MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis

Teresa Zulueta-Coarasa1,; Florian Jug; Aastha   Mathur; Josh Moore; Arrate Muñoz-Barrutia; Liviu Anita; Kola Babalola; Peter Bankhead; Perrine Gilloteaux; Nodar Gogoberidze; Martin Jones; Gerard J. Kleywegt; Paul Korir; Anna Kreshuk; Aybuke Kupcu Yoldas; Luca Marconato; Kedar Narayan; Nils Norlan; Bugra Oezdemir; Jessica Riesterer; Norman Rzepka; Ugis Sarkans; Beatriz Serrano; Christian  Tischer; Virginie Uhlmann; Vladimir Ulman; Matthew Hartley

doi:10.1038/s41592-025-02835-8

MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis

Teresa Zulueta-Coarasa1,, Florian Jug, Aastha Mathur, Josh Moore, Arrate Muñoz-Barrutia, Liviu Anita, Kola Babalola, Peter Bankhead, Perrine Gilloteaux, Nodar Gogoberidze, Martin Jones, Gerard J. Kleywegt, Paul Korir, Anna Kreshuk, Aybuke Kupcu Yoldas, Luca Marconato, Kedar Narayan, Nils Norlan, Bugra Oezdemir, Jessica RiestererNorman Rzepka, Ugis Sarkans, Beatriz Serrano, Christian Tischer, Virginie Uhlmann, Vladimir Ulman, Matthew Hartley

Research output: Contribution to journal › Article › peer-review

Abstract

Artificial Intelligence methods are powerful tools for biological image analysis and
processing. High-quality annotated images are key to training and developing new methods, but access to such data is often hindered by the lack of standards for sharing datasets. We brought together community experts in a workshop to develop guidelines to improve the reuse of bioimages and annotations for AI applications. These include standards on data formats, metadata, data presentation and sharing, and incentives to generate new datasets. We are positive that the MIFA (Metadata, Incentives, Formats, and Accessibility) recommendations will accelerate the development of AI tools for bioimage analysis by facilitating access to high quality training data.

Original language	English
Journal	Nature Methods
Early online date	15 Sept 2025
DOIs	https://doi.org/10.1038/s41592-025-02835-8
Publication status	E-pub ahead of print - 15 Sept 2025

Access to Document

10.1038/s41592-025-02835-8

Embargoed Document

2025-10-MIFA Accepted manuscript
Accepted author manuscript, 3.41 MB
Embargo ends: 15/03/26

Cite this

Zulueta-Coarasa1, T., Jug, F., Mathur, A., Moore, J., Muñoz-Barrutia, A., Anita, L., Babalola, K., Bankhead, P., Gilloteaux, P., Gogoberidze, N., Jones, M., Kleywegt, G. J., Korir, P., Kreshuk, A., Yoldas, A. K., Marconato, L., Narayan, K., Norlan, N., Oezdemir, B., ... Hartley, M. (2025). MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis. Nature Methods. Advance online publication. https://doi.org/10.1038/s41592-025-02835-8

@article{f1a41a2f161a45aa95bb6f24fadd6653,

title = "MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis",

abstract = "Artificial Intelligence methods are powerful tools for biological image analysis and processing. High-quality annotated images are key to training and developing new methods, but access to such data is often hindered by the lack of standards for sharing datasets. We brought together community experts in a workshop to develop guidelines to improve the reuse of bioimages and annotations for AI applications. These include standards on data formats, metadata, data presentation and sharing, and incentives to generate new datasets. We are positive that the MIFA (Metadata, Incentives, Formats, and Accessibility) recommendations will accelerate the development of AI tools for bioimage analysis by facilitating access to high quality training data.",

