Metrics reloaded: recommendations for image analysis validation

Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. In biomedical image analysis, chosen performance metrics often do not reflect the domain interest, and thus fail to adequately measure scientific progress and hinder translation o...

Full description

Saved in:
Bibliographic Details
Main Authors: Maier-Hein, Lena (Author) , Reinke, Annika (Author) , Godau, Patrick (Author) , Tizabi, Minu (Author) , Christodoulou, Evangelia (Author) , Isensee, Fabian (Author) , Wiesenfarth, Manuel (Author) , Kavur, A. Emre (Author) , Baumgartner, Michael (Author) , Eisenmann, Matthias (Author) , Heckmann-Nötzel, Doreen (Author) , Rädsch, Tim (Author) , Kopp-Schneider, Annette (Author) , Kreshuk, Anna (Author) , Maier-Hein, Klaus H. (Author) , Sáez Rodríguez, Julio (Author) , Jäger, Paul F. (Author)
Format: Article (Journal)
Language:English
Published: 12 February 2024
In: Nature methods
Year: 2024, Volume: 21, Issue: 2, Pages: 195-212
ISSN:1548-7105
DOI:10.1038/s41592-023-02151-z
Online Access:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1038/s41592-023-02151-z
Verlag, lizenzpflichtig, Volltext: https://www.nature.com/articles/s41592-023-02151-z
Get full text
Author Notes:Lena Maier-Hein, Annika Reinke, Patrick Godau, Minu D. Tizabi, Florian Buettner, Evangelia Christodoulou, Ben Glocker, Fabian Isensee, Jens Kleesiek, Michal Kozubek, Mauricio Reyes, Michael A. Riegler, Manuel Wiesenfarth, A. Emre Kavur, Carole H. Sudre, Michael Baumgartner, Matthias Eisenmann, Doreen Heckmann-Nötzel, Tim Rädsch, Laura Acion, Michela Antonelli, Tal Arbel, Spyridon Bakas, Arriel Benis, Matthew B. Blaschko, M. Jorge Cardoso, Veronika Cheplygina, Beth A. Cimini, Gary S. Collins, Keyvan Farahani, Luciana Ferrer, Adrian Galdran, Bram van Ginneken, Robert Haase, Daniel A. Hashimoto, Michael M. Hoffman, Merel Huisman, Pierre Jannin, Charles E. Kahn, Dagmar Kainmueller, Bernhard Kainz, Alexandros Karargyris, Alan Karthikesalingam, Florian Kofler, Annette Kopp-Schneider, Anna Kreshuk, Tahsin Kurc, Bennett A. Landman, Geert Litjens, Amin Madani, Klaus Maier-Hein, Anne L. Martel, Peter Mattson, Erik Meijering, Bjoern Menze, Karel G. M. Moons, Henning Müller, Brennan Nichyporuk, Felix Nickel, Jens Petersen, Nasir Rajpoot, Nicola Rieke, Julio Saez-Rodriguez, Clara I. Sánchez, Shravya Shetty, Maarten van Smeden, Ronald M. Summers, Abdel A. Taha, Aleksei Tiulpin, Sotirios A. Tsaftaris, Ben Van Calster, Gaël Varoquaux, Paul F. Jäger

MARC

LEADER 00000caa a2200000 c 4500
001 1882382722
003 DE-627
005 20250901164100.0
007 cr uuu---uuuuu
008 240304s2024 xx |||||o 00| ||eng c
024 7 |a 10.1038/s41592-023-02151-z  |2 doi 
035 |a (DE-627)1882382722 
035 |a (DE-599)KXP1882382722 
035 |a (OCoLC)1425199950 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 33  |2 sdnb 
100 1 |a Maier-Hein, Lena  |d 1980-  |e VerfasserIn  |0 (DE-588)1075029252  |0 (DE-627)832869899  |0 (DE-576)190090804  |4 aut 
