{"id":122,"date":"2021-10-27T14:20:42","date_gmt":"2021-10-27T12:20:42","guid":{"rendered":"https:\/\/sc21.icm.edu.pl\/?page_id=122"},"modified":"2021-12-14T13:20:28","modified_gmt":"2021-12-14T12:20:28","slug":"ai-in-biomedical-imaging","status":"publish","type":"page","link":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/","title":{"rendered":"AI in biomedical imaging"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-page\" data-elementor-id=\"122\" class=\"elementor elementor-122\">\n\t\t\t\t\t\t<section class=\"has_eae_slider elementor-section elementor-top-section elementor-element elementor-element-4b3f834e elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-eae-slider=\"95335\" data-id=\"4b3f834e\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"has_eae_slider elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-46eef378\" data-eae-slider=\"77652\" data-id=\"46eef378\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-436d88a1 elementor-position-right elementor-vertical-align-middle footer-menu-items elementor-widget elementor-widget-image-box\" data-id=\"436d88a1\" data-element_type=\"widget\" data-widget_type=\"image-box.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-image-box-wrapper\"><figure class=\"elementor-image-box-img\"><img fetchpriority=\"high\" decoding=\"async\" width=\"400\" height=\"400\" src=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png\" class=\"attachment-full size-full wp-image-3288\" alt=\"\" srcset=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png 400w, https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski-300x300.png 300w, https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski-150x150.png 150w\" sizes=\"(max-width: 400px) 100vw, 400px\" \/><\/figure><div class=\"elementor-image-box-content\"><h3 class=\"elementor-image-box-title\">Norbert Kapi\u0144ski , PhD<\/h3><p class=\"elementor-image-box-description\"><i aria-hidden=\"true\" class=\"far fa-arrow-alt-circle-right contact-arrow\"><\/i><a href=\"#1\">Email<\/a>\n<br>\n<i aria-hidden=\"true\" class=\"far fa-arrow-alt-circle-right contact-arrow\"><\/i><a href=\"https:\/\/www.linkedin.com\/in\/nkn187\/\" target=\"_blank\">LinkedIn<\/a><\/p><\/div><\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-45e0eea6 linear-wipe elementor-widget elementor-widget-heading\" data-id=\"45e0eea6\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">AI in biomedical imaging\n<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4aa5eec elementor-widget elementor-widget-heading\" data-id=\"4aa5eec\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">DL &amp; ML in medical imaging and radiology<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-128acc0 elementor-widget elementor-widget-text-editor\" data-id=\"128acc0\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Artificial Intelligence in medical imaging<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-45fd8c3 elementor-widget elementor-widget-text-editor\" data-id=\"45fd8c3\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Development and use of deep learning (DL) techniques and machine learning (ML) in medical imaging and radiology. From image enhancement, through segmentation, detection, diagnostics automation, decision support, up to structuring reporting, optimization of radiological workflows (e.g. triage),<br \/>analysis automation and study report generation. Main fields of interest: MRI, CT, Ultrasound, orthopaedics \/ lower limb, chest \/ lungs.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-bc129e2 elementor-widget elementor-widget-text-editor\" data-id=\"bc129e2\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tAutomatic diagnosis of the Achilles tendon in Magnetic Resonance Imaging\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-0a7aac8 elementor-widget elementor-widget-text-editor\" data-id=\"0a7aac8\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tThanks to the use of convolutional neural networks (CNNs) as well as machine learning and statistics methods, we developed tools to  automate the evaluation of the Achilles tendon in MR imaging. Based on the image data, numerical indicators are generated quantifying the condition of individual features of the imaged tissues. Doctors can use point diagnostic information, as well as monitor the healing process and observe the process against the statistical background.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2009f0a elementor-widget elementor-widget-text-editor\" data-id=\"2009f0a\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tComputed Tomography chest screening for early detection of health threats\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4fda380 elementor-widget elementor-widget-text-editor\" data-id=\"4fda380\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tComplex CNN-based models, including multiple autoencoders, GANs and Cycle-GANs, are often applied to solve radiological and clinical problems in terms of low data availability, low labelling quality or the need of explainability. Such networks allow us for explainable studies of chest screening examinations targeted at finding early life-threatening symptoms and assessment of death risk, with the clinically meaningful answers of the model.   \t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c3291ac elementor-widget elementor-widget-text-editor\" data-id=\"c3291ac\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tConvolutional Neural Networks explainability with advanced visualization\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-0f20842 elementor-widget elementor-widget-text-editor\" data-id=\"0f20842\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tDevelopment of visual analysis techniques for opening the \u201eblack box\u201d of deep Convolutional Neural Networks. Generic access to AI model data formats implemented in VisNow platform and combined with dedicated CNN structure visualization modules allow for monitoring of network parameters evolution in iterative training and neural activation paths during inference. Visual explanation of CNN structure leads to better understanding of the model and its application. In medical imaging it helps to understand the low and high level image structures contribution to model decisions. \t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"has_eae_slider elementor-section elementor-top-section elementor-element elementor-element-2eeacd96 elementor-hidden-mobile elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-eae-slider=\"67483\" data-id=\"2eeacd96\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"has_eae_slider elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-956fb6d\" data-eae-slider=\"24510\" data-id=\"956fb6d\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-7c0b3cf9 elementor-widget elementor-widget-text-editor\" data-id=\"7c0b3cf9\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>USE CASE:<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1ad1d50f elementor-widget elementor-widget-text-editor\" data-id=\"1ad1d50f\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tMortality risk prediction and pathology detection in chest medical imaging screening with deep learning techniques\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2d31ca16 elementor-widget elementor-widget-text-editor\" data-id=\"2d31ca16\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Executor: Interdisciplinary Centre for Mathematical and Computational Modelling, University of Warsaw (ICM)<br \/>HPC resources: Rysy (NVIDIA GPU cluster), Tetyda (Lustre storage)<br \/>Principal investigator:\u00a0\u00a0Norbert Kapi\u0144ski (PhD), <a class=\"c-link\" tabindex=\"-1\" href=\"mailto:n.kapinski@icm.edu.pl\" target=\"_blank\" rel=\"noopener noreferrer\" data-stringify-link=\"mailto:n.kapinski@icm.edu.pl\" data-sk=\"tooltip_parent\" aria-haspopup=\"menu\" aria-expanded=\"false\" data-remove-tab-index=\"true\">n.kapinski@icm.edu.pl<\/a><br \/>Project type: Scientific<br \/>Project status: Preliminary work<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-de520cc eae--hr-position-center eae--vr-position-middle elementor-widget elementor-widget-eae-thumbgallery\" data-id=\"de520cc\" data-element_type=\"widget\" data-settings=\"{&quot;thumb_slides_per_view&quot;:4,&quot;thumb_slides_per_view_tablet&quot;:3,&quot;thumb_slides_per_view_mobile&quot;:2,&quot;thumb_space_between&quot;:10,&quot;thumb_space_between_tablet&quot;:10,&quot;thumb_space_between_mobile&quot;:5,&quot;slider_space_between&quot;:{&quot;unit&quot;:&quot;px&quot;,&quot;size&quot;:10,&quot;sizes&quot;:[]},&quot;slider_space_between_tablet&quot;:{&quot;unit&quot;:&quot;px&quot;,&quot;size&quot;:15,&quot;sizes&quot;:[]},&quot;slider_space_between_mobile&quot;:{&quot;unit&quot;:&quot;px&quot;,&quot;size&quot;:10,&quot;sizes&quot;:[]}}\" data-widget_type=\"eae-thumbgallery.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\n\t\t<div class=\"eae-swiper-outer-wrapper eae-swiper\" data-swiper-settings=\"{&quot;breakpoints_value&quot;:{&quot;mobile&quot;:320,&quot;tablet&quot;:768,&quot;desktop&quot;:1025},&quot;thumbs&quot;:{&quot;spaceBetween&quot;:{&quot;mobile&quot;:5,&quot;tablet&quot;:10,&quot;default&quot;:10},&quot;slidesPerView&quot;:{&quot;mobile&quot;:2,&quot;tablet&quot;:3,&quot;default&quot;:4}},&quot;effect&quot;:&quot;slide&quot;,&quot;speed&quot;:500,&quot;autoplay&quot;:{&quot;duration&quot;:5000,&quot;reverseDirection&quot;:false,&quot;slider_direction&quot;:&quot;ltr&quot;,&quot;disableOnInteraction&quot;:true},&quot;pauseOnHover&quot;:&quot;yes&quot;,&quot;spaceBetween&quot;:{&quot;mobile&quot;:1,&quot;tablet&quot;:1,&quot;default&quot;:10},&quot;loop&quot;:&quot;yes&quot;,&quot;navigation&quot;:&quot;yes&quot;,&quot;clickable&quot;:false,&quot;keyboard&quot;:true}\">\n\t\t\t\t\t   \n\t\t\t<div class=\"eae-swiper-container swiper\">\n\t\t\t\t<div class=\"eae-swiper-wrapper swiper-wrapper slider_vertical_wrapper\">\n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-2aa981d eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/Pluca_schemat_ENG_black-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-070d4a8 eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka101-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-d1805ce eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka02-full.jpg\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-f087813 eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka104-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-4fa57b9 eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka102-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t \n\t\t\t\t<\/div>\n\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t<div class = \" eae-swiper-button eae-slider-nav-button  eae-swiper-button-prev\">\n\t\t\t\t\t\t\t<i class=\"fa fa-angle-left\"><\/i>\t\t\t\t\t\t<\/div>\t\n\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class = \"eae-swiper-button eae-slider-nav-button eae-swiper-button-next\">\n\t\t\t\t\t\t\t<i class=\"fa fa-angle-right\"><\/i>\t\t\t\t\t\t<\/div>\t\n\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\n\t\t\t<\/div>\t\n\t\t\n\t\t\t<div class=\"eae-thumb-container swiper eae-gallery-thumbs eae-thumb-horizontal-bottom\">\n\t\t\t\t<div class=\"eae-thumb-wrapper swiper-wrapper\">\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-2aa981d eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/Pluca_schemat_ENG_black.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-070d4a8 eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka101-e1636964737443.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-d1805ce eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka02.jpg);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-f087813 eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka104.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-4fa57b9 eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka102.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t \n\t\t\t\t<\/div>\n\n\t\t\t\t\n\t\t\t<\/div>\n\n\t\t\t\t<\/div> \n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-38def71 row-in-infini elementor-tabs-view-horizontal elementor-widget elementor-widget-tabs\" data-id=\"38def71\" data-element_type=\"widget\" data-widget_type=\"tabs.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<div class=\"elementor-tabs\">\n\t\t\t<div class=\"elementor-tabs-wrapper\" role=\"tablist\" >\n\t\t\t\t\t\t\t\t\t<div id=\"elementor-tab-title-5961\" class=\"elementor-tab-title elementor-tab-desktop-title\" aria-selected=\"true\" data-tab=\"1\" role=\"tab\" tabindex=\"0\" aria-controls=\"elementor-tab-content-5961\" aria-expanded=\"false\">What problem the project tackles?<\/div>\n\t\t\t\t\t\t\t\t\t<div id=\"elementor-tab-title-5962\" class=\"elementor-tab-title elementor-tab-desktop-title\" aria-selected=\"false\" data-tab=\"2\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-5962\" aria-expanded=\"false\">Project goal and main tasks<\/div>\n\t\t\t\t\t\t\t\t\t<div id=\"elementor-tab-title-5963\" class=\"elementor-tab-title elementor-tab-desktop-title\" aria-selected=\"false\" data-tab=\"3\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-5963\" aria-expanded=\"false\">HPC infrrastructure usage<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t\t<div class=\"elementor-tabs-content-wrapper\" role=\"tablist\" aria-orientation=\"vertical\">\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-tab-title elementor-tab-mobile-title\" aria-selected=\"true\" data-tab=\"1\" role=\"tab\" tabindex=\"0\" aria-controls=\"elementor-tab-content-5961\" aria-expanded=\"false\">What problem the project tackles?<\/div>\n\t\t\t\t\t<div id=\"elementor-tab-content-5961\" class=\"elementor-tab-content elementor-clearfix\" data-tab=\"1\" role=\"tabpanel\" aria-labelledby=\"elementor-tab-title-5961\" tabindex=\"0\" hidden=\"false\"><p>The project uses Artificial Intelligence techniques, including deep machine learning methods, to analyze medical X-ray images and low-dose Computed Tomography from screening imaging for the early detection of pathological changes and prediction of the risk of death.<\/p><\/div>\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-tab-title elementor-tab-mobile-title\" aria-selected=\"false\" data-tab=\"2\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-5962\" aria-expanded=\"false\">Project goal and main tasks<\/div>\n\t\t\t\t\t<div id=\"elementor-tab-content-5962\" class=\"elementor-tab-content elementor-clearfix\" data-tab=\"2\" role=\"tabpanel\" aria-labelledby=\"elementor-tab-title-5962\" tabindex=\"0\" hidden=\"hidden\"><p>The aim of the project is to show that with the help of the Artificial Intelligence explanatory techniques used for the analysis of medical images\u00a0(Trustworthy AI), it is\u00a0possible to detect pathological changes early and assess the risk of death in screening tests, while providing clinically viable explanation for system output. The main tasks of the project include the development and validation of\u00a0artificial intelligence models implementing the above tasks, and the development of explainability methods to understand the results of the model, in particular in terms of the impact\u00a0of low-level and high-level image information.<\/p><\/div>\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-tab-title elementor-tab-mobile-title\" aria-selected=\"false\" data-tab=\"3\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-5963\" aria-expanded=\"false\">HPC infrrastructure usage<\/div>\n\t\t\t\t\t<div id=\"elementor-tab-content-5963\" class=\"elementor-tab-content elementor-clearfix\" data-tab=\"3\" role=\"tabpanel\" aria-labelledby=\"elementor-tab-title-5963\" tabindex=\"0\" hidden=\"hidden\"><p>The basic computing technology for deep machine learning methods, in particular in the field of convolutional neural networks (CNNs), are tensor\u00a0calculations. One of the most efficient architectures performing tensor operations is the GPU. Due to the number and size of data, model\u00a0sizes, and the need for iterative multi-epoch computations, supercomputing-class solutions best address the needs, providing computing power of\u00a0GPUs, large resources of host and GPU memory, as well as quick access to data storage resources and metadata databases.