{"id":32637,"date":"2026-08-04T20:20:12","date_gmt":"2026-08-04T18:20:12","guid":{"rendered":"https:\/\/sano.science\/?p=32637"},"modified":"2026-08-18T20:42:05","modified_gmt":"2026-08-18T18:42:05","slug":"teaching-ai-to-fill-in-the-gaps","status":"publish","type":"post","link":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/","title":{"rendered":"Teaching\u00a0AI to\u00a0fill in the gaps"},"content":{"rendered":"\n<p class=\"eplus-wrapper wp-block-paragraph\">At the Health AI &amp; CyberSec Summit in Warsaw, Sano\u2019s Wojciech Szyma\u0144ski showed how to make 3D medical image reconstruction more robust \u2014 and introduced Defect API, a way to train models without hand-labelled data.<\/p>\n\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n\n<p class=\"eplus-wrapper wp-block-paragraph\">On 16 June, Sano researcher Wojciech Szyma\u0144ski took the stage at the Health AI &amp; CyberSec Summit (HAICS) in Warsaw, a meeting held at the Copernicus Science Centre that brought together two communities not often in the same room: healthcare cybersecurity and healthcare AI. Drawing on the Hack Summit and Data Science Summit, the programme spanned data protection, medical imaging and the growing role of artificial intelligence across the health and pharmaceutical sector.<\/p>\n\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n\n<p class=\"eplus-wrapper wp-block-paragraph\">Wojciech presented part of his PhD research on a stubborn problem in medical imaging: how to make the reconstruction of 3D anatomical structures more reliable. Scans are rarely perfect. Parts of an organ or vessel can be missing, noisy or distorted, and a model trained to rebuild them usually needs large collections of examples that a person has painstakingly labelled by hand. Preparing those labels is slow and costly, and it limits how far such models can scale.<\/p>\n\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n<div class=\"wp-block-columns eplus-wrapper is-layout-flex wp-container-core-columns-is-layout-8f761849 wp-block-columns-is-layout-flex eplus-styles-uid-ad04af\">\n<div class=\"wp-block-column eplus-wrapper is-layout-flow wp-block-column-is-layout-flow\">\n<figure class=\"wp-block-image size-large eplus-wrapper\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"1024\" src=\"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-1024x1024.jpg\" alt=\"\" class=\"wp-image-32638\" srcset=\"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-1024x1024.jpg 1024w, https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-300x300.jpg 300w, https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-150x150.jpg 150w, https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-768x768.jpg 768w, https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg 1200w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n\n<div class=\"wp-block-column is-vertically-aligned-center eplus-wrapper is-layout-flow wp-block-column-is-layout-flow\">\n<p class=\"eplus-wrapper wp-block-paragraph\">Wojciech presented part of his PhD research on a stubborn problem in medical imaging: how to make the reconstruction of 3D anatomical structures more reliable. Scans are rarely perfect. Parts of an organ or vessel can be missing, noisy or distorted, and a model trained to rebuild them usually needs large collections of examples that a person has painstakingly labelled by hand. Preparing those labels is slow and costly, and it limits how far such models can scale.<\/p>\n<\/div>\n<\/div>\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n\n<p class=\"eplus-wrapper wp-block-paragraph\">His answer is a framework called Defect API. Instead of relying on hand-prepared data, it generates synthetic structural defects during training: it deliberately damages complete examples in configurable, realistic ways, then asks the model to restore them. Because the correct answer is simply the undamaged original, the model can learn on its own, an approach known as self-supervised learning, with nobody labelling anything. The framework works with both voxel grids and point clouds, two common ways of representing 3D shapes, and it has been connected to well-established architectures including nnU-Net and Point Transformer V3.<\/p>\n\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n\n<p class=\"eplus-wrapper wp-block-paragraph\">Tested on large, varied medical datasets, the method pointed to a clear lesson: the diversity and realism of the training defects strongly shape how robust the model becomes and how well it generalises to data it has not seen before. Put plainly, the quality of the &#8220;damage&#8221; a model learns from can matter as much as the model itself.<\/p>\n\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n\n<p class=\"eplus-wrapper wp-block-paragraph\">Wojciech closed with a live demonstration of the full pipeline, from training the model and tracking its progress in TensorBoard to an interactive 3D view of the reconstructed anatomy. Seeing the whole chain run end to end, rather than a set of final numbers on a slide, made the method easy to grasp for an audience drawn from very different backgrounds.