author = "Teresa Zulueta-Coarasa1, and Florian Jug and Aastha Mathur and Josh Moore and Arrate Mu{\~n}oz-Barrutia and Liviu Anita and Kola Babalola and Peter Bankhead and Perrine Gilloteaux and Nodar Gogoberidze and Martin Jones and Kleywegt, \{Gerard J.\} and Paul Korir and Anna Kreshuk and Yoldas, \{Aybuke Kupcu\} and Luca Marconato and Kedar Narayan and Nils Norlan and Bugra Oezdemir and Jessica Riesterer and Norman Rzepka and Ugis Sarkans and Beatriz Serrano and Christian Tischer and Virginie Uhlmann and Vladimir Ulman and Matthew Hartley",

year = "2025",

month = sep,

day = "15",

doi = "10.1038/s41592-025-02835-8",

language = "English",

journal = "Nature Methods",

issn = "1548-7091",

publisher = "Nature Publishing Group",

}

Zulueta-Coarasa1, T, Jug, F, Mathur, A, Moore, J, Muñoz-Barrutia, A, Anita, L, Babalola, K, Bankhead, P, Gilloteaux, P, Gogoberidze, N, Jones, M, Kleywegt, GJ, Korir, P, Kreshuk, A, Yoldas, AK, Marconato, L, Narayan, K, Norlan, N, Oezdemir, B, Riesterer, J, Rzepka, N, Sarkans, U, Serrano, B, Tischer, C, Uhlmann, V, Ulman, V & Hartley, M 2025, 'MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis', Nature Methods. https://doi.org/10.1038/s41592-025-02835-8

TY - JOUR

T1 - MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis

AU - Zulueta-Coarasa1,, Teresa

AU - Jug, Florian

AU - Mathur, Aastha

AU - Moore, Josh

AU - Muñoz-Barrutia, Arrate

AU - Anita, Liviu

AU - Babalola, Kola

AU - Bankhead, Peter

AU - Gilloteaux, Perrine

AU - Gogoberidze, Nodar

AU - Jones, Martin

AU - Kleywegt, Gerard J.

AU - Korir, Paul

AU - Kreshuk, Anna

AU - Yoldas, Aybuke Kupcu

AU - Marconato, Luca

AU - Narayan, Kedar

AU - Norlan, Nils

AU - Oezdemir, Bugra

AU - Riesterer, Jessica

AU - Rzepka, Norman

AU - Sarkans, Ugis

AU - Serrano, Beatriz

AU - Tischer, Christian

AU - Uhlmann, Virginie

AU - Ulman, Vladimir

AU - Hartley, Matthew

PY - 2025/9/15

Y1 - 2025/9/15

N2 - Artificial Intelligence methods are powerful tools for biological image analysis and processing. High-quality annotated images are key to training and developing new methods, but access to such data is often hindered by the lack of standards for sharing datasets. We brought together community experts in a workshop to develop guidelines to improve the reuse of bioimages and annotations for AI applications. These include standards on data formats, metadata, data presentation and sharing, and incentives to generate new datasets. We are positive that the MIFA (Metadata, Incentives, Formats, and Accessibility) recommendations will accelerate the development of AI tools for bioimage analysis by facilitating access to high quality training data.

AB - Artificial Intelligence methods are powerful tools for biological image analysis and processing. High-quality annotated images are key to training and developing new methods, but access to such data is often hindered by the lack of standards for sharing datasets. We brought together community experts in a workshop to develop guidelines to improve the reuse of bioimages and annotations for AI applications. These include standards on data formats, metadata, data presentation and sharing, and incentives to generate new datasets. We are positive that the MIFA (Metadata, Incentives, Formats, and Accessibility) recommendations will accelerate the development of AI tools for bioimage analysis by facilitating access to high quality training data.

U2 - 10.1038/s41592-025-02835-8

DO - 10.1038/s41592-025-02835-8

M3 - Article

SN - 1548-7091

JO - Nature Methods

JF - Nature Methods

ER -

MIFA: Metadata, Incentives, Formats, and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis

Abstract

Access to Document

Embargoed Document

Fingerprint

Cite this