NEČASOVÁ, Tereza, Ninon BURGOS and David SVOBODA. Validation and evaluation metrics for medical and biomedical image synthesis. In Ninon Burgos, David Svoboda. Biomedical Image Synthesis and Simulation - Methods and Applications. 1st ed. Neuveden: Elsevier, 2022, p. 573-600. The MICCAI Society book Series. ISBN 978-0-12-824349-7. Available from: https://dx.doi.org/10.1016/B978-0-12-824349-7.00032-3.
Other formats:   BibTeX LaTeX RIS
Basic information
Original name Validation and evaluation metrics for medical and biomedical image synthesis
Authors NEČASOVÁ, Tereza (203 Czech Republic, guarantor, belonging to the institution), Ninon BURGOS (250 France) and David SVOBODA (203 Czech Republic, belonging to the institution).
Edition 1st ed. Neuveden, Biomedical Image Synthesis and Simulation - Methods and Applications, p. 573-600, 28 pp. The MICCAI Society book Series, 2022.
Publisher Elsevier
Other information
Original language English
Type of outcome Chapter(s) of a specialized book
Field of Study 10201 Computer sciences, information science, bioinformatics
Country of publisher United States of America
Confidentiality degree is not subject to a state or trade secret
Publication form printed version "print"
WWW URL
RIV identification code RIV/00216224:14330/22:00126257
Organization unit Faculty of Informatics
ISBN 978-0-12-824349-7
Doi http://dx.doi.org/10.1016/B978-0-12-824349-7.00032-3
Keywords in English Image synthesis; Image simulation; Similarity;Distance metrics;Image datasets
Tags cbia-web
Tags International impact
Changed by Changed by: RNDr. Pavel Šmerk, Ph.D., učo 3880. Changed: 28/3/2023 11:40.
Abstract
Synthetic image data play an important role in the verification of medical and biomedical image analysis algorithms. However, the usage of such data strongly relies on their quality and plausibility. Despite the emergence of many frameworks for image synthesis in recent years, the quality of the generated images has not been sufficiently assessed in many cases, or the methodology varied across the publications. If we want to use synthetic image data for the verification of biomedical analysis tools, then the images should resemble the real ones as much as possible with evidence about their similarity. Initially, the hardware available for simulations was limited. Therefore, the validation was not under the scope of interest. With the technological improvements, the expectations put on synthetic data have arisen. Proper validation of synthetic image data is nowadays becoming essential.
Links
MUNI/A/1145/2021, interní kód MUName: Rozsáhlé výpočetní systémy: modely, aplikace a verifikace XI. (Acronym: SV-FI MAV XI.)
Investor: Masaryk University
PrintDisplayed: 25/4/2024 13:39