<\/p>\n\n\n\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n\n\n\n<p class=\"eplus-wrapper wp-block-paragraph\">For Sano, the talk is a good example of how careful methodological work in AI feeds directly into the tools that clinicians and researchers may one day rely on. We congratulate Wojciech on the presentation.<\/p>\n","protected":false},"excerpt":"At the Health AI &amp; CyberSec Summit in Warsaw, Sano\u2019s Wojciech Szyma\u0144ski showed how to make 3D medical image reconstruction more robust \u2014 and introduced Defect API, a way to train models without hand-labelled data. On 16 June, Sano researcher Wojciech Szyma\u0144ski took the stage at the Health AI &amp; CyberSec Summit (HAICS) in Warsaw, [&hellip;]","author":8,"featured_media":32638,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"inline_featured_image":false,"editor_plus_post_options":"{}","editor_plus_copied_stylings":"{}","footnotes":""},"categories":[1],"tags":[],"class_list":["post-32637","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v28.2 (Yoast SEO v28.2) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>Teaching\u00a0AI to\u00a0fill in the gaps - Centre for Computational Personalized Medicine<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Teaching\u00a0AI to\u00a0fill in the gaps\" \/>\n<meta property=\"og:description\" content=\"At the Health AI &amp; CyberSec Summit in Warsaw, Sano\u2019s Wojciech Szyma\u0144ski showed how to make 3D medical image reconstruction more robust \u2014 and introduced Defect API, a way to train models without hand-labelled data. On 16 June, Sano researcher Wojciech Szyma\u0144ski took the stage at the Health AI &amp; CyberSec Summit (HAICS) in Warsaw, [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/\" \/>\n<meta property=\"og:site_name\" content=\"Centre for Computational Personalized Medicine\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/sano.science\/\" \/>\n<meta property=\"article:published_time\" content=\"2026-08-04T18:20:12+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-08-18T18:42:05+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"1200\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Sano\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@sanoscience\" \/>\n<meta name=\"twitter:site\" content=\"@sanoscience\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sano\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/\"},\"author\":{\"name\":\"Sano\",\"@id\":\"https:\\\/\\\/sano.science\\\/#\\\/schema\\\/person\\\/561b69b48b6a8f7904aed06df5d03c98\"},\"headline\":\"Teaching\u00a0AI to\u00a0fill in the gaps\",\"datePublished\":\"2026-08-04T18:20:12+00:00\",\"dateModified\":\"2026-08-18T18:42:05+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/\"},\"wordCount\":521,\"publisher\":{\"@id\":\"https:\\\/\\\/sano.science\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/sano.science\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/HAICS_1.jpg\",\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/\",\"url\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/\",\"name\":\"Teaching\u00a0AI to\u00a0fill in the gaps - Centre for Computational Personalized Medicine\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/sano.science\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/sano.science\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/HAICS_1.jpg\",\"datePublished\":\"2026-08-04T18:20:12+00:00\",\"dateModified\":\"2026-08-18T18:42:05+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#primaryimage\",\"url\":\"https:\\\/\\\/sano.science\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/HAICS_1.jpg\",\"contentUrl\":\"https:\\\/\\\/sano.science\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/HAICS_1.jpg\",\"width\":1200,\"height\":1200},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/sano.science\\\/teaching-ai-to-fill-in-the-gaps\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/sano.science\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Teaching\u00a0AI to\u00a0fill in the gaps\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/sano.science\\\/#website\",\"url\":\"https:\\\/\\\/sano.science\\\/\",\"name\":\"Centre for Computational Personalized Medicine\",\"description\":\"Sano \u2013 Centre for Computational Medicine\",\"publisher\":{\"@id\":\"https:\\\/\\\/sano.science\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/sano.science\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/sano.science\\\/#organization\",\"name\":\"Sano \u2013 Centre for Computational