Using computational pathology and molecular subtyping information in bladder cancer clinical practice is hampered. The first step to feel this gap is providing a proper database that includes whole slide images, molecular data, and clinical records.<\/h3>\n<\/blockquote>\n<\/h2>\n\n\t\t\n\t\t\t\n\t\t\t\t\n\t\t\t
\n\t\t\t<\/div>\n\t\t<\/div> \n\t\t\n\t\t\tWe (EUCRG lab) aim to fill the gap between molecular subtyping information and bladder cancer clinical practice. The first step is done by providing a proper dataset.<\/span>\n\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n <\/p>\n
Computational pathology has shown promising results in improving diagnosis and prognosis in bladder cancer by developing several AI-based applications such as grading, staging, and predicting clinical outcome. Molecular subtyping of bladder cancer patients also could be a predictive factor by classifying patients with similar clinical outcome, and several studies predicted molecular subtypes by analyzing histopathological slides. Nevertheless, using computational pathology and molecular subtyping information in bladder cancer clinical practice is hampered. The first step to feel this gap is providing a proper database that includes whole slide images, molecular data, and clinical records. In this post, I am introducing the database we have developed to feel the mentioned gap.<\/span><\/p>\nWhat are the aims?<\/span><\/strong><\/p>\nTwo main aim have been following in our research:<\/span><\/p>\n\n- How to predict response to treatment from H&E whole slide images using AI-based image analysis methods?<\/span><\/li>\n
- How to predict molecular subtype from H&E whole slide images using AI-based image analysis methods?<\/span><\/li>\n<\/ul>\n
What do we have?<\/span><\/strong><\/p>\nWe (EUCRG lab, Fig 1) included 1150 high-risk non-muscle-invasive bladder cancer (HR-NMIBC) patients who received BCG treatment from 2000-2018 in six different hospitals. <\/span>H&E slides of the primary tumors were scanned. During quality control (QC), WSIs with out-of-focus regions were re-scanned. A uropathologist classified all slides for the grade, stage, and CIS, which, together with follow-up information, were considered weak labels for WSIs. Strong labels were created by delineating predictive areas for progression on a WSI \u2013 e.g., grade, stage, and CIS \u2013 and were confirmed by a uropathologist. A consensus was made to annotate at least 20 strong labels per WSI. RNA isolation has been done from up to five punches from FFPE blocks. Here you can find an overview of what have we collected so far:<\/span><\/p>\n\n- Clinicopathological data (e.g., grade, stage, age, gender, smoking status, variant histology) for 11500 patients.<\/span><\/li>\n
- Whole slide images from primary tumor samples for 1150 patients.<\/span><\/li>\n
- Annotation on the whole slide images which represent grade, invasive area, tumor and stroma immune cell infiltrated areas, tissue types, and artifacts for 100 patients.<\/span><\/li>\n
- RNA-seq data for 287 patients with corresponding novel molecular subtype classification, which was recently developed in our lab.<\/span><\/li>\n<\/ul>\n
Here, we provide an extensive dataset that could readily be applied to develop clinically translatable computational pathology tools predicting clinical outcome and detecting molecular subtypes from primary BC WSIs. When looking into computational pathology aspects, the time to prepare a WSI dataset should be considered.<\/p>\n
<\/span><\/span><\/em><\/p>\nFarbod Khoraminia<\/span><\/span> – ESR10<\/em><\/p>\n[\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":"
Computational pathology has shown promising results in improving diagnosis and prognosis in bladder cancer by developing several AI-based applications such as grading, staging, and predicting clinical outcome. <\/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":[10],"tags":[],"class_list":["post-3143","post","type-post","status-publish","format-standard","hentry","category-about-project"],"_links":{"self":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts\/3143","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=3143"}],"version-history":[{"count":0,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts\/3143\/revisions"}],"wp:attachment":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/media?parent=3143"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/categories?post=3143"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/tags?post=3143"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\n\t\t\t<\/p>\n
Computational pathology has shown promising results in improving diagnosis and prognosis in bladder cancer by developing several AI-based applications such as grading, staging, and predicting clinical outcome. Molecular subtyping of bladder cancer patients also could be a predictive factor by classifying patients with similar clinical outcome, and several studies predicted molecular subtypes by analyzing histopathological slides. Nevertheless, using computational pathology and molecular subtyping information in bladder cancer clinical practice is hampered. The first step to feel this gap is providing a proper database that includes whole slide images, molecular data, and clinical records. In this post, I am introducing the database we have developed to feel the mentioned gap.<\/span><\/p>\n What are the aims?<\/span><\/strong><\/p>\n Two main aim have been following in our research:<\/span><\/p>\n What do we have?<\/span><\/strong><\/p>\n We (EUCRG lab, Fig 1) included 1150 high-risk non-muscle-invasive bladder cancer (HR-NMIBC) patients who received BCG treatment from 2000-2018 in six different hospitals. <\/span>H&E slides of the primary tumors were scanned. During quality control (QC), WSIs with out-of-focus regions were re-scanned. A uropathologist classified all slides for the grade, stage, and CIS, which, together with follow-up information, were considered weak labels for WSIs. Strong labels were created by delineating predictive areas for progression on a WSI \u2013 e.g., grade, stage, and CIS \u2013 and were confirmed by a uropathologist. A consensus was made to annotate at least 20 strong labels per WSI. RNA isolation has been done from up to five punches from FFPE blocks. Here you can find an overview of what have we collected so far:<\/span><\/p>\n Here, we provide an extensive dataset that could readily be applied to develop clinically translatable computational pathology tools predicting clinical outcome and detecting molecular subtypes from primary BC WSIs. When looking into computational pathology aspects, the time to prepare a WSI dataset should be considered.<\/p>\n <\/span><\/span><\/em><\/p>\n Farbod Khoraminia<\/span><\/span> – ESR10<\/em><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":" Computational pathology has shown promising results in improving diagnosis and prognosis in bladder cancer by developing several AI-based applications such as grading, staging, and predicting clinical outcome. <\/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":[10],"tags":[],"class_list":["post-3143","post","type-post","status-publish","format-standard","hentry","category-about-project"],"_links":{"self":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts\/3143","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=3143"}],"version-history":[{"count":0,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/posts\/3143\/revisions"}],"wp:attachment":[{"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/media?parent=3143"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/categories?post=3143"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cvblab.webs.upv.es\/clarify-project.eu\/wp-json\/wp\/v2\/tags?post=3143"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\n
\n