<br \/><br \/>Artificial Intelligence requires a lot of data sets and metadata (e.g. labels) for training and validation. Often in the order of tens or even hundreds of\u00a0thousands of images and related records. In case of medical imaging data derived from 3D imaging (eg Computed Tomography), a single data sample is often\u00a0the entire imaging stack (the so-called series), with a size of 100MB (eg. 200 2D slices 512x512x2B each). The training set of tens of thousands data samples often requires\u00a0tens of TB, and the process of training the model requires multiple reads of the whole set. Moreover, metadata, and in particular labels, are often\u00a0assigned at the level of single slices or individual pixels of the study, what requires adequate database resources to efficiently store and search metadata\u00a0resources in counts several orders of magnitude larger than the counts of the studies.<br \/><br \/>The task of chest screening analysis undertaken in the Project was based on data resources covering 90,000 Computed Tomography image series (including CDAS NLST study database). Due to dataset size\u00a0and model complexity, solution to the following problems was required: efficient metadata management at the level of single images\u00a0(18 million) for experiment planning, efficient access to data sets, efficient calculations, and large GPU memory for the model footprint and the training batch.<br \/><br \/>A dedicated database solution was prepared, integrating labeling database with the medical image storage system (PACS) for metadata management\u00a0and image data addressing, based on fast data resources (SSD \/\u00a0NVME) and in-memory databases (IMDB). The &#8222;Rysy&#8221; cluster based on NVIDIA V100 32GB GPU cards was used for the computations &#8211; ensuring high efficiency of\u00a0calculations and large memory resources for the model. The calculations were performed in Python programming language with the TensorFlow environment. Due to high I\/O intensity\u00a0(high data reading ratio in relation to the calculation time) and the large size of the training and validation set, it was necessary to provide\u00a0hierarchical\u00a0data access model for the GPU computing system. The data was stored entirely in Lustre file system (&#8222;Tetyda&#8221;) and asynchronously allocated to fast\u00a0local resources (SSD \/ NVME) of the computing cluster.<\/p><p>The use of HPC infrastructure:<\/p><ul><li>allowed training models with greater complexity at higher;<\/li><li>batch-size values;<br \/>reduced the metadata database response time several orders of magnitude;<\/li><li>reduced data access time several orders of magnitude.<\/li><\/ul><\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"has_eae_slider elementor-section elementor-top-section elementor-element elementor-element-37dd4e3 elementor-hidden-mobile elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-eae-slider=\"96601\" data-id=\"37dd4e3\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"has_eae_slider elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-440b740c\" data-eae-slider=\"43017\" data-id=\"440b740c\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-2b3998f8 elementor-widget elementor-widget-text-editor\" data-id=\"2b3998f8\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Papers<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-37486dfe elementor-align-left elementor-icon-list--layout-traditional elementor-list-item-link-full_width elementor-widget elementor-widget-icon-list\" data-id=\"37486dfe\" data-element_type=\"widget\" data-widget_type=\"icon-list.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<ul class=\"elementor-icon-list-items\">\n\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/arxiv.org\/abs\/1909.05687\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Late fusion of deep learning and hand-crafted features for Achilles tendon healing monitoring<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/link.springer.com\/chapter\/10.1007\/978-3-030-32875-7_8\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Monitoring Achilles Tendon Healing Progress in Ultrasound Imaging with Convolutional Neural Networks<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/Monitoring%20of%20the%20Achilles%20tendon%20healing%20process.pdf\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Monitoring of the Achilles tendon healing process: can artificial intelligence be helpful?