Machine learning classification meets migraine: recommendations for study evaluation

Abstract The integration of machine learning (ML) classification techniques into migraine research has offered new insights into the pathophysiology and classification of migraine types and subtypes. However, inconsistencies in study design, lack of methodological transparency, and the absence of ex...

Full description

Saved in:
Bibliographic Details
Main Authors: Igor Petrušić, Andrej Savić, Katarina Mitrović, Nebojša Bačanin, Gabriele Sebastianelli, Daniele Secci, Gianluca Coppola
Format: Article
Language:English
Published: BMC 2024-12-01
Series:The Journal of Headache and Pain
Subjects:
Online Access:https://doi.org/10.1186/s10194-024-01924-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846112384227213312
author Igor Petrušić
Andrej Savić
Katarina Mitrović
Nebojša Bačanin
Gabriele Sebastianelli
Daniele Secci
Gianluca Coppola
author_facet Igor Petrušić
Andrej Savić
Katarina Mitrović
Nebojša Bačanin
Gabriele Sebastianelli
Daniele Secci
Gianluca Coppola
author_sort Igor Petrušić
collection DOAJ
description Abstract The integration of machine learning (ML) classification techniques into migraine research has offered new insights into the pathophysiology and classification of migraine types and subtypes. However, inconsistencies in study design, lack of methodological transparency, and the absence of external validation limit the impact and reproducibility of such studies. This paper presents a framework of six essential recommendations for evaluating ML-based classification in migraine research: (1) group homogenization by clinical phenotype, attack frequency, comorbidity, therapy, and demographics; (2) defining adequate sample size; (3) quality control of collected and preprocessed data; (4) transparent training, testing, and performance evaluation of ML models, including strategies for data splitting, overfitting control, and feature selection; (5) interpretability of results with clinical relevance; and (6) open data and code sharing to facilitate reproducibility. These recommendations aim to balance the trade-off between model generalization and precision while encouraging collaborative standardization across the ML and headache communities. Furthermore, this framework intends to stimulate discussion toward forming a consortium to establish definitive guidelines for ML-based classification research in migraine field.
format Article
id doaj-art-f93ed33d770042b1939107d17f92b0f1
institution Kabale University
issn 1129-2377
language English
publishDate 2024-12-01
publisher BMC
record_format Article
series The Journal of Headache and Pain
spelling doaj-art-f93ed33d770042b1939107d17f92b0f12024-12-22T12:38:00ZengBMCThe Journal of Headache and Pain1129-23772024-12-012511710.1186/s10194-024-01924-xMachine learning classification meets migraine: recommendations for study evaluationIgor Petrušić0Andrej Savić1Katarina Mitrović2Nebojša Bačanin3Gabriele Sebastianelli4Daniele Secci5Gianluca Coppola6Laboratory for Advanced Analysis of Neuroimages, Faculty of Physical Chemistry, University of BelgradeScience and Research Centre, School of Electrical Engineering, University of Belgrade, University of BelgradeDepartment of Information Technologies, Faculty of Technical Sciences Čačak, University of KragujevacDepartment of Informatics and Computing, Singidunum UniversityDepartment of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome Polo Pontino ICOTDepartment of Engineering and Architecture, University of ParmaDepartment of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome Polo Pontino ICOTAbstract The integration of machine learning (ML) classification techniques into migraine research has offered new insights into the pathophysiology and classification of migraine types and subtypes. However, inconsistencies in study design, lack of methodological transparency, and the absence of external validation limit the impact and reproducibility of such studies. This paper presents a framework of six essential recommendations for evaluating ML-based classification in migraine research: (1) group homogenization by clinical phenotype, attack frequency, comorbidity, therapy, and demographics; (2) defining adequate sample size; (3) quality control of collected and preprocessed data; (4) transparent training, testing, and performance evaluation of ML models, including strategies for data splitting, overfitting control, and feature selection; (5) interpretability of results with clinical relevance; and (6) open data and code sharing to facilitate reproducibility. These recommendations aim to balance the trade-off between model generalization and precision while encouraging collaborative standardization across the ML and headache communities. Furthermore, this framework intends to stimulate discussion toward forming a consortium to establish definitive guidelines for ML-based classification research in migraine field.https://doi.org/10.1186/s10194-024-01924-xBenchmarkMachine learning classification modelsData qualityModel interpretabilityModel reproducibilityMigraine types
spellingShingle Igor Petrušić
Andrej Savić
Katarina Mitrović
Nebojša Bačanin
Gabriele Sebastianelli
Daniele Secci
Gianluca Coppola
Machine learning classification meets migraine: recommendations for study evaluation
The Journal of Headache and Pain
Benchmark
Machine learning classification models
Data quality
Model interpretability
Model reproducibility
Migraine types
title Machine learning classification meets migraine: recommendations for study evaluation
title_full Machine learning classification meets migraine: recommendations for study evaluation
title_fullStr Machine learning classification meets migraine: recommendations for study evaluation
title_full_unstemmed Machine learning classification meets migraine: recommendations for study evaluation
title_short Machine learning classification meets migraine: recommendations for study evaluation
title_sort machine learning classification meets migraine recommendations for study evaluation
topic Benchmark
Machine learning classification models
Data quality
Model interpretability
Model reproducibility
Migraine types
url https://doi.org/10.1186/s10194-024-01924-x
work_keys_str_mv AT igorpetrusic machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation
AT andrejsavic machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation
AT katarinamitrovic machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation
AT nebojsabacanin machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation
AT gabrielesebastianelli machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation
AT danielesecci machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation
AT gianlucacoppola machinelearningclassificationmeetsmigrainerecommendationsforstudyevaluation