{"id":574,"date":"2025-05-12T10:16:07","date_gmt":"2025-05-12T08:16:07","guid":{"rendered":"https:\/\/websites.fraunhofer.de\/ki-fortschrittszentrum\/?post_type=projekt&#038;p=574"},"modified":"2025-09-02T11:34:00","modified_gmt":"2025-09-02T09:34:00","slug":"information-extraction-from-medical-publications","status":"publish","type":"projekt","link":"https:\/\/www.ki-fortschrittszentrum.de\/en\/projekt\/informationsextraktion-aus-medizinischen-publikationen\/","title":{"rendered":"Information extraction from medical publications"},"content":{"rendered":"<div class=\"wp-block-stackable-columns stk-block-columns stk-block stk-86663cd stk-block-background stk--has-background-overlay\" data-block-id=\"86663cd\"><style>.stk-86663cd {background-image:url(https:\/\/www.ki-fortschrittszentrum.de\/wp-content\/uploads\/2025\/05\/Informtionsextraktion-aus-medizinischen-Publikationen-Titelbild.jpg) !important;min-height:520px !important;align-items:flex-end !important;padding-bottom:0px !important;margin-bottom:0px !important;display:flex !important;}<\/style><div class=\"stk-row stk-inner-blocks stk-block-content stk-content-align stk-86663cd-column\">\n<div class=\"wp-block-stackable-column stk-block-column stk-column stk-block stk-5a28863\" data-v=\"4\" data-block-id=\"5a28863\"><div class=\"stk-column-wrapper stk-block-column__content stk-container stk-5a28863-container stk--no-background stk--no-padding\"><div class=\"stk-block-content stk-inner-blocks stk-5a28863-inner-blocks\">\n<div class=\"wp-block-stackable-columns stk-block-columns stk-block stk-3e4895d\" data-block-id=\"3e4895d\"><div class=\"stk-row stk-inner-blocks stk-block-content stk-content-align stk-3e4895d-column\">\n<div class=\"wp-block-stackable-column stk-block-column stk-column stk-block stk-a094f1e\" data-v=\"4\" data-block-id=\"a094f1e\"><style>@media screen and (min-width:690px){.stk-a094f1e {flex:var(--stk-flex-grow, 1) 1 calc(30% - var(--stk-column-gap, 0px) * 2 \/ 3 ) !important;}}<\/style><div class=\"stk-column-wrapper stk-block-column__content stk-container stk-a094f1e-container stk--no-background stk--no-padding\"><div class=\"stk-block-content stk-inner-blocks stk-a094f1e-inner-blocks\"><\/div><\/div><\/div>\n\n\n\n<div class=\"wp-block-stackable-column stk-block-column stk-column stk-block stk-ba57710\" data-v=\"4\" data-block-id=\"ba57710\"><style>@media screen and (min-width:690px){.stk-ba57710 {flex:var(--stk-flex-grow, 1) 1 calc(30% - var(--stk-column-gap, 0px) * 2 \/ 3 ) !important;}}<\/style><div class=\"stk-column-wrapper stk-block-column__content stk-container stk-ba57710-container stk--no-background stk--no-padding\"><div class=\"stk-block-content stk-inner-blocks stk-ba57710-inner-blocks\"><\/div><\/div><\/div>\n\n\n\n<div class=\"wp-block-stackable-column stk-block-column stk-column stk-block stk-4764f29 stk-block-background\" data-v=\"4\" data-block-id=\"4764f29\"><style>.stk-4764f29 {align-self:flex-end !important;background-color:var(--theme-palette-color-8, #ffffff) !important;padding-top:0px !important;padding-right:0px !important;padding-bottom:0px !important;padding-left:0px !important;}.stk-4764f29-inner-blocks{justify-content:flex-end !important;}.stk-4764f29:before{background-color:var(--theme-palette-color-8, #ffffff) !important;}@media screen and (min-width:690px){.stk-4764f29 {flex:var(--stk-flex-grow, 1) 1 calc(40% - var(--stk-column-gap, 0px) * 2 \/ 3 ) !important;}}<\/style><div class=\"stk-column-wrapper stk-block-column__content stk-container stk-4764f29-container stk--no-background stk--no-padding\"><div class=\"stk--column-flex stk-block-content stk-inner-blocks stk-4764f29-inner-blocks\">\n<div class=\"wp-block-greenshift-blocks-container gspb_container gspb_container-gsbp-933ef16\" id=\"gspb_container-id-gsbp-933ef16\">\n<div class=\"wp-block-stackable-image stk-block-image has-text-align-left stk-block stk-1690316\" data-block-id=\"1690316\"><style>.stk-1690316 {margin-bottom:36px !important;}.stk-1690316 .stk-img-wrapper{width:50% !important;}<\/style><figure><span