<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/arxiv.org\/abs\/1806.05091\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Estimating Achilles tendon healing progress with convolutional neural networks<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"has_eae_slider elementor-section elementor-top-section elementor-element elementor-element-1da7a9c elementor-hidden-desktop elementor-hidden-tablet elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-eae-slider=\"29333\" data-id=\"1da7a9c\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"has_eae_slider elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-070a508\" data-eae-slider=\"61349\" data-id=\"070a508\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-96a8a71 elementor-widget elementor-widget-text-editor\" data-id=\"96a8a71\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>USE CASE:<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3ec3e75 elementor-widget elementor-widget-text-editor\" data-id=\"3ec3e75\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\tMortality risk prediction and pathology detection in chest medical imaging screening with deep learning techniques\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-e6aafc9 elementor-widget elementor-widget-text-editor\" data-id=\"e6aafc9\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Executor: Interdisciplinary Centre for Mathematical and Computational Modelling, University of Warsaw (ICM)<br \/>HPC resources: Rysy (NVIDIA GPU cluster), Tetyda (Lustre storage)<br \/>Principal investigator:\u00a0\u00a0Norbert Kapi\u0144ski (PhD), <a class=\"c-link\" tabindex=\"-1\" href=\"mailto:n.kapinski@icm.edu.pl\" target=\"_blank\" rel=\"noopener noreferrer\" data-stringify-link=\"mailto:n.kapinski@icm.edu.pl\" data-sk=\"tooltip_parent\" aria-haspopup=\"menu\" aria-expanded=\"false\" data-remove-tab-index=\"true\">n.kapinski@icm.edu.pl<\/a><br \/>Project type: Scientific<br \/>Project status: Preliminary work<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-56d92b1 eae--hr-position-center eae--vr-position-middle elementor-widget elementor-widget-eae-thumbgallery\" data-id=\"56d92b1\" data-element_type=\"widget\" data-settings=\"{&quot;thumb_slides_per_view&quot;:4,&quot;thumb_slides_per_view_tablet&quot;:3,&quot;thumb_slides_per_view_mobile&quot;:2,&quot;thumb_space_between&quot;:10,&quot;thumb_space_between_tablet&quot;:10,&quot;thumb_space_between_mobile&quot;:5,&quot;slider_space_between&quot;:{&quot;unit&quot;:&quot;px&quot;,&quot;size&quot;:10,&quot;sizes&quot;:[]},&quot;slider_space_between_tablet&quot;:{&quot;unit&quot;:&quot;px&quot;,&quot;size&quot;:15,&quot;sizes&quot;:[]},&quot;slider_space_between_mobile&quot;:{&quot;unit&quot;:&quot;px&quot;,&quot;size&quot;:10,&quot;sizes&quot;:[]}}\" data-widget_type=\"eae-thumbgallery.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\n\t\t<div class=\"eae-swiper-outer-wrapper eae-swiper\" data-swiper-settings=\"{&quot;breakpoints_value&quot;:{&quot;mobile&quot;:320,&quot;tablet&quot;:768,&quot;desktop&quot;:1025},&quot;thumbs&quot;:{&quot;spaceBetween&quot;:{&quot;mobile&quot;:5,&quot;tablet&quot;:10,&quot;default&quot;:10},&quot;slidesPerView&quot;:{&quot;mobile&quot;:2,&quot;tablet&quot;:3,&quot;default&quot;:4}},&quot;effect&quot;:&quot;slide&quot;,&quot;speed&quot;:500,&quot;autoplay&quot;:{&quot;duration&quot;:5000,&quot;reverseDirection&quot;:false,&quot;slider_direction&quot;:&quot;ltr&quot;,&quot;disableOnInteraction&quot;:true},&quot;pauseOnHover&quot;:&quot;yes&quot;,&quot;spaceBetween&quot;:{&quot;mobile&quot;:1,&quot;tablet&quot;:1,&quot;default&quot;:10},&quot;loop&quot;:&quot;yes&quot;,&quot;navigation&quot;:&quot;yes&quot;,&quot;clickable&quot;:false,&quot;keyboard&quot;:true}\">\n\t\t\t\t\t   \n\t\t\t<div class=\"eae-swiper-container swiper\">\n\t\t\t\t<div class=\"eae-swiper-wrapper swiper-wrapper slider_vertical_wrapper\">\n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-2aa981d eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/Pluca_schemat_ENG_black-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-070d4a8 eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka101-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-d1805ce eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka02-full.jpg\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-f087813 eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka104-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t\n\n\t\t\t\t\t<div class=\"elementor-repeater-item-4fa57b9 eae-swiper-slide swiper-slide\">\n\t\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t<a class=\"eae-slide-inner\" href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka102-full.png\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/a>\t\t\t\t\t\t\t<\/div>\n\t\t\t\t   \n\t\t\t\t\t \n\t\t\t\t<\/div>\n\t\t\t   \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t<div class = \" eae-swiper-button eae-slider-nav-button  eae-swiper-button-prev\">\n\t\t\t\t\t\t\t<i class=\"fa fa-angle-left\"><\/i>\t\t\t\t\t\t<\/div>\t\n\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class = \"eae-swiper-button eae-slider-nav-button eae-swiper-button-next\">\n\t\t\t\t\t\t\t<i class=\"fa fa-angle-right\"><\/i>\t\t\t\t\t\t<\/div>\t\n\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\n\t\t\t<\/div>\t\n\t\t\n\t\t\t<div class=\"eae-thumb-container swiper eae-gallery-thumbs eae-thumb-horizontal-bottom\">\n\t\t\t\t<div class=\"eae-thumb-wrapper swiper-wrapper\">\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-2aa981d eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/Pluca_schemat_ENG_black.