.889 and accuracy = 0.833), and CHE (AUC = 0.837 and accuracy = 0.730), significantly outperforming original published models (<i>p</i> < 0.001 for LLN/SLN and <i>p</i> = 0.023 for MBC accuracy). When original published models were re-evaluated using ten-fold cross-validation, their performance decreased substantially: LLN (AUC = 0.783 and accuracy = 0.731), SLN (AUC = 0.748 and accuracy = 0.714), MBC (AUC = 0.764 and accuracy = 0.711), and CHE (AUC = 0.755 and accuracy = 0.677), further highlighting RadiomiX’s methodological advantages. <b>Conclusions:</b> Systematically testing model combinations using RadiomiX has led to significant improvements in performance. This emphasizes the potential of automated ML as a step towards better-performing and more reliable radiomic models.
format Article
id doaj-art-b6b59a7f2bb44e06a83f0ecf92af4ed6
institution Kabale University
issn 2075-4418
language English
publishDate 2025-08-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-b6b59a7f2bb44e06a83f0ecf92af4ed62025-08-20T04:00:50ZengMDPI AGDiagnostics2075-44182025-08-011515196810.3390/diagnostics15151968RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in ReplicabilityHarel Kotler0Luca Bergamin1Fabio Aiolli2Elena Scagliori3Angela Grassi4Giulia Pasello5Alessandra Ferro6Francesca Caumo7Gisella Gennaro8Breast Radiology, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, ItalyDepartment of Mathematics, University of Padova, 35128 Padua, ItalyDepartment of Mathematics, University of Padova, 35128 Padua, ItalyRadiology, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, ItalyClinical Research Unit, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, ItalyMedical Oncology 2, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, ItalyMedical Oncology 2, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, ItalyBreast Radiology, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, ItalyBreast Radiology, Veneto Institute of Oncology IOV-IRCCS, 35128 Padua, Italy<b>Background/Objectives:</b> To simplify the decision-making process in radiomics by employing RadiomiX, an algorithm designed to automatically identify the best model combination and validate them across multiple environments was developed, thus enhancing the reliability of results. <b>Methods</b>: RadiomiX systematically tests classifier and feature selection method combinations known to be suitable for radiomic datasets to determine the best-performing configuration across multiple train–test splits and K-fold cross-validation. The framework was validated on four public retrospective radiomics datasets including lung nodules, metastatic breast cancer, and hepatic encephalopathy using CT, PET/CT, and MRI modalities. Model performance was assessed using the area under the receiver-operating-characteristic curve (AUC) and accuracy metrics. <b>Results:</b> RadiomiX achieved superior performance across four datasets: LLN (AUC = 0.850 and accuracy = 0.785), SLN (AUC = 0.845 and accuracy = 0.754), MBC (AUC = 0.889 and accuracy = 0.833), and CHE (AUC = 0.837 and accuracy = 0.730), significantly outperforming original published models (<i>p</i> < 0.001 for LLN/SLN and <i>p</i> = 0.023 for MBC accuracy). When original published models were re-evaluated using ten-fold cross-validation, their performance decreased substantially: LLN (AUC = 0.783 and accuracy = 0.731), SLN (AUC = 0.748 and accuracy = 0.714), MBC (AUC = 0.764 and accuracy = 0.711), and CHE (AUC = 0.755 and accuracy = 0.677), further highlighting RadiomiX’s methodological advantages. <b>Conclusions:</b> Systematically testing model combinations using RadiomiX has led to significant improvements in performance. This emphasizes the potential of automated ML as a step towards better-performing and more reliable radiomic models.https://www.mdpi.com/2075-4418/15/15/1968radiomicslung cancerbreast cancerhepatic encephalopathymachine learningreplicability
spellingShingle Harel Kotler
Luca Bergamin
Fabio Aiolli
Elena Scagliori
Angela Grassi
Giulia Pasello
Alessandra Ferro
Francesca Caumo
Gisella Gennaro
RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability
Diagnostics
radiomics
lung cancer
breast cancer
hepatic encephalopathy
machine learning
replicability
title RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability
title_full RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability
title_fullStr RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability
title_full_unstemmed RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability
title_short RadiomiX for Radiomics Analysis: Automated Approaches to Overcome Challenges in Replicability
title_sort radiomix for radiomics analysis automated approaches to overcome challenges in replicability
topic radiomics
lung cancer
breast cancer
hepatic encephalopathy
machine learning
replicability
url https://www.mdpi.com/2075-4418/15/15/1968
work_keys_str_mv AT harelkotler radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT lucabergamin radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT fabioaiolli radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT elenascagliori radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT angelagrassi radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT giuliapasello radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT alessandraferro radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT francescacaumo radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability
AT gisellagennaro radiomixforradiomicsanalysisautomatedapproachestoovercomechallengesinreplicability