A1 Vertaisarvioitu alkuperäisartikkeli tieteellisessä lehdessä
Comparison of Automatic Segmentation and Preprocessing Approaches for Dynamic Total-Body 3D Pet Images with Different Pet Tracers
Tekijät: Jaakkola, Maria K.; Rivera Pineda, Marcela Xiomara; Diaz, Rafael; Rantala, Maria; Jalo, Anna; Karpijoki, Henri; Saari, Teemu; Maaniitty, Teemu; Keller, Thomas; Louhi, Heli; Wahlroos, Saara; Haaparanta-Solin, Merja; Solin, Olof; Hentila, Jaakko; Helin, Jatta S.; Nissinen, Tuuli A.; Eskola, Olli; Rajander, Johan; Knuuti, Juhani; Virtanen, Kirsi A.; Hannukainen, Jarna C.; Lopez-Picon, Francisco; Klen, Riku
Kustantaja: Springer Science and Business Media LLC
Kustannuspaikka: NEW YORK
Julkaisuvuosi: 2025
Journal: Journal of Imaging Informatics in Medicine
Tietokannassa oleva lehden nimi: Journal of Imaging Informatics in Medicine
Lehden akronyymi: J IMAGING INFORM MED
Sivujen määrä: 18
ISSN: 2948-2925
eISSN: 2948-2933
DOI: https://doi.org/10.1007/s10278-025-01540-4
Verkko-osoite: https://doi.org/10.1007/s10278-025-01540-4
Rinnakkaistallenteen osoite: https://research.utu.fi/converis/portal/detail/Publication/498516080
Segmentation is a routine step in PET image analysis, and few automatic tools have been developed for it. However, excluding supervised methods with their own limitations, they are typically designed for older, small images and the implementations are no longer publicly available. Here, we test if different commonly used building blocks of the automatic methods work with large modern total-body PET images. Dynamic total-body images from five different datasets are used for evaluation purposes, and the tested algorithms cover wide range of different preprocessing approaches and unsupervised segmentation methods. The validation is done by comparing the obtained segments to manually drawn ones using Jaccard index, Dice score, precision, and recall as measures of match. Out of the 17 considered segmentation methods, only 6 were computationally usable and provided enough segments for the needs of this study. Among these six feasible methods, hierarchical clustering and HDBSCAN had systematically the lowest Jaccard indices with the manual segmentations, whereas both GMM and k-means had median Jaccards of 0.58 over different organ segments and data sets. GMM outperformed k-means in human data, but with rat images, the two methods had equally good performance k-means having slightly stronger precision and GMM recall. We conclude that most of the commonly used unsupervised segmentation methods are computationally infeasible with the modern PET images, classical clustering algorithms k-means and especially Gaussian mixture model being the most promising candidates for further method development. Even though preprocessing, particularly denoising, improved the results, small organs remained difficult to segment.
Ladattava julkaisu This is an electronic reprint of the original article. |  |
Julkaisussa olevat rahoitustiedot:
The work of MKJ has been supported by donation funds of Faculty of Medicine at University of Turku, and the Finnish Cultural Foundation.Also, MR received support from donation funds of Faculty of Medicine at University of Turku.JCH reports funding from The Academy of Finland (decision 317332), the Finnish Cultural Foundation, the Finnish Cultural Foundation Varsinais-Suomi Regional Fund, the Diabetes Research Foundation of Finland, and State Research Funding/Hospital District of Southwest Finland.KAV report funding from The Academy of Finland (decision 343410), Sigrid Juselius Foundation and State Research Funding/Hospital District of Southwest Finland.JH reports funding from The Finnish Cultural Foundation Varsinais-Suomi Regional Fund, Kyllikki and Uolevi Lehikoinen Foundation, and Finnish Cultural Foundation.