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-070d4a8 eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka101-e1636964737443.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-d1805ce eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka02.jpg);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-f087813 eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka104.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t   \n\t\t\t\t\t\t<div class=\"elementor-repeater-item-4fa57b9 eae-thumb-slide swiper-slide\" style=\"background-image : url(https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/plucka102.png);\">\n\t\t\t\t\t\t\t\t<div class='eae-fit-aspect-ratio'><\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t \n\t\t\t\t<\/div>\n\n\t\t\t\t\n\t\t\t<\/div>\n\n\t\t\t\t<\/div> \n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5c92b5d row-in-infini elementor-tabs-view-horizontal elementor-widget elementor-widget-tabs\" data-id=\"5c92b5d\" data-element_type=\"widget\" data-widget_type=\"tabs.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<div class=\"elementor-tabs\">\n\t\t\t<div class=\"elementor-tabs-wrapper\" role=\"tablist\" >\n\t\t\t\t\t\t\t\t\t<div id=\"elementor-tab-title-9701\" class=\"elementor-tab-title elementor-tab-desktop-title\" aria-selected=\"true\" data-tab=\"1\" role=\"tab\" tabindex=\"0\" aria-controls=\"elementor-tab-content-9701\" aria-expanded=\"false\">What problem the project tackles?<\/div>\n\t\t\t\t\t\t\t\t\t<div id=\"elementor-tab-title-9702\" class=\"elementor-tab-title elementor-tab-desktop-title\" aria-selected=\"false\" data-tab=\"2\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-9702\" aria-expanded=\"false\">Project goal and main tasks<\/div>\n\t\t\t\t\t\t\t\t\t<div id=\"elementor-tab-title-9703\" class=\"elementor-tab-title elementor-tab-desktop-title\" aria-selected=\"false\" data-tab=\"3\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-9703\" aria-expanded=\"false\">HPC infrrastructure usage<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t\t<div class=\"elementor-tabs-content-wrapper\" role=\"tablist\" aria-orientation=\"vertical\">\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-tab-title elementor-tab-mobile-title\" aria-selected=\"true\" data-tab=\"1\" role=\"tab\" tabindex=\"0\" aria-controls=\"elementor-tab-content-9701\" aria-expanded=\"false\">What problem the project tackles?<\/div>\n\t\t\t\t\t<div id=\"elementor-tab-content-9701\" class=\"elementor-tab-content elementor-clearfix\" data-tab=\"1\" role=\"tabpanel\" aria-labelledby=\"elementor-tab-title-9701\" tabindex=\"0\" hidden=\"false\"><p>The project uses Artificial Intelligence techniques, including deep machine learning methods, to analyze medical X-ray images and low-dose Computed Tomography from screening imaging for the early detection of pathological changes and prediction of the risk of death.<\/p><\/div>\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-tab-title elementor-tab-mobile-title\" aria-selected=\"false\" data-tab=\"2\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-9702\" aria-expanded=\"false\">Project goal and main tasks<\/div>\n\t\t\t\t\t<div id=\"elementor-tab-content-9702\" class=\"elementor-tab-content elementor-clearfix\" data-tab=\"2\" role=\"tabpanel\" aria-labelledby=\"elementor-tab-title-9702\" tabindex=\"0\" hidden=\"hidden\"><p>The aim of the project is to show that with the help of the Artificial Intelligence explanatory techniques used for the analysis of medical images\u00a0(Trustworthy AI), it is\u00a0possible to detect pathological changes early and assess the risk of death in screening tests, while providing clinically viable explanation for system output. The main tasks of the project include the development and validation of\u00a0artificial intelligence models implementing the above tasks, and the development of explainability methods to understand the results of the model, in particular in terms of the impact\u00a0of low-level and high-level image information.<\/p><\/div>\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-tab-title elementor-tab-mobile-title\" aria-selected=\"false\" data-tab=\"3\" role=\"tab\" tabindex=\"-1\" aria-controls=\"elementor-tab-content-9703\" aria-expanded=\"false\">HPC infrrastructure usage<\/div>\n\t\t\t\t\t<div id=\"elementor-tab-content-9703\" class=\"elementor-tab-content elementor-clearfix\" data-tab=\"3\" role=\"tabpanel\" aria-labelledby=\"elementor-tab-title-9703\" tabindex=\"0\" hidden=\"hidden\"><p>The basic computing technology for deep machine learning methods, in particular in the field of convolutional neural networks (CNNs), are tensor\u00a0calculations. One of the most efficient architectures performing tensor operations is the GPU. Due to the number and size of data, model\u00a0sizes, and the need for iterative multi-epoch computations, supercomputing-class solutions best address the needs, providing computing power of\u00a0GPUs, large resources of host and GPU memory, as well as quick access to data storage resources and metadata databases.