class=\"stk-img-wrapper stk-image--shape-stretch\"><img loading=\"lazy\" decoding=\"async\" class=\"stk-img\" src=\"https:\/\/www.ki-fortschrittszentrum.de\/wp-content\/uploads\/2025\/05\/KI-Fortschrittszentrum_349_QC_Informationsextraktion-aus-medizinischen-Publikationen_Logo.png\" width=\"50\" height=\"300\"\/><\/span><\/figure><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-97d4068\" data-block-id=\"97d4068\"><style>.stk-97d4068 {margin-bottom:12px !important;}.stk-97d4068 .stk-block-text__text{font-size:16px !important;line-height:1.4em !important;font-weight:500 !important;font-style:italic !important;font-family:Constantia, Lucida Bright, Lucidabright, \"Lucida Serif\", Lucida, \"DejaVu Serif\", \"Bitstream Vera Serif\", \"Liberation Serif\", Georgia, serif !important;}@media screen and (max-width:999px){.stk-97d4068 .stk-block-text__text{font-size:16px !important;}}<\/style><p class=\"stk-block-text__text\">Extracting information from medical publications is a demanding task that takes a lot of time. A tool that supports this task and helps us to significantly reduce the effort would be a great improvement.<\/p><\/div>\n\n\n\n<div class=\"wp-block-greenshift-blocks-container gspb_container gspb_container-gsbp-eafe0ac\" id=\"gspb_container-id-gsbp-eafe0ac\">\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-25dc0f8\" data-block-id=\"25dc0f8\"><style>.stk-25dc0f8 {padding-top:0px !important;padding-bottom:0px !important;margin-top:0px !important;margin-bottom:0px !important;}.stk-25dc0f8 .stk-block-text__text{font-size:15px !important;}@media screen and (max-width:999px){.stk-25dc0f8 .stk-block-text__text{font-size:15px !important;}}<\/style><p class=\"stk-block-text__text has-text-align-left\"><\/p><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-d634667\" data-block-id=\"d634667\"><style>.stk-d634667 {align-items:flex-start !important;padding-top:0px !important;padding-bottom:0px !important;margin-top:0px !important;margin-bottom:0px !important;display:flex !important;}.stk-d634667 .stk-block-text__text{font-size:15px !important;font-weight:200 !important;}@media screen and (max-width:999px){.stk-d634667 .stk-block-text__text{font-size:15px !important;}}<\/style><p class=\"stk-block-text__text has-text-align-left\">W\u00f6rwag Pharma<\/p><\/div>\n<\/div>\n<\/div>\n\n\n\n<div class=\"wp-block-greenshift-blocks-container gspb_container gspb_container-gsbp-1fa6a74\" id=\"gspb_container-id-gsbp-1fa6a74\">\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-be1fdb3\" data-block-id=\"be1fdb3\"><style>.stk-be1fdb3 {padding-top:0px !important;padding-bottom:12px !important;margin-top:0px !important;margin-bottom:0px !important;}.stk-be1fdb3 .stk-block-text__text{font-size:15px !important;color:#ffffff80 !important;}@media screen and (max-width:999px){.stk-be1fdb3 .stk-block-text__text{font-size:15px !important;}}<\/style><p class=\"stk-block-text__text has-text-color has-text-align-left\">Contact at the AI Innovation Center<\/p><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-24ffa5e\" data-block-id=\"24ffa5e\"><style>.stk-24ffa5e {padding-top:0px !important;padding-bottom:0px !important;margin-top:0px !important;margin-bottom:0px !important;}.stk-24ffa5e .stk-block-text__text{font-size:15px !important;color:#ffffff !important;}@media screen and (max-width:999px){.stk-24ffa5e .stk-block-text__text{font-size:15px !important;}}<\/style><p class=\"stk-block-text__text has-text-color has-text-align-left\">Janina Bierkandt<\/p><\/div>\n\n\n\n<div class=\"wp-block-stackable-button-group stk-block-button-group stk-block stk-af007b4\" data-block-id=\"af007b4\"><style>.stk-af007b4 {padding-top:0px !important;padding-right:0px !important;padding-bottom:0px !important;padding-left:0px !important;margin-top:0px !important;margin-right:0px !important;margin-bottom:0px !important;margin-left:0px !important;}<\/style><div class=\"stk-row stk-inner-blocks stk-block-content stk-button-group\">\n<div class=\"wp-block-stackable-button stk-block-button is-style-plain stk-block