Medicine\",\"alternateName\":\"Sano\",\"url\":\"https:\\\/\\\/sano.science\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/sano.science\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/sano.science\\\/wp-content\\\/uploads\\\/2024\\\/05\\\/logo_sano_podstawowe.png\",\"contentUrl\":\"https:\\\/\\\/sano.science\\\/wp-content\\\/uploads\\\/2024\\\/05\\\/logo_sano_podstawowe.png\",\"width\":700,\"height\":265,\"caption\":\"Sano \u2013 Centre for Computational Medicine\"},\"image\":{\"@id\":\"https:\\\/\\\/sano.science\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/sano.science\\\/\",\"https:\\\/\\\/x.com\\\/sanoscience\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/sanoscience\\\/\",\"https:\\\/\\\/www.youtube.com\\\/channel\\\/UCDZ_8TcjMWUG2ZcgKKgfpwQ\",\"https:\\\/\\\/bsky.app\\\/profile\\\/sanoscience.bsky.social\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/sano.science\\\/#\\\/schema\\\/person\\\/561b69b48b6a8f7904aed06df5d03c98\",\"name\":\"Sano\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/3349c3947cb21b81e9745a6442fe4ec43961bb077198eb416cbdc66ec45ea97c?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/3349c3947cb21b81e9745a6442fe4ec43961bb077198eb416cbdc66ec45ea97c?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/3349c3947cb21b81e9745a6442fe4ec43961bb077198eb416cbdc66ec45ea97c?s=96&d=mm&r=g\",\"caption\":\"Sano\"},\"url\":\"https:\\\/\\\/sano.science\\\/author\\\/s15to\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Teaching\u00a0AI to\u00a0fill in the gaps - Centre for Computational Personalized Medicine","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/","og_locale":"en_US","og_type":"article","og_title":"Teaching\u00a0AI to\u00a0fill in the gaps","og_description":"At the Health AI &amp; CyberSec Summit in Warsaw, Sano\u2019s Wojciech Szyma\u0144ski showed how to make 3D medical image reconstruction more robust \u2014 and introduced Defect API, a way to train models without hand-labelled data. On 16 June, Sano researcher Wojciech Szyma\u0144ski took the stage at the Health AI &amp; CyberSec Summit (HAICS) in Warsaw, [&hellip;]","og_url":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/","og_site_name":"Centre for Computational Personalized Medicine","article_publisher":"https:\/\/www.facebook.com\/sano.science\/","article_published_time":"2026-08-04T18:20:12+00:00","article_modified_time":"2026-08-18T18:42:05+00:00","og_image":[{"width":1200,"height":1200,"url":"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg","type":"image\/jpeg"}],"author":"Sano","twitter_card":"summary_large_image","twitter_creator":"@sanoscience","twitter_site":"@sanoscience","twitter_misc":{"Written by":"Sano","Est. reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#article","isPartOf":{"@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/"},"author":{"name":"Sano","@id":"https:\/\/sano.science\/#\/schema\/person\/561b69b48b6a8f7904aed06df5d03c98"},"headline":"Teaching\u00a0AI to\u00a0fill in the gaps","datePublished":"2026-08-04T18:20:12+00:00","dateModified":"2026-08-18T18:42:05+00:00","mainEntityOfPage":{"@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/"},"wordCount":521,"publisher":{"@id":"https:\/\/sano.science\/#organization"},"image":{"@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#primaryimage"},"thumbnailUrl":"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg","inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/","url":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/","name":"Teaching\u00a0AI to\u00a0fill in the gaps - Centre for Computational Personalized Medicine","isPartOf":{"@id":"https:\/\/sano.science\/#website"},"primaryImageOfPage":{"@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#primaryimage"},"image":{"@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#primaryimage"},"thumbnailUrl":"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg","datePublished":"2026-08-04T18:20:12+00:00","dateModified":"2026-08-18T18:42:05+00:00","breadcrumb":{"@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#primaryimage","url":"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg","contentUrl":"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1.jpg","width":1200,"height":1200},{"@type":"BreadcrumbList","@id":"https:\/\/sano.science\/teaching-ai-to-fill-in-the-gaps\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/sano.science\/"},{"@type":"ListItem","position":2,"name":"Teaching\u00a0AI to\u00a0fill in the gaps"}]},{"@type":"WebSite","@id":"https:\/\/sano.science\/#website","url":"https:\/\/sano.science\/","name":"Centre for Computational Personalized Medicine","description":"Sano \u2013 Centre for Computational Medicine","publisher":{"@id":"https:\/\/sano.science\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/sano.science\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/sano.science\/#organization","name":"Sano \u2013 Centre for Computational