<br \/><br \/>Artificial Intelligence requires a lot of data sets and metadata (e.g. labels) for training and validation. Often in the order of tens or even hundreds of\u00a0thousands of images and related records. In case of medical imaging data derived from 3D imaging (eg Computed Tomography), a single data sample is often\u00a0the entire imaging stack (the so-called series), with a size of 100MB (eg. 200 2D slices 512x512x2B each). The training set of tens of thousands data samples often requires\u00a0tens of TB, and the process of training the model requires multiple reads of the whole set. Moreover, metadata, and in particular labels, are often\u00a0assigned at the level of single slices or individual pixels of the study, what requires adequate database resources to efficiently store and search metadata\u00a0resources in counts several orders of magnitude larger than the counts of the studies.<br \/><br \/>The task of chest screening analysis undertaken in the Project was based on data resources covering 90,000 Computed Tomography image series (including CDAS NLST study database). Due to dataset size\u00a0and model complexity, solution to the following problems was required: efficient metadata management at the level of single images\u00a0(18 million) for experiment planning, efficient access to data sets, efficient calculations, and large GPU memory for the model footprint and the training batch.<br \/><br \/>A dedicated database solution was prepared, integrating labeling database with the medical image storage system (PACS) for metadata management\u00a0and image data addressing, based on fast data resources (SSD \/\u00a0NVME) and in-memory databases (IMDB). The &#8222;Rysy&#8221; cluster based on NVIDIA V100 32GB GPU cards was used for the computations &#8211; ensuring high efficiency of\u00a0calculations and large memory resources for the model. The calculations were performed in Python programming language with the TensorFlow environment. Due to high I\/O intensity\u00a0(high data reading ratio in relation to the calculation time) and the large size of the training and validation set, it was necessary to provide\u00a0hierarchical\u00a0data access model for the GPU computing system. The data was stored entirely in Lustre file system (&#8222;Tetyda&#8221;) and asynchronously allocated to fast\u00a0local resources (SSD \/ NVME) of the computing cluster.<\/p><p>The use of HPC infrastructure:<\/p><ul><li>allowed training models with greater complexity at higher;<\/li><li>batch-size values;<br \/>reduced the metadata database response time several orders of magnitude;<\/li><li>reduced data access time several orders of magnitude.<\/li><\/ul><\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d0428d1 elementor-widget elementor-widget-text-editor\" data-id=\"d0428d1\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Papers<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-231b238 elementor-align-left elementor-icon-list--layout-traditional elementor-list-item-link-full_width elementor-widget elementor-widget-icon-list\" data-id=\"231b238\" data-element_type=\"widget\" data-widget_type=\"icon-list.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<ul class=\"elementor-icon-list-items\">\n\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/arxiv.org\/abs\/1909.05687\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Late fusion of deep learning and hand-crafted features for Achilles tendon healing monitoring<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/link.springer.com\/chapter\/10.1007\/978-3-030-32875-7_8\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Monitoring Achilles Tendon Healing Progress in Ultrasound Imaging with Convolutional Neural Networks<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/Monitoring%20of%20the%20Achilles%20tendon%20healing%20process.pdf\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Monitoring of the Achilles tendon healing process: can artificial intelligence be helpful?<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/arxiv.org\/abs\/1806.05091\" target=\"_blank\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<i aria-hidden=\"true\" class=\"fas fa-book\"><\/i>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\">Estimating Achilles tendon healing progress with convolutional neural networks<\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Norbert Kapi\u0144ski , PhD Email LinkedIn AI in biomedical imaging DL &#038; ML in medical imaging and radiology Artificial Intelligence in medical imaging Development and use of deep learning (DL) techniques and machine learning (ML) in medical imaging and radiology. From image enhancement, through segmentation, detection, diagnostics automation, decision support, up to structuring reporting, optimization [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-122","page","type-page","status-publish","hentry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.1.1 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>AI in biomedical imaging - SC21 ICM Univ of Warsaw #811<\/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:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/\" \/>\n<meta property=\"og:locale\" content=\"pl_PL\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI in biomedical imaging - SC21 ICM Univ of Warsaw #811\" \/>\n<meta property=\"og:description\" content=\"Norbert Kapi\u0144ski , PhD Email LinkedIn AI in biomedical imaging DL &#038; ML in medical imaging and radiology Artificial Intelligence in medical imaging