stk-1cf8bea\" data-block-id=\"1cf8bea\"><style>.stk-1cf8bea .stk-button{padding-top:0px !important;padding-right:0px !important;padding-bottom:0px !important;padding-left:0px !important;background:transparent !important;}.stk-1cf8bea .stk-button:hover:after{background:transparent !important;opacity:1 !important;}:where(.stk-hover-parent:hover,  .stk-hover-parent.stk--is-hovered) .stk-1cf8bea .stk-button:after{background:transparent !important;opacity:1 !important;}.stk-1cf8bea .stk-button__inner-text{font-size:15px !important;color:var(--theme-palette-color-8, #ffffff) !important;font-weight:200 !important;}@media screen and (max-width:999px){.stk-1cf8bea .stk-button__inner-text{font-size:15px !important;}}<\/style><a class=\"stk-link stk-button stk--hover-effect-darken\" href=\"mailto:janina.bierkandt@iao.fraunhofer.de\" title=\"janina.bierkandt@iao.fraunhofer.de\"><span class=\"has-text-color stk-button__inner-text\">janina.bierkandt@iao.fraunhofer.de<\/span><\/a><\/div>\n<\/div><\/div>\n<\/div>\n<\/div><\/div><\/div>\n<\/div><\/div>\n<\/div><\/div><\/div>\n<\/div><\/div>\n\n\n\n<div class=\"wp-block-stackable-spacer stk-block-spacer stk--no-padding stk-block stk-56a43b6\" data-block-id=\"56a43b6\"><\/div>\n\n\n\n<div class=\"wp-block-stackable-columns stk-block-columns stk-block stk-fc04401\" data-block-id=\"fc04401\"><style>.stk-fc04401 {padding-right:24px !important;padding-left:24px !important;}<\/style><div class=\"stk-row stk-inner-blocks stk-block-content stk-content-align stk-fc04401-column\">\n<div class=\"wp-block-stackable-column stk-block-column stk-column stk-block stk-a194320\" data-v=\"4\" data-block-id=\"a194320\"><div class=\"stk-column-wrapper stk-block-column__content stk-container stk-a194320-container stk--no-background stk--no-padding\"><div class=\"stk-block-content stk-inner-blocks stk-a194320-inner-blocks\"><div data-block=\"hook:1248\" class=\"alignfull\"><article id=\"post-1248\" class=\"post-1248\"><div class=\"entry-content is-layout-constrained\">\n<div class=\"wp-block-stackable-button-group stk-block-button-group stk-block stk-fcd1a8a\" data-block-id=\"fcd1a8a\"><style>.stk-fcd1a8a {margin-bottom:24px !important;}<\/style><div class=\"stk-row stk-inner-blocks stk-block-content stk-button-group\">\n<div class=\"wp-block-stackable-icon-button stk-block-icon-button stk-block stk-82470ba is-style-ghost\" data-block-id=\"82470ba\"><style>.stk-82470ba .stk-button{background:transparent !important;}.stk-82470ba .stk-button:hover{background:transparent !important;opacity:1 !important;}:where(.stk-hover-parent:hover,  .stk-hover-parent.stk--is-hovered) .stk-82470ba .stk-button:after{background:transparent !important;opacity:1 !important;}.stk-82470ba .stk-button:before{border-style:solid !important;border-color:var(--theme-palette-color-9, #264e5d) !important;}.stk-82470ba .stk-button .stk--inner-svg svg:last-child, .stk-82470ba .stk-button .stk--inner-svg svg:last-child :is(g, path, rect, polygon, ellipse){fill:var(--theme-palette-color-11, #006e92) !important;}<\/style><a class=\"stk-link stk-button stk--hover-effect-darken\" href=\"javascript:window.history.back();\" title=\"Back\"><span class=\"stk--svg-wrapper\"><div class=\"stk--inner-svg\"><svg xmlns=\"http:\/\/www.w3.org\/2000\/svg\" viewbox=\"0 0 512 512\" aria-hidden=\"true\" width=\"32\" height=\"32\"><path d=\"M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 288 480 288c17.7 0 32-14.3 32-32s-14.3-32-32-32l-370.7 0 73.4-73.