245 1 0 |a Metrics reloaded  |b recommendations for image analysis validation  |c Lena Maier-Hein, Annika Reinke, Patrick Godau, Minu D. Tizabi, Florian Buettner, Evangelia Christodoulou, Ben Glocker, Fabian Isensee, Jens Kleesiek, Michal Kozubek, Mauricio Reyes, Michael A. Riegler, Manuel Wiesenfarth, A. Emre Kavur, Carole H. Sudre, Michael Baumgartner, Matthias Eisenmann, Doreen Heckmann-Nötzel, Tim Rädsch, Laura Acion, Michela Antonelli, Tal Arbel, Spyridon Bakas, Arriel Benis, Matthew B. Blaschko, M. Jorge Cardoso, Veronika Cheplygina, Beth A. Cimini, Gary S. Collins, Keyvan Farahani, Luciana Ferrer, Adrian Galdran, Bram van Ginneken, Robert Haase, Daniel A. Hashimoto, Michael M. Hoffman, Merel Huisman, Pierre Jannin, Charles E. Kahn, Dagmar Kainmueller, Bernhard Kainz, Alexandros Karargyris, Alan Karthikesalingam, Florian Kofler, Annette Kopp-Schneider, Anna Kreshuk, Tahsin Kurc, Bennett A. Landman, Geert Litjens, Amin Madani, Klaus Maier-Hein, Anne L. Martel, Peter Mattson, Erik Meijering, Bjoern Menze, Karel G. M. Moons, Henning Müller, Brennan Nichyporuk, Felix Nickel, Jens Petersen, Nasir Rajpoot, Nicola Rieke, Julio Saez-Rodriguez, Clara I. Sánchez, Shravya Shetty, Maarten van Smeden, Ronald M. Summers, Abdel A. Taha, Aleksei Tiulpin, Sotirios A. Tsaftaris, Ben Van Calster, Gaël Varoquaux, Paul F. Jäger 
264 1 |c 12 February 2024 
300 |a 18 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
500 |a Gesehen am 04.03.2024 
520 |a Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. In biomedical image analysis, chosen performance metrics often do not reflect the domain interest, and thus fail to adequately measure scientific progress and hinder translation of ML techniques into practice. To overcome this, we created Metrics Reloaded, a comprehensive framework guiding researchers in the problem-aware selection of metrics. Developed by a large international consortium in a multistage Delphi process, it is based on the novel concept of a problem fingerprint—a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), dataset and algorithm output. On the basis of the problem fingerprint, users are guided through the process of choosing and applying appropriate validation metrics while being made aware of potential pitfalls. Metrics Reloaded targets image analysis problems that can be interpreted as classification tasks at image, object or pixel level, namely image-level classification, object detection, semantic segmentation and instance segmentation tasks. To improve the user experience, we implemented the framework in the Metrics Reloaded online tool. Following the convergence of ML methodology across application domains, Metrics Reloaded fosters the convergence of validation methodology. Its applicability is demonstrated for various biomedical use cases. 
650 4 |a Education 
650 4 |a Medical research 
700 1 |a Reinke, Annika  |d 1993-  |e VerfasserIn  |0 (DE-588)1219684252  |0 (DE-627)1735662143  |4 aut 
700 1 |8 1\p  |a Godau, Patrick  |d 1993-  |e VerfasserIn  |0 (DE-588)1179282515  |0 (DE-627)1066788200  |0 (DE-576)518174212  |4 aut 
700 1 |a Tizabi, Minu  |d 1992-  |e VerfasserIn  |0 (DE-588)1140542494  |0 (DE-627)89854632X  |0 (DE-576)493859608  |4 aut 
700 1 |a Christodoulou, Evangelia  |e VerfasserIn  |4 aut 
700 1 |a Isensee, Fabian  |d 1990-  |e VerfasserIn  |0 (DE-588)1207568430  |0 (DE-627)1694044998  |4 aut 
700 1 |a Wiesenfarth, Manuel  |e VerfasserIn  |4 aut 
700 1 |a Kavur, A. Emre  |e VerfasserIn  |4 aut 
700 1 |a Baumgartner, Michael  |d 1996-  |e VerfasserIn  |0 (DE-588)1338678140  |0 (DE-627)1898470472  |4 aut 
700 1 |8 2\p  |a Eisenmann, Matthias  |e VerfasserIn  |0 (DE-588)1219684139  |0 (DE-627)1735661740  |4 aut 
700 1 |a Heckmann-Nötzel, Doreen  |e VerfasserIn  |4 aut 
700 1 |a Rädsch, Tim  |e VerfasserIn  |4 aut 
700 1 |a Kopp-Schneider, Annette  |e VerfasserIn  |0 (DE-588)1119160545  |0 (DE-627)872460444  |0 (DE-576)178153206  |4 aut 