Medicine","alternateName":"Sano","url":"https:\/\/sano.science\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/sano.science\/#\/schema\/logo\/image\/","url":"https:\/\/sano.science\/wp-content\/uploads\/2024\/05\/logo_sano_podstawowe.png","contentUrl":"https:\/\/sano.science\/wp-content\/uploads\/2024\/05\/logo_sano_podstawowe.png","width":700,"height":265,"caption":"Sano \u2013 Centre for Computational Medicine"},"image":{"@id":"https:\/\/sano.science\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/sano.science\/","https:\/\/x.com\/sanoscience","https:\/\/www.linkedin.com\/company\/sanoscience\/","https:\/\/www.youtube.com\/channel\/UCDZ_8TcjMWUG2ZcgKKgfpwQ","https:\/\/bsky.app\/profile\/sanoscience.bsky.social"]},{"@type":"Person","@id":"https:\/\/sano.science\/#\/schema\/person\/561b69b48b6a8f7904aed06df5d03c98","name":"Sano","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/3349c3947cb21b81e9745a6442fe4ec43961bb077198eb416cbdc66ec45ea97c?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/3349c3947cb21b81e9745a6442fe4ec43961bb077198eb416cbdc66ec45ea97c?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/3349c3947cb21b81e9745a6442fe4ec43961bb077198eb416cbdc66ec45ea97c?s=96&d=mm&r=g","caption":"Sano"},"url":"https:\/\/sano.science\/author\/s15to\/"}]}},"acf":[],"gutenberg_blocks":[{"blockName":"custom-styles","attrs":{"styles":""}},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-T7VCQ9","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">At the Health AI &amp; CyberSec Summit in Warsaw, Sano\u2019s Wojciech Szyma\u0144ski showed how to make 3D medical image reconstruction more robust \u2014 and introduced Defect API, a way to train models without hand-labelled data.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">At the Health AI &amp; CyberSec Summit in Warsaw, Sano\u2019s Wojciech Szyma\u0144ski showed how to make 3D medical image reconstruction more robust \u2014 and introduced Defect API, a way to train models without hand-labelled data.<\/p>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-vD5oV1","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">On 16 June, Sano researcher Wojciech Szyma\u0144ski took the stage at the Health AI &amp; CyberSec Summit (HAICS) in Warsaw, a meeting held at the Copernicus Science Centre that brought together two communities not often in the same room: healthcare cybersecurity and healthcare AI. Drawing on the Hack Summit and Data Science Summit, the programme spanned data protection, medical imaging and the growing role of artificial intelligence across the health and pharmaceutical sector.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">On 16 June, Sano researcher Wojciech Szyma\u0144ski took the stage at the Health AI &amp; CyberSec Summit (HAICS) in Warsaw, a meeting held at the Copernicus Science Centre that brought together two communities not often in the same room: healthcare cybersecurity and healthcare AI. Drawing on the Hack Summit and Data Science Summit, the programme spanned data protection, medical imaging and the growing role of artificial intelligence across the health and pharmaceutical sector.<\/p>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-mjgb61","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">Wojciech presented part of his PhD research on a stubborn problem in medical imaging: how to make the reconstruction of 3D anatomical structures more reliable. Scans are rarely perfect. Parts of an organ or vessel can be missing, noisy or distorted, and a model trained to rebuild them usually needs large collections of examples that a person has painstakingly labelled by hand. Preparing those labels is slow and costly, and it limits how far such models can scale.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">Wojciech presented part of his PhD research on a stubborn problem in medical imaging: how to make the reconstruction of 3D anatomical structures more reliable. Scans are rarely perfect. Parts of an organ or vessel can be missing, noisy or distorted, and a model trained to rebuild them usually needs large collections of examples that a person has painstakingly labelled by hand. Preparing those labels is slow and costly, and it limits how far such models can scale.<\/p>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/columns","attrs":{"verticalAlignment":null,"epStylingOptions":{"epCustomColumnsResponsiveEnabled":true,"epCustomColumnsReverseResponsiveEnabled":true,"epCustomColumnsSpacingResponsiveEnabled":true,"epCustomColumns":{"target":"wp-block-column","responsive":true,"hover":false,"options":[{"custom":true,"control":"ColumnToggle"},{"label":"Responsive Columns","control":"Range","attribute":"epCustomColumns","defaults":{"tablet":"2","mobile":"1"},"css":"flex-basis","customValue":"calc(( 100% - ({{Range:epCustomColumnsSpacing:auto:0px}} * ({{value}} - 1))) \/ {{value}}) !important","props":{"max":6,"min":1,"supportedUnits":[]},"condition":{"query":[{"field":"attributes.className","compare":"IN","value":"ep-custom-column"},[{"relation":"OR","query":[{"field":"viewport","compare":"EQUAL","value":"tablet"},{"field":"viewport","compare":"EQUAL","value":"mobile"}]}]]}}]},"epCustomColumnsReverse":{"target":"","responsive":true,"hover":false,"options":[{"label":"Columns