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-128 128z\"><\/path><\/svg><\/div><\/span><\/a><\/div>\n<\/div><\/div>\n<\/div><\/article><\/div>\n\n\n\n<div class=\"wp-block-stackable-heading stk-block-heading stk-block-heading--v2 stk-block stk-5cbd537\" id=\"quick-check\" data-block-id=\"5cbd537\"><style>.stk-5cbd537 {margin-bottom:0px !important;}<\/style><h1 class=\"stk-block-heading__text\">Information extraction from medical publications<\/h1><\/div>\n\n\n\n<div class=\"wp-block-stackable-icon-label stk-block-icon-label stk-block stk-9c593c0\" id=\"quick-check\" data-block-id=\"9c593c0\"><style>.stk-9c593c0 .stk-inner-blocks{gap:8px !important;}<\/style><div class=\"stk-row stk-inner-blocks stk-block-content\">\n<div class=\"wp-block-stackable-icon stk-block-icon has-text-align-left stk-block stk-624347f\" data-block-id=\"624347f\"><style>.stk-624347f .stk--svg-wrapper .stk--inner-svg svg:last-child{height:16px !important;width:16px !important;}.stk-624347f .stk--svg-wrapper .stk--inner-svg svg:last-child, .stk-624347f .stk--svg-wrapper .stk--inner-svg svg:last-child :is(g, path, rect, polygon, ellipse){fill:var(--theme-palette-color-10, #25bae2) !important;}<\/style><span class=\"stk--svg-wrapper\"><div class=\"stk--inner-svg\"><svg style=\"height:0;width:0\"><defs><lineargradient id=\"linear-gradient-624347f\" x1=\"0\" x2=\"100%\" y1=\"0\" y2=\"0\"><stop offset=\"0%\" style=\"stop-opacity:1;stop-color:var(--linear-gradient-624347-f-color-1)\"><\/stop><stop offset=\"100%\" style=\"stop-opacity:1;stop-color:var(--linear-gradient-624347-f-color-2)\"><\/stop><\/lineargradient><\/defs><\/svg><svg data-prefix=\"fa\" data-icon=\"star\" class=\"svg-inline--fa fa-star fa-w-18\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" viewbox=\"0 0 576 512\" aria-hidden=\"true\" width=\"32\" height=\"32\"><path fill=\"currentColor\" d=\"M259.3 17.8L194 150.2 47.9 171.5c-26.2 3.8-36.7 36.1-17.7 54.6l105.7 103-25 145.5c-4.5 26.3 23.2 46 46.4 33.7L288 439.6l130.7 68.7c23.2 12.2 50.9-7.4 46.4-33.7l-25-145.5 105.7-103c19-18.5 8.5-50.8-17.7-54.6L382 150.2 316.7 17.8c-11.7-23.6-45.6-23.9-57.4 0z\"><\/path><\/svg><\/div><\/span><\/div>\n\n\n\n<div class=\"wp-block-stackable-heading stk-block-heading stk-block-heading--v2 stk-block stk-cdc2d3f\" id=\"span-data-stk-dynamic-current-page-post-taxonomy-term-projektformat-contenteditable-false-class-stk-dynamic-content-quick-check-span\" data-block-id=\"cdc2d3f\"><p class=\"stk-block-heading__text\">Quick Check<\/p><\/div>\n<\/div><\/div>\n\n\n\n<div class=\"wp-block-stackable-heading stk-block-heading stk-block-heading--v2 stk-block stk-c21bed1\" id=\"ausgangssituation\" data-block-id=\"c21bed1\"><h2 class=\"stk-block-heading__text\">Initial situation<\/h2><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-65e3563\" data-block-id=\"65e3563\"><style>.stk-65e3563 {column-count:1 !important;}<\/style><p class=\"stk-block-text__text\">When preparing approval documents (core dossiers), specific information from medical studies must be compiled. This information includes the study design, the purpose of the study, the indication, the drugs used, the size of the participant group, the study results and the conclusion. Various terms and formulations are used in medical studies. Therefore, extracting this information from the documents requires expert knowledge in the pharmaceutical field and takes a lot of time. Tools that support the extraction process and can simplify and accelerate it are desirable.<\/p><\/div>\n\n\n\n<div class=\"wp-block-stackable-heading stk-block-heading stk-block-heading--v2 stk-block stk-8c26049\" id=\"losungsidee\" data-block-id=\"8c26049\"><h2 class=\"stk-block-heading__text\">Solution idea<\/h2><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-ab6a506\" data-block-id=\"ab6a506\"><style>.stk-ab6a506 {column-count:1 !important;}<\/style><p class=\"stk-block-text__text\">In the Quick Check, a common understanding of the documents and the information they contain was first developed. Test documents were used to examine the extraction features and identify formulations that provide indications of certain feature values. Attention was also paid to the extent to which the document structure can be used to determine the best candidates for feature values. Furthermore, it was discussed how the extraction functions can be made available to users and how the extracted information should be stored in order to achieve the greatest possible benefit.