700 1 |a Kreshuk, Anna  |e VerfasserIn  |0 (DE-588)1031765751  |0 (DE-627)737325941  |0 (DE-576)369550420  |4 aut 
700 1 |a Maier-Hein, Klaus H.  |d 1980-  |e VerfasserIn  |0 (DE-588)1100551875  |0 (DE-627)85946461X  |0 (DE-576)333771222  |4 aut 
700 1 |a Sáez Rodríguez, Julio  |d 1978-  |e VerfasserIn  |0 (DE-588)133764362  |0 (DE-627)555766632  |0 (DE-576)300083114  |4 aut 
700 1 |a Jäger, Paul F.  |e VerfasserIn  |4 aut 
773 0 8 |i Enthalten in  |t Nature methods  |d London [u.a.] : Nature Publishing Group, 2004  |g 21(2024), 2, Seite 195-212  |h Online-Ressource  |w (DE-627)397615310  |w (DE-600)2163081-1  |w (DE-576)118489089  |x 1548-7105  |7 nnas  |a Metrics reloaded recommendations for image analysis validation 
773 1 8 |g volume:21  |g year:2024  |g number:2  |g pages:195-212  |g extent:18  |a Metrics reloaded recommendations for image analysis validation 
856 4 0 |u https://doi.org/10.1038/s41592-023-02151-z  |x Verlag  |x Resolving-System  |z lizenzpflichtig  |3 Volltext 
856 4 0 |u https://www.nature.com/articles/s41592-023-02151-z  |x Verlag  |z lizenzpflichtig  |3 Volltext 
883 |8 1\p  |a cgwrk  |d 20250802  |q DE-101  |u https://d-nb.info/provenance/plan#cgwrk 
883 |8 2\p  |a cgwrk  |d 20241001  |q DE-101  |u https://d-nb.info/provenance/plan#cgwrk 
951 |a AR 
992 |a 20240304 
993 |a Article 
994 |a 2024 
998 |g 133764362  |a Sáez Rodríguez, Julio  |m 133764362:Sáez Rodríguez, Julio  |d 910000  |d 912900  |e 910000PS133764362  |e 912900PS133764362  |k 0/910000/  |k 1/910000/912900/  |p 64 
998 |g 1100551875  |a Maier-Hein, Klaus H.  |m 1100551875:Maier-Hein, Klaus H.  |d 910000  |d 911400  |e 910000PM1100551875  |e 911400PM1100551875  |k 0/910000/  |k 1/910000/911400/  |p 51 
998 |g 1119160545  |a Kopp-Schneider, Annette  |m 1119160545:Kopp-Schneider, Annette  |d 50000  |e 50000PK1119160545  |k 0/50000/  |p 45 
998 |g 1338678140  |a Baumgartner, Michael  |m 1338678140:Baumgartner, Michael  |d 110000  |e 110000PB1338678140  |k 0/110000/  |p 9 
998 |g 1207568430  |a Isensee, Fabian  |m 1207568430:Isensee, Fabian  |p 8 
998 |g 1179282515  |a Godau, Patrick  |m 1179282515:Godau, Patrick  |d 110000  |e 110000PG1179282515  |k 0/110000/  |p 3 
998 |g 1219684252  |a Reinke, Annika  |m 1219684252:Reinke, Annika  |d 50000  |e 50000PR1219684252  |k 0/50000/  |p 2 
998 |g 1075029252  |a Maier-Hein, Lena  |m 1075029252:Maier-Hein, Lena  |d 110000  |e 110000PM1075029252  |k 0/110000/  |p 1  |x j 
999 |a KXP-PPN1882382722  |e 4495373161 
BIB |a Y 
SER |a journal 
JSO |a {"language":["eng"],"type":{"bibl":"article-journal","media":"Online-Ressource"},"origin":[{"dateIssuedDisp":"12 February 2024","dateIssuedKey":"2024"}],"physDesc":[{"extent":"18 S."}],"person":[{"display":"Maier-Hein, Lena","roleDisplay":"VerfasserIn","given":"Lena","role":"aut","family":"Maier-Hein"},{"family":"Reinke","role":"aut","given":"Annika","display":"Reinke, Annika","roleDisplay":"VerfasserIn"},{"display":"Godau, Patrick","roleDisplay":"VerfasserIn","given":"Patrick","role":"aut","family":"Godau"},{"role":"aut","family":"Tizabi","given":"Minu","roleDisplay":"VerfasserIn","display":"Tizabi, Minu"},{"roleDisplay":"VerfasserIn","display":"Christodoulou, Evangelia","given":"Evangelia","family":"Christodoulou","role":"aut"},{"family":"Isensee","role":"aut","roleDisplay":"VerfasserIn","display":"Isensee, Fabian","given":"Fabian"},{"given":"Manuel","display":"Wiesenfarth, Manuel","roleDisplay":"VerfasserIn","role":"aut","family