{"id":3506,"date":"2025-09-02T11:12:58","date_gmt":"2025-09-02T09:12:58","guid":{"rendered":"https:\/\/neuraldesigner.com\/learning\/lung-cancer-recurrence\/"},"modified":"2026-02-19T09:58:58","modified_gmt":"2026-02-19T08:58:58","slug":"lung-cancer-recurrence","status":"publish","type":"learning","link":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/","title":{"rendered":"Predicting cancer relapse with machine learning"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"3506\" class=\"elementor elementor-3506\" data-elementor-post-type=\"learning\">\n\t\t\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-49bbefec elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"49bbefec\" 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=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-66049ea4\" data-id=\"66049ea4\" 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-062a793 elementor-widget elementor-widget-heading\" data-id=\"062a793\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Introduction<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d7bc345 elementor-widget elementor-widget-text-editor\" data-id=\"d7bc345\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\" data-start=\"115\" data-end=\"248\">Machine learning can help physicians predict lung cancer recurrence, enabling earlier interventions and personalized treatment.<\/p><p style=\"text-align: justify;\" data-start=\"250\" data-end=\"377\">Recurrence after surgery is a major challenge, making accurate risk assessment essential for follow-up and therapy decisions.<\/p><p style=\"text-align: justify;\" data-start=\"379\" data-end=\"497\">We developed a neural network model using gene expression and clinical data to estimate 5-year relapse risk.<\/p><p style=\"text-align: justify;\" data-start=\"499\" data-end=\"658\">Using 335 patients (<a href=\"https:\/\/www.ncbi.nlm.nih.gov\/geo\/\">GEO<\/a> dataset), the model achieved an AUC of 0.84\u00a0and 76.3% accuracy, demonstrating strong potential as a clinical decision-support tool.<\/p><p style=\"text-align: justify;\" data-start=\"660\" data-end=\"750\">Healthcare professionals can test this methodology by downloading Neural Designer.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-5277af6 e-flex e-con-boxed e-con e-parent\" data-id=\"5277af6\" data-element_type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-0499bc3 elementor-widget__width-initial boton_descarga elementor-widget-mobile__width-initial elementor-widget elementor-widget-button\" data-id=\"0499bc3\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"https:\/\/www.neuraldesigner.com\/downloads\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-icon\">\n\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-download\" viewBox=\"0 0 512 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M216 0h80c13.3 0 24 10.7 24 24v168h87.7c17.8 0 26.7 21.5 14.1 34.1L269.7 378.3c-7.5 7.5-19.8 7.5-27.3 0L90.1 226.1c-12.6-12.6-3.7-34.1 14.1-34.1H192V24c0-13.3 10.7-24 24-24zm296 376v112c0 13.3-10.7 24-24 24H24c-13.3 0-24-10.7-24-24V376c0-13.3 10.7-24 24-24h146.7l49 49c20.1 20.1 52.5 20.1 72.6 0l49-49H488c13.3 0 24 10.7 24 24zm-124 88c0-11-9-20-20-20s-20 9-20 20 9 20 20 20 20-9 20-20zm64 0c0-11-9-20-20-20s-20 9-20 20 9 20 20 20 20-9 20-20z\"><\/path><\/svg>\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">Download<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-6a830d9 elementor-widget elementor-widget-heading\" data-id=\"6a830d9\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Contents<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-754961b elementor-widget elementor-widget-text-editor\" data-id=\"754961b\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The following index outlines the steps for performing the analysis.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-035dd5e e-grid e-con-boxed e-con e-parent\" data-id=\"035dd5e\" data-element_type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-8de2f9d elementor-align-center elementor-widget elementor-widget-button\" data-id=\"8de2f9d\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#model_type\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">1.Model type<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-24e8da1 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"24e8da1\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#dataset\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">2.Dataset<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-9514ee9 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"9514ee9\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#neural_network\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">3.Neural network<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-749b54b elementor-align-center elementor-widget elementor-widget-button\" data-id=\"749b54b\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#training_strategy\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">4.Training strategy<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4be8e17 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"4be8e17\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#model_selection\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">5.Model selection<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4db418d elementor-align-center elementor-widget elementor-widget-button\" data-id=\"4db418d\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#testing_analysis\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">6.Testing analysis<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c07ba41 