<\/p><\/div>\n\n\n\n<div class=\"wp-block-stackable-heading stk-block-heading stk-block-heading--v2 stk-block stk-76475ee\" id=\"nutzen\" data-block-id=\"76475ee\"><h2 class=\"stk-block-heading__text\">Benefit<\/h2><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-3e86ca0\" data-block-id=\"3e86ca0\"><p class=\"stk-block-text__text\">The results achieved in the quick check show that it is possible to extract the desired information from the documents with the available AI approaches in order to support the medical writers in their tasks. The potential benefits should be evaluated in more detail using a larger volume of documents. In the first step, the aim should not be to extract the required features fully automatically. Instead, an assistance function with a correction option should be implemented. The correction data can later be used to further train the AI and achieve better results.<\/p><\/div>\n\n\n\n<div class=\"wp-block-stackable-heading stk-block-heading stk-block-heading--v2 stk-block stk-1e747a7\" id=\"umsetzung-der-ki-applikation\" data-block-id=\"1e747a7\"><h2 class=\"stk-block-heading__text\">Implementation of the AI application<\/h2><\/div>\n\n\n\n<div class=\"wp-block-stackable-text stk-block-text stk-block stk-d745445\" data-block-id=\"d745445\"><p class=\"stk-block-text__text\">The existing methods for feature extraction were evaluated and the most suitable methods were selected for each feature. AI models were created for the features study design, medication and indication, adapted using freely available language models and made available for initial tests in Fraunhofer IAO's existing Thorpedo software framework. The AI models were combined with rule-based approaches (e.g. keyword lists) to create a hybrid approach in order to identify the relevant sections and sentences for the individual features.<\/p><\/div>\n<\/div><\/div><\/div>\n<\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>","protected":false},"excerpt":{"rendered":"<p>It was evaluated how the preparation of marketing authorization dossiers (core dossiers) can be supported by the extraction of information from medical studies.<\/p>","protected":false},"author":4,"featured_media":576,"template":"","format":"standard","meta":{"_acf_changed":true,"_gspb_post_css":".gspb_container-id-gsbp-1fa6a74,.gspb_container-id-gsbp-933ef16,.gspb_container-id-gsbp-eafe0ac{flex-direction:column;box-sizing:border-box}#gspb_container-id-gsbp-1fa6a74.gspb_container>p:last-of-type,#gspb_container-id-gsbp-933ef16.gspb_container>p:last-of-type,#gspb_container-id-gsbp-eafe0ac.gspb_container>p:last-of-type{margin-bottom:0}#gspb_container-id-gsbp-933ef16.gspb_container{position:relative;padding:24px}#gspb_container-id-gsbp-eafe0ac.gspb_container{position:relative;display:block;margin-left:auto}#gspb_container-id-gsbp-1fa6a74.gspb_container{position:relative;display:block;margin:0;padding:24px;background-color:var(--wp--preset--color--palette-color-11, var(--theme-palette-color-11, #006e92))}"},"bereich":[20,22],"institut":[],"projektformat":[14],"class_list":["post-574","projekt","type-projekt","status-publish","format-standard","has-post-thumbnail","hentry","bereich-entwicklung-innovation","bereich-verwaltung-organisation","projektformat-quick-check"],"blocksy_meta":[],"acf":[],"_links":{"self":[{"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/projekt\/574","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/projekt"}],"about":[{"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/types\/projekt"}],"author":[{"embeddable":true,"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/users\/4"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/media\/576"}],"wp:attachment":[{"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/media?parent=574"}],"wp:term":[{"taxonomy":"bereich","embeddable":true,"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/bereich?post=574"},{"taxonomy":"institut","embeddable":true,"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/institut?post=574"},{"taxonomy":"projektformat","embeddable":true,"href":"https:\/\/www.ki-fortschrittszentrum.de\/en\/wp-json\/wp\/v2\/projektformat?post=574"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}