Order","control":"ButtonGroup","attribute":"epCustomColumnsReverse","css":"flex-direction","customValue":"{{value}}","condition":{"relation":"AND","query":[{"field":"attributes.className","compare":"IN","value":"ep-custom-column"},[{"relation":"OR","query":[[{"relation":"AND","query":[{"field":"attributes.epCustomColumns.mobile","controlledBy":"Range","compare":"IN","value":"1"},{"field":"viewport","compare":"EQUAL","value":"mobile"}]}],[{"relation":"AND","query":[{"field":"attributes.epCustomColumns.tablet","controlledBy":"Range","compare":"IN","value":"1"},{"field":"viewport","compare":"EQUAL","value":"tablet"}]}]]}]]},"defaults":{"tablet":"column","mobile":"column"},"props":{"layout":"stacked","options":[{"label":"Default","value":"column"},{"label":"Reverse","value":"column-reverse"}]}}]},"epCustomColumnsSpacing":{"target":"","responsive":true,"hover":false,"options":[{"label":"Columns Gap","control":"Range","attribute":"epCustomColumnsSpacing","defaults":{"desktop":"32px","tablet":"32px","mobile":"32px"},"css":"gap","props":{"max":100,"min":0,"supportedUnits":["px"]},"condition":{"query":[{"field":"attributes.className","compare":"IN","value":"ep-custom-column"}]}}]},"savedStyling":"","clientId":"649e1c60-72b3-4386-94e2-70832f1aa217"},"epCustomColumns":{"desktop":{"value":"","important":false,"unit":"%"},"tablet":{"value":2,"important":false,"unit":""},"tabletModified":true,"mobile":{"value":1,"important":false,"unit":""},"mobileModified":true},"epCustomColumnsSpacing":{"desktop":{"value":32,"important":false,"unit":"px"},"tablet":{"value":32,"important":false,"unit":"px"},"tabletModified":true,"mobile":{"value":32,"important":false,"unit":"px"},"mobileModified":true},"epCustomColumnsReverse":{"desktop":"","tablet":"column","tabletModified":true,"mobile":"column","mobileModified":true},"epAnimationGeneratedClass":"edplus_anim-8RP2Xp","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[{"blockName":"core\/column","attrs":{"epAnimationGeneratedClass":"edplus_anim-qogEx6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[{"blockName":"core\/image","attrs":{"id":32638,"sizeSlug":"large","linkDestination":"none","epAnimationGeneratedClass":"edplus_anim-jNTDn6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<figure class=\"wp-block-image size-large eplus-wrapper\"><img src=\"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-1024x1024.jpg\" alt=\"\" class=\"wp-image-32638\"\/><\/figure>\n","innerContent":["\n<figure class=\"wp-block-image size-large eplus-wrapper\"><img src=\"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-1024x1024.jpg\" alt=\"\" class=\"wp-image-32638\"\/><\/figure>\n"]}],"innerHTML":"\n<div class=\"wp-block-column eplus-wrapper\"><\/div>\n","innerContent":["\n<div class=\"wp-block-column eplus-wrapper\">",null,"<\/div>\n"]},{"blockName":"core\/column","attrs":{"verticalAlignment":"center","epAnimationGeneratedClass":"edplus_anim-8UKunF","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-GHW0T9","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">Wojciech presented part of his PhD research on a stubborn problem in medical imaging: how to make the reconstruction of 3D anatomical structures more reliable. Scans are rarely perfect. Parts of an organ or vessel can be missing, noisy or distorted, and a model trained to rebuild them usually needs large collections of examples that a person has painstakingly labelled by hand. Preparing those labels is slow and costly, and it limits how far such models can scale.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">Wojciech presented part of his PhD research on a stubborn problem in medical imaging: how to make the reconstruction of 3D anatomical structures more reliable. Scans are rarely perfect. Parts of an organ or vessel can be missing, noisy or distorted, and a model trained to rebuild them usually needs large collections of examples that a person has painstakingly labelled by hand. Preparing those labels is slow and costly, and it limits how far such models can scale.<\/p>\n"]}],"innerHTML":"\n<div class=\"wp-block-column is-vertically-aligned-center eplus-wrapper\"><\/div>\n","innerContent":["\n<div class=\"wp-block-column is-vertically-aligned-center eplus-wrapper\">",null,"<\/div>\n"]}],"innerHTML":"<div class=\"wp-block-columns eplus-wrapper eplus-styles-uid-ad04af\">\n\n<\/div>","innerContent":["\n<div class=\"wp-block-columns eplus-wrapper\">",null,"\n\n",null,"<\/div>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-JhEFcL","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">His answer is a framework called Defect API. Instead of relying on hand-prepared data, it generates synthetic structural defects during training: it deliberately damages complete examples in configurable, realistic ways, then asks the model to restore them. Because the correct answer is simply the undamaged original, the model can learn on its own, an approach known as self-supervised learning, with nobody labelling anything. The framework works with both voxel grids and point clouds, two common ways of representing 3D shapes, and it has been connected to well-established architectures including nnU-Net and Point Transformer V3.