Development and use of deep learning (DL) techniques and machine learning (ML) in medical imaging and radiology. From image enhancement, through segmentation, detection, diagnostics automation, decision support, up to structuring reporting, optimization [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/\" \/>\n<meta property=\"og:site_name\" content=\"SC21 ICM Univ of Warsaw #811\" \/>\n<meta property=\"article:modified_time\" content=\"2021-12-14T12:20:28+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Szacowany czas czytania\" \/>\n\t<meta name=\"twitter:data1\" content=\"8 minut\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/\",\"url\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/\",\"name\":\"AI in biomedical imaging - SC21 ICM Univ of Warsaw #811\",\"isPartOf\":{\"@id\":\"https:\/\/sc21.icm.edu.pl\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png\",\"datePublished\":\"2021-10-27T12:20:42+00:00\",\"dateModified\":\"2021-12-14T12:20:28+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#breadcrumb\"},\"inLanguage\":\"pl-PL\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"pl-PL\",\"@id\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#primaryimage\",\"url\":\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png\",\"contentUrl\":\"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png\",\"width\":400,\"height\":400},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Strona g\u0142\u00f3wna\",\"item\":\"https:\/\/sc21.icm.edu.pl\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI in biomedical imaging\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/sc21.icm.edu.pl\/#website\",\"url\":\"https:\/\/sc21.icm.edu.pl\/\",\"name\":\"SC21 ICM Univ of Warsaw #811\",\"description\":\"SC21 ICM Univ of Warsaw #811\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/sc21.icm.edu.pl\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"pl-PL\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"AI in biomedical imaging - SC21 ICM Univ of Warsaw #811","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:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/","og_locale":"pl_PL","og_type":"article","og_title":"AI in biomedical imaging - SC21 ICM Univ of Warsaw #811","og_description":"Norbert Kapi\u0144ski , PhD Email LinkedIn AI in biomedical imaging DL &#038; ML in medical imaging and radiology Artificial Intelligence in medical imaging Development and use of deep learning (DL) techniques and machine learning (ML) in medical imaging and radiology. From image enhancement, through segmentation, detection, diagnostics automation, decision support, up to structuring reporting, optimization [&hellip;]","og_url":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/","og_site_name":"SC21 ICM Univ of Warsaw #811","article_modified_time":"2021-12-14T12:20:28+00:00","og_image":[{"url":"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png","type":"","width":"","height":""}],"twitter_card":"summary_large_image","twitter_misc":{"Szacowany czas czytania":"8 minut"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/","url":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/","name":"AI in biomedical imaging - SC21 ICM Univ of Warsaw #811","isPartOf":{"@id":"https:\/\/sc21.icm.edu.pl\/#website"},"primaryImageOfPage":{"@id":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#primaryimage"},"image":{"@id":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#primaryimage"},"thumbnailUrl":"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png","datePublished":"2021-10-27T12:20:42+00:00","dateModified":"2021-12-14T12:20:28+00:00","breadcrumb":{"@id":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#breadcrumb"},"inLanguage":"pl-PL","potentialAction":[{"@type":"ReadAction","target":["https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/"]}]},{"@type":"ImageObject","inLanguage":"pl-PL","@id":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#primaryimage","url":"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png","contentUrl":"https:\/\/sc21.icm.edu.pl\/wp-content\/uploads\/2021\/11\/norbert-kapinski.png","width":400,"height":400},{"@type":"BreadcrumbList","@id":"https:\/\/sc21.icm.edu.pl\/index.php\/ai-in-biomedical-imaging\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Strona g\u0142\u00f3wna","item":"https:\/\/sc21.icm.edu.pl\/"},{"@type":"ListItem","position":2,"name":"AI in biomedical imaging"}]},{"@type":"WebSite","@id":"https:\/\/sc21.icm.edu.pl\/#website","url":"https:\/\/sc21.icm.edu.pl\/","name":"SC21 ICM Univ of Warsaw #811","description":"SC21 ICM Univ of Warsaw #811","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/sc21.icm.edu.pl\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"pl-PL"}]}},"_links":{"self":[{"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/pages\/122","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/comments?post=122"}],"version-history":[{"count":61,"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/pages\/122\/revisions"}],"predecessor-version":[{"id":4339,"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/pages\/122\/revisions\/4339"}],"wp:attachment":[{"href":"https:\/\/sc21.icm.edu.pl\/index.php\/wp-json\/wp\/v2\/media?parent=122"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}