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"c07ba41\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"#model_deployment\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">7.Model deployment<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c5df658 elementor-widget elementor-widget-heading\" data-id=\"c5df658\" data-element_type=\"widget\" id=\"model_type\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">1. Model type<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5788fa8 elementor-widget elementor-widget-text-editor\" data-id=\"5788fa8\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<ul>\n \t<li data-start=\"64\" data-end=\"134\" style=\"text-align: justify;\"><strong data-start=\"64\" data-end=\"81\">Problem type:<\/strong> Binary <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/neural-networks-applications#Classification\">classification<\/a> (lung relapse or no relapse)<\/li>\n \t<li data-start=\"64\" data-end=\"134\" style=\"text-align: justify;\"><strong data-start=\"136\" data-end=\"145\">Goal:<\/strong> Model the probability of lung relapse based on gene expression and clinical data using AI and machine learning to support clinical decision-making.<\/li>\n<\/ul>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d38077f elementor-widget elementor-widget-heading\" data-id=\"d38077f\" data-element_type=\"widget\" id=\"dataset\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">2. Data set<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-36e5834 elementor-widget elementor-widget-text-editor\" data-id=\"36e5834\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<h3>Data source<\/h3>\n<p style=\"text-align: justify;\">The lung_cancer_relapse.csv file has the data for this example.<\/p>\n\n<p style=\"text-align: justify;\">The target variable can only be binary in a classification model: 0 (false, no) or 1 (true, yes).<\/p>\n\n<p style=\"text-align: justify;\">The number of rows (instances) in the data set is 18388, and the number of columns (variables) is 11747.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-77118e3 elementor-widget elementor-widget-heading\" data-id=\"77118e3\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Variables<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4ada5a8 elementor-widget elementor-widget-text-editor\" data-id=\"4ada5a8\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The following list summarizes the variables used in the lung cancer relapse model:<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3aa1202 elementor-widget elementor-widget-heading\" data-id=\"3aa1202\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">Clinical features<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-9dea5e4 elementor-widget elementor-widget-text-editor\" data-id=\"9dea5e4\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<ul data-start=\"467\" data-end=\"1425\"><li style=\"text-align: justify;\" data-start=\"467\" data-end=\"558\"><p data-start=\"469\" data-end=\"558\"><strong data-start=\"469\" data-end=\"485\">month (0\u201360)<\/strong> \u2013 Time from the patient\u2019s surgical procedure until month 60 (5 years).<\/p><\/li><li style=\"text-align: justify;\" data-start=\"559\" data-end=\"611\"><p data-start=\"561\" data-end=\"611\"><strong data-start=\"561\" data-end=\"576\">source_name<\/strong> \u2013 Hospital source of the sample.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"612\" data-end=\"641\"><p data-start=\"614\" data-end=\"641\"><strong data-start=\"614\" data-end=\"621\">sex<\/strong> \u2013 Male or female.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"642\" data-end=\"695\"><p data-start=\"644\" data-end=\"695\"><strong data-start=\"644\" data-end=\"661\">age_in_months<\/strong> \u2013 Age of the patient in months.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"696\" data-end=\"763\"><p data-start=\"698\" data-end=\"763\"><strong data-start=\"698\" data-end=\"706\">race<\/strong> \u2013 Patient race, categorized by shared physical traits.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"764\" data-end=\"870\"><p data-start=\"766\" data-end=\"870\"><strong data-start=\"766\" data-end=\"816\">clinical treatment adjuvant chemotherapy (0\/1)<\/strong> \u2013 Whether or not the patient received chemotherapy.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"871\" data-end=\"977\"><p data-start=\"873\" data-end=\"977\"><strong data-start=\"873\" data-end=\"923\">clinical treatment adjuvant radiotherapy (0\/1)<\/strong> \u2013 Whether or not the patient received radiotherapy.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"978\" data-end=\"1076\"><p data-start=\"980\" data-end=\"1076\"><strong data-start=\"980\" data-end=\"1013\">pathological_nodes (0, 1, 2+)<\/strong> \u2013 Lymph node involvement relative to the TNM classification.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"1077\" data-end=\"1176\"><p data-start=\"1079\" data-end=\"1176\"><strong data-start=\"1079\" data-end=\"1108\">pathological_tumour (1\u20134)<\/strong> \u2013 Extent of the primary tumor relative to the TNM classification.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"1177\" data-end=\"1233\"><p data-start=\"1179\" data-end=\"1233\"><strong data-start=\"1179\" data-end=\"1204\">smoking_history (0\/1)<\/strong> \u2013 Patient smoking history.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"1234\" data-end=\"1317\"><p data-start=\"1236\" data-end=\"1317\"><strong data-start=\"1236\" data-end=\"1256\">surgical_margins<\/strong> \u2013 Margin of non-tumorous tissue around the resected tumor.