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">His answer is a framework called Defect API. Instead of relying on hand-prepared data, it generates synthetic structural defects during training: it deliberately damages complete examples in configurable, realistic ways, then asks the model to restore them. Because the correct answer is simply the undamaged original, the model can learn on its own, an approach known as self-supervised learning, with nobody labelling anything. The framework works with both voxel grids and point clouds, two common ways of representing 3D shapes, and it has been connected to well-established architectures including nnU-Net and Point Transformer V3.<\/p>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-n3Ad83","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">Tested on large, varied medical datasets, the method pointed to a clear lesson: the diversity and realism of the training defects strongly shape how robust the model becomes and how well it generalises to data it has not seen before. Put plainly, the quality of the \"damage\" a model learns from can matter as much as the model itself.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">Tested on large, varied medical datasets, the method pointed to a clear lesson: the diversity and realism of the training defects strongly shape how robust the model becomes and how well it generalises to data it has not seen before. Put plainly, the quality of the \"damage\" a model learns from can matter as much as the model itself.<\/p>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-OuFt8X","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">Wojciech closed with a live demonstration of the full pipeline, from training the model and tracking its progress in TensorBoard to an interactive 3D view of the reconstructed anatomy. Seeing the whole chain run end to end, rather than a set of final numbers on a slide, made the method easy to grasp for an audience drawn from very different backgrounds.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">Wojciech closed with a live demonstration of the full pipeline, from training the model and tracking its progress in TensorBoard to an interactive 3D view of the reconstructed anatomy. Seeing the whole chain run end to end, rather than a set of final numbers on a slide, made the method easy to grasp for an audience drawn from very different backgrounds.<\/p>\n"]},{"blockName":"core\/spacer","attrs":{"height":"30px","epAnimationGeneratedClass":"edplus_anim-BT6je6","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n","innerContent":["\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer eplus-wrapper\"><\/div>\n"]},{"blockName":"core\/paragraph","attrs":{"epAnimationGeneratedClass":"edplus_anim-CjhiLz","epGeneratedClass":"eplus-wrapper"},"innerBlocks":[],"innerHTML":"\n<p class=\" eplus-wrapper\">For Sano, the talk is a good example of how careful methodological work in AI feeds directly into the tools that clinicians and researchers may one day rely on. We congratulate Wojciech on the presentation.<\/p>\n","innerContent":["\n<p class=\" eplus-wrapper\">For Sano, the talk is a good example of how careful methodological work in AI feeds directly into the tools that clinicians and researchers may one day rely on. We congratulate Wojciech on the presentation.<\/p>\n"]}],"meta_data":{"has_thumbnail_pattern":false,"share_on_social_media":{"has_social_media":false}},"featured_image":{"url":"https:\/\/sano.science\/wp-content\/uploads\/2026\/08\/HAICS_1-1024x1024.jpg"},"main_category":{"name":"Uncategorized"},"prev_page":{"slug":"adam-sulek-at-thursday-gathering-263-ai-reshaping-daily-life"},"next_page":{"slug":"sanos-legal-team-at-icail-2026"},"_links":{"self":[{"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/posts\/32637","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/comments?post=32637"}],"version-history":[{"count":9,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/posts\/32637\/revisions"}],"predecessor-version":[{"id":32649,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/posts\/32637\/revisions\/32649"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/media\/32638"}],"wp:attachment":[{"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/media?parent=32637"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/categories?post=32637"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sano.science\/index.php\/wp-json\/wp\/v2\/tags?post=32637"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}