":"Wiesenfarth"},{"family":"Kavur","role":"aut","roleDisplay":"VerfasserIn","display":"Kavur, A. Emre","given":"A. Emre"},{"display":"Baumgartner, Michael","roleDisplay":"VerfasserIn","given":"Michael","family":"Baumgartner","role":"aut"},{"display":"Eisenmann, Matthias","roleDisplay":"VerfasserIn","given":"Matthias","role":"aut","family":"Eisenmann"},{"role":"aut","family":"Heckmann-Nötzel","roleDisplay":"VerfasserIn","display":"Heckmann-Nötzel, Doreen","given":"Doreen"},{"role":"aut","family":"Rädsch","display":"Rädsch, Tim","roleDisplay":"VerfasserIn","given":"Tim"},{"role":"aut","family":"Kopp-Schneider","given":"Annette","display":"Kopp-Schneider, Annette","roleDisplay":"VerfasserIn"},{"given":"Anna","display":"Kreshuk, Anna","roleDisplay":"VerfasserIn","family":"Kreshuk","role":"aut"},{"family":"Maier-Hein","role":"aut","given":"Klaus H.","roleDisplay":"VerfasserIn","display":"Maier-Hein, Klaus H."},{"role":"aut","family":"Sáez Rodríguez","given":"Julio","display":"Sáez Rodríguez, Julio","roleDisplay":"VerfasserIn"},{"role":"aut","family":"Jäger","given":"Paul F.","roleDisplay":"VerfasserIn","display":"Jäger, Paul F."}],"id":{"doi":["10.1038/s41592-023-02151-z"],"eki":["1882382722"]},"relHost":[{"id":{"issn":["1548-7105"],"eki":["397615310"],"zdb":["2163081-1"]},"pubHistory":["1.2004 -"],"recId":"397615310","title":[{"title":"Nature methods","title_sort":"Nature methods","subtitle":"techniques for life scientists and chemists"}],"note":["Gesehen am 14. August 2018"],"disp":"Metrics reloaded recommendations for image analysis validationNature methods","language":["eng"],"part":{"extent":"18","volume":"21","year":"2024","issue":"2","text":"21(2024), 2, Seite 195-212","pages":"195-212"},"origin":[{"publisher":"Nature Publishing Group","dateIssuedDisp":"2004-","publisherPlace":"London [u.a.]","dateIssuedKey":"2004"}],"type":{"media":"Online-Ressource","bibl":"periodical"},"physDesc":[{"extent":"Online-Ressource"}]}],"recId":"1882382722","title":[{"title_sort":"Metrics reloaded","title":"Metrics reloaded","subtitle":"recommendations for image analysis validation"}],"note":["Gesehen am 04.03.2024"],"name":{"displayForm":["Lena Maier-Hein, Annika Reinke, Patrick Godau, Minu D. Tizabi, Florian Buettner, Evangelia Christodoulou, Ben Glocker, Fabian Isensee, Jens Kleesiek, Michal Kozubek, Mauricio Reyes, Michael A. Riegler, Manuel Wiesenfarth, A. Emre Kavur, Carole H. Sudre, Michael Baumgartner, Matthias Eisenmann, Doreen Heckmann-Nötzel, Tim Rädsch, Laura Acion, Michela Antonelli, Tal Arbel, Spyridon Bakas, Arriel Benis, Matthew B. Blaschko, M. Jorge Cardoso, Veronika Cheplygina, Beth A. Cimini, Gary S. Collins, Keyvan Farahani, Luciana Ferrer, Adrian Galdran, Bram van Ginneken, Robert Haase, Daniel A. Hashimoto, Michael M. Hoffman, Merel Huisman, Pierre Jannin, Charles E. Kahn, Dagmar Kainmueller, Bernhard Kainz, Alexandros Karargyris, Alan Karthikesalingam, Florian Kofler, Annette Kopp-Schneider, Anna Kreshuk, Tahsin Kurc, Bennett A. Landman, Geert Litjens, Amin Madani, Klaus Maier-Hein, Anne L. Martel, Peter Mattson, Erik Meijering, Bjoern Menze, Karel G. M. Moons, Henning Müller, Brennan Nichyporuk, Felix Nickel, Jens Petersen, Nasir Rajpoot, Nicola Rieke, Julio Saez-Rodriguez, Clara I. Sánchez, Shravya Shetty, Maarten van Smeden, Ronald M. Summers, Abdel A. Taha, Aleksei Tiulpin, Sotirios A. Tsaftaris, Ben Van Calster, Gaël Varoquaux, Paul F. Jäger"]}} 
SRT |a MAIERHEINLMETRICSREL1220