[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121105-en":3,"doc-seo-121105-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121105,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Efficacy of MRI data harmonization in the age of machine learning - a multicenter study across 36 datasets","Pooling publicly available MRI data across multiple sites enables large, diverse cohorts, increases statistical power, and supports data reuse with machine learning. Multicenter harmonization is used to reduce confounding from non-biological variability, yet applying harmonization to the full dataset before training can cause data leakage and inflate model performance. This work measures harmonization efficacy and proposes a harmonizer transformer implementing ComBat within an ML preprocessing pipeline to avoid leakage. Using 36-site T1-weighted MRI from 1740 healthy subjects, site effects were reduced and leakage impact in age prediction was demonstrated.","[www. nature.com/scientificdata](www. nature.com/scientificdata)  \nOPEN  \nAnAlySIS  \nEfficacy of MRI data harmonization in the age of machine learning: a multicenter study across 36 datasets  \nChiara Marzi1,2, Marco Giannelli3, Andrea Barucci2, Carlo Tessa4, Mario Mascalchi5,6 & Stefano Diciotti7,8 ✉  \nPooling publicly-available MRI data from multiple sites allows to assemble extensive groups of subjects, increase statistical power, and promote data reuse with machine learning techniques. The harmonization of multicenter data is necessary to reduce the confounding effect associated with nonbiological sources of variability in the data. However, when applied to the entire dataset before machine learning, the harmonization leads to data leakage, because information outside the training set may affect model building, and potentially falsely overestimate performance. We propose a 1) measurement of the efficacy of data harmonization; 2) harmonizer transformer, i.e., an implementation of the ComBat harmonization allowing its encapsulation among the preprocessing steps of a machine learning pipeline, avoiding data leakage by design. We tested these tools using brain T1-weighted MRI data from 1740 healthy subjects acquired at 36 sites. After harmonization, the site effect was removed or reduced, and we showed the data leakage effect in predicting individual age from MRI data, highlighting that introducing the harmonizer transformer into a machine learning pipeline allows for avoiding data leakage by design.  \nIntroduction  \nIn recent years there has been an increasing trend toward data sharing in neuroimaging research communities, leading to a rising number of public neuroimaging databases and collaborative multicenter initiatives 1–4. Indeed, pooling MRI data from multiple sites provides an opportunity to assemble more extensive and diverse groups of subjects2,3,5,6, increase statistical power3,7–10, and study rare disorders and subtle effects11, 12. However, a major drawback of combining neuroimaging data across sites is the introduction of confounding effects due to non-biological variability in the data, typically related to image acquisition hardware and protocol. Indeed, properties of MRI such as scanner field strength, radiofrequency coil type, gradients coil characteristics, hardware, image reconstruction algorithm, and non-standardized acquisition protocol parameters can introduce unwanted technical variability, also reflected in MRI-derived features13–15.  \nThe harmonization of multicenter data, defined as applying mathematical and statistical concepts to reduce unwanted site variability while maintaining the biological content, is, therefore, necessary to ensure the success of cooperative analyses. Currently, among the harmonization methods for tabular data available to the neuroimaging scientific community, ComBat is one of the most widely used7, 12, 16–34. The ComBat model was first introduced in gene expression analysis as a batch-effect correction tool to remove unwanted variation associated  \n1Department of Statistics, Computer Science and Applications “Giuseppe Parenti”, University of Florence, 50134, florence, italy. 2“Nello Carrara” Institute of Applied Physics (IFAC), National Research Council (CNR), 50019, Sesto fiorentino, florence, italy. 3Unit of Medical Physics, Pisa University Hospital “Azienda Ospedaliero-Universitaria Pisana”, 56126, Pisa, Italy. 4Radiology Unit Apuane e Lunigiana, Azienda USL Toscana Nord Ovest, 54100, Massa, italy. 5Department of Experimental and Clinical Biomedical Sciences “Mario Serio”, University of Florence, 50139, florence, italy. 6 Division of epidemiology and clinical Governance, institute for Study, Prevention and netwoRk in Oncology (ISPRO), 50139, Florence, Italy. 7 Department of electrical, electronic, and information engineering“Guglielmo Marconi” -DEI, University of Bologna, 47522, Cesena, Italy. 8Alma Mater Research Institute for HumanCentered Artificial Intelligence, Unive","cbCaij65UM28mlTs","https://ap.wps.com/l/cbCaij65UM28mlTs","pdf",6361838,1,27,"English","en",105,"# Introduction\n## Data sharing and pooling in neuroimaging\n## Non-biological variability and need for harmonization\n## ComBat and current focus of prior work\n# Proposed approach and rationale\n## Measuring harmonization efficacy\n## Harmonizer transformer for ML pipelines\n## Data leakage when harmonizing before splitting\n# Study setup and evaluation\n## Dataset and imaging modality\n## Effects on site variability and age prediction","[{\"question\":\"Why is MRI data harmonization necessary in multicenter machine learning studies?\",\"answer\":\"To reduce confounding variability introduced by non-biological differences across sites, such as scanner and acquisition-related factors, while preserving biological signal.\"},{\"question\":\"How does harmonizing the entire dataset before training lead to data leakage?\",\"answer\":\"When harmonization uses information from outside the training set, that information can influence model building, which can falsely overestimate predictive performance.\"},{\"question\":\"What is the harmonizer transformer and how does it prevent data leakage?\",\"answer\":\"It wraps ComBat harmonization inside the preprocessing steps of an ML pipeline so harmonization is applied within the training framework design, avoiding leakage by construction.\"}]","Efficacy of MRI data harmonization in the age of machine learning - a multicenter study across 36 datasets | PDF",1785733754,68,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"efficacy-of-mri-data-harmonization-in-the-age-of-machine-learning-a-multicenter-study-across-36-datasets","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/efficacy-of-mri-data-harmonization-in-the-age-of-machine-learning-a-multicenter-study-across-36-datasets/121105/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is MRI data harmonization necessary in multicenter machine learning studies?","Question",{"text":75,"@type":76},"To reduce confounding variability introduced by non-biological differences across sites, such as scanner and acquisition-related factors, while preserving biological signal.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does harmonizing the entire dataset before training lead to data leakage?",{"text":80,"@type":76},"When harmonization uses information from outside the training set, that information can influence model building, which can falsely overestimate predictive performance.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the harmonizer transformer and how does it prevent data leakage?",{"text":84,"@type":76},"It wraps ComBat harmonization inside the preprocessing steps of an ML pipeline so harmonization is applied within the training framework design, avoiding leakage by construction.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]