The paper \u201cProbabilistic Modeling of Inter- and Intra-observer Variability in Medical Image Segmentation\u201d by Arne Schmidt, Pablo Morales-\u00c1lvarez, and Rafael Molina, was accepted for ICCV 2023 in Paris.<\/h3>\n<\/blockquote>\n
<\/p>\n\n\t\t
\n\t\t\t<\/p>\n
I am very happy to announce that our paper “Probabilistic Modeling of Inter- and Intra-observer Variability in Medical Image Segmentation” was accepted for ICCV 2023!<\/p>\n
In the medical field, experts often disagree when assessing images. As studies in histopathology have shown, even the same medical expert often comes to a different conclusion when assessing the same image twice. Therefore, we hope to make the diagnostic processes more reproducible with AI. But the missing ‘ground truth’ also imposes a major challenge when training AI models.<\/p>\n
To tackle this challenge, we propose a Probabilistic Inter-Observer and iNtra-Observer variation NetwOrk (Pionono). Pionono, inspired by the probabilistic U-Net architecture, models the labeling behavior of each annotator by a multivariate Gaussian distribution optimized in an end-to-end training process, as shown in the figure. This allows for a robust training procedure with imperfect labels and accurate prediction of possible ambiguities.<\/p>\n
The preprint of our article is available at: https:\/\/arxiv.org\/abs\/2307.11397<\/a><\/span><\/p>\n \u00a0Thanks to my great supervisors Rafael Molina<\/a><\/span> and Pablo Morales \u00c1lvarez<\/a><\/span> for the great collaboration and all the successful projects in the last three years.<\/p>\n <\/p>\n The proposed Pionono model. The labeling behaviour of each annotator is represented by a multivariate Gaussian distribution. The drawn samples of these distributions are concatenated with the extracted features and then fed into the segmentation head. The output simulates the inter- and intra-observer variability of annotations and is optimized using the real annotations of each rater. The model is trained end-to-end with a combination of log-likelihood loss (LL) and Kulback Leibler (KL) divergence between posterior and prior, combined in the overall ELBO loss<\/em><\/p>\n <\/p>\n Arne Schmidt – ESR8<\/em><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":" Arne Schmidt is Early Stage Researcher (ESR) 8 of the CLARIFY project at the Universidad de Granada.<\/p>\n","protected":false},"author":8,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"on","_et_pb_old_content":"","_et_gb_content_width":"","_eb_attr":"","footnotes":""},"categories":[13],"tags":[],"class_list":["post-3397","post","type-post","status-publish","format-standard","hentry","category-publications"],"_links":{"self":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts\/3397","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/comments?post=3397"}],"version-history":[{"count":0,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts\/3397\/revisions"}],"wp:attachment":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/media?parent=3397"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/categories?post=3397"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/tags?post=3397"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-content\/uploads\/2023\/07\/Imagen1_esr8-300x242.png\")
<\/p>\n