[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120200-en":3,"doc-seo-120200-105":30,"detail-sidebar-cat-0-en-105":92},{"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},120200,687197207639,"Asher","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Reproducibility in Machine Learning for Medical Imaging - Chapter 21 - Introduction","Reproducibility is presented as a cornerstone of the scientific method, enabling replicated findings to become durable knowledge and helping address a widespread reproducibility crisis in research. The chapter offers an introduction to reproducibility for machine learning in medical imaging by distinguishing key types, specifying requirements to achieve them, and clarifying their utility. It concludes by highlighting practical benefits and advocating a nondogmatic approach to implementation in research workflows.","Chapter 21  \nReproducibility in Machine Learning for Medical Imaging Olivier Colliot, Elina Thibeau-Sutre, and Ninon Burgos  \nAbstract  \nReproducibility is a cornerstone of science, as the replication of findings is the process through which they become knowledge. It is widely considered that many fields of science are undergoing a reproducibility crisis. This has led to the publications of various guidelines in order to improve research reproducibility.  \nThis didactic chapter intends at being an introduction to reproducibility for researchers in the field of machine learning for medical imaging. We first distinguish between different types of reproducibility. For each of them, we aim at defining it, at describing the requirements to achieve it, and at discussing its utility. The chapter ends with a discussion on the benefits of reproducibility and with a plea for a nondogmatic approach to this concept and its implementation in research practice.  \nKey words Reproducibility, Replicability, Reliability, Repeatability, Open science, Machine learning, Artificial intelligence, Deep learning, Medical imaging  \n1 Introduction  \nReproducibility is at the core of the scientific method. In its general and most common meaning, it corresponds to the ability to reproduce the findings ofa given experimental study. This is a necessary (but not sufficient) condition for a scientific statement to become accepted as new knowledge. Let’s illustrate this with a simple example, considering the following statement: “the volume of the hippocampus is, on average, smaller in patients with Alzheimer’s disease (AD) than in healthy people of comparable age.” Such statement was the conclusion of studies which measured such volume from magnetic resonance images (MRI) . To the best of our knowledge, the first study to assert this was that of Seab et al  \n[1]. This was later reproduced by many other studies (e.g., [2, 3]) . It is now widely accepted, which would not have been the case ifthe study had proven impossible to reproduce. Note that, as stated above, this is a necessary but not a sufficient condition. Indeed, there could be other reasons for such statement not to be considered as knowledge. For instance, let’s imagine that some other  \nOlivier Colliot (ed.), Machine Learning for Brain Disorders, Neuromethods, vol. 197, [https://doi.org/10.1007/978-1-0716-3195-9_21](https://doi.org/10.1007/978-1-0716-3195-9_21) ,© The Author(s) 2023  \n632 Olivier Colliot et al.  \nresearchers discover that there is an artifact that is systematically present in the MRI of patients with AD and which leads to erroneous volume estimation. Then, the statement could not be considered new knowledge even though it had been reproduced several times.  \nMachine learning (ML) is, in part, an experimental science. This is not the case of the entirety of the discipline, part of which is theoretical (for instance, mathematical proofs of convergence or of approximation capabilities of different classes of models) or methodological (the invention of a new approach) . Nevertheless, since ML ultimately aims at solving practical problems, its experimental component is essential. Typically, one would want to be able to make statements ofthe type described above from an experimental study. Here is an example of such statement: “this ML model (for instance, a specific convolutional neural network [CNN] architecture), using MRI data as input, is capable of classifying AD patients and healthy controls with an accuracy superior to 80% .”In order to end an article with such a statement, one needs to conduct an experimental study. For such findings to become knowledge, it needs to be subsequently reproduced. Of course, this statement is unlikely to be universal, and one would want to know under which conditions it holds: for instance, is it restricted to a specific class of MRI scanners, to specific disease stages, to specific age ranges?  \nBox 1: Glossary  \nThe readers will find the definition of the terms we us","cbCaiiHkE0NwHyWq","https://ap.wps.com/l/cbCaiiHkE0NwHyWq","pdf",640523,1,23,"English","en",105,"# Introduction\n## Reproducibility as a scientific method\n## Reproducibility in machine learning experiments\n## Core concepts glossary","[{\"question\":\"What does reproducibility mean in the context of scientific research?\",\"answer\":\"Reproducibility refers to the ability to reproduce the findings of a given experimental study, which is necessary for acceptance as new knowledge. It is required but not sufficient, since other issues can prevent findings from becoming valid knowledge.\"},{\"question\":\"Why is reproducibility important for machine learning in medical imaging?\",\"answer\":\"Machine learning is partly an experimental science, especially because it aims to solve practical problems. Experimental findings from ML models must be reproduced, and conditions such as scanner type, disease stage, and age range determine where claims hold.\"},{\"question\":\"What key terms are used to describe reproducibility in the chapter?\",\"answer\":\"The chapter treats reproducibility, replicability, and repeatability as synonyms. It also defines related concepts such as original studies, replication studies, research artifacts, claims, limitations, methods, code, software dependencies, and data types including public and semi-public data.\"}]","Reproducibility in Machine Learning for Medical Imaging - Chapter 21 - Introduction | PDF",1785728684,58,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"reproducibility-in-machine-learning-for-medical-imaging-chapter-21-introduction","",{"@graph":36,"@context":86},[37,54,69],{"@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/reproducibility-in-machine-learning-for-medical-imaging-chapter-21-introduction/120200/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What does reproducibility mean in the context of scientific research?","Question",{"text":76,"@type":77},"Reproducibility refers to the ability to reproduce the findings of a given experimental study, which is necessary for acceptance as new knowledge. It is required but not sufficient, since other issues can prevent findings from becoming valid knowledge.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Why is reproducibility important for machine learning in medical imaging?",{"text":81,"@type":77},"Machine learning is partly an experimental science, especially because it aims to solve practical problems. Experimental findings from ML models must be reproduced, and conditions such as scanner type, disease stage, and age range determine where claims hold.",{"name":83,"@type":74,"acceptedAnswer":84},"What key terms are used to describe reproducibility in the chapter?",{"text":85,"@type":77},"The chapter treats reproducibility, replicability, and repeatability as synonyms. It also defines related concepts such as original studies, replication studies, research artifacts, claims, limitations, methods, code, software dependencies, and data types including public and semi-public data.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]