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"1318\" data-end=\"1425\"><p data-start=\"1320\" data-end=\"1425\"><strong data-start=\"1320\" data-end=\"1346\">histologic_grade (0\u20132)<\/strong> \u2013 Description of how abnormal the cancer cells\/tissue look under a microscope.<\/p><\/li><\/ul>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-574c6d1 elementor-widget elementor-widget-heading\" data-id=\"574c6d1\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">Gene expression features<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c87dbd6 elementor-widget elementor-widget-text-editor\" data-id=\"c87dbd6\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<ul>\n \t<li style=\"text-align: justify;\"><strong data-start=\"1460\" data-end=\"1518\">RAD51, ADGRF5, COCH, SLC2A1, CLU, ZDHHC7, LRFN4, AP2A2<\/strong> \u2013 Expression levels of selected genes significantly correlated with lung cancer relapse.<\/li>\n<\/ul>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-663cdbb elementor-widget elementor-widget-heading\" data-id=\"663cdbb\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">Target variable<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-43966ca elementor-widget elementor-widget-text-editor\" data-id=\"43966ca\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<ul data-start=\"1631\" data-end=\"1731\">\n \t<li data-start=\"1631\" data-end=\"1731\" style=\"text-align: justify;\">\n<p data-start=\"1633\" data-end=\"1731\"><strong data-start=\"1633\" data-end=\"1653\">relapse (yes or no)<\/strong> \u2013 0 = no relapse, 1 = patient experienced lung cancer relapse within 5 years.<\/p>\n<\/li>\n<\/ul>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-51f2bdd elementor-widget elementor-widget-text-editor\" data-id=\"51f2bdd\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<h3>Instances<\/h3>\n<p style=\"text-align: justify;\">The dataset\u2019s\u00a0<a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/data-set#Instances\">instances<\/a>\u00a0(one per patient, including input and target variables) are split into training (60%), validation (20%), and testing (20%) subsets by default, adjustable as needed.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4fb66b8 elementor-widget elementor-widget-text-editor\" data-id=\"4fb66b8\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<h3>Variables distributions<\/h3>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1c36aa9 elementor-widget elementor-widget-text-editor\" data-id=\"1c36aa9\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/data-set#Distributions\">distributions<\/a> of all variables can be analyzed; the figure shows a pie chart of patients with and without relapse.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-117743e elementor-widget elementor-widget-image\" data-id=\"117743e\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img fetchpriority=\"high\" decoding=\"async\" width=\"600\" height=\"350\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_pie.webp\" class=\"attachment-large size-large wp-image-16591\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_pie.webp 600w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_pie-300x175.webp 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-bc71f2a elementor-widget elementor-widget-text-editor\" data-id=\"bc71f2a\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">As depicted in the image, 38.79% of patients experienced a relapse, while 61.21% did not.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-b39ef28 elementor-widget elementor-widget-text-editor\" data-id=\"b39ef28\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<h3>Input-target correlations<\/h3>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-77d3bb7 elementor-widget elementor-widget-text-editor\" data-id=\"77d3bb7\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/data-set#InputsTargetsCorrelations\" target=\"_blank\" rel=\"noopener\">input-target correlations<\/a> indicate which clinical and pathological factors most influence whether a patient will experience a relapse, making them particularly relevant for our analysis.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-08c38a8 elementor-widget elementor-widget-image\" data-id=\"08c38a8\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"530\" height=\"547\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_cors.webp\" class=\"attachment-large size-large wp-image-16590\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_cors.webp 530w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_cors-291x300.webp 291w\" sizes=\"(max-width: 530px) 100vw, 530px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-7625cb4 elementor-widget elementor-widget-text-editor\" data-id=\"7625cb4\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">Here, the most correlated variables with relapse are <strong>month<\/strong>, <strong>pathological nodes<\/strong>, and <strong>SLC2A1<\/strong>.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1fed09c elementor-widget elementor-widget-heading\" data-id=\"1fed09c\" data-element_type=\"widget\" id=\"neural_network\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">3. Neural network<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ae168cf elementor-widget elementor-widget-text-editor\" data-id=\"ae168cf\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">A neural network is an artificial intelligence model inspired by how the human brain processes information.<\/p>\n\n<p style=\"text-align: justify;\">It is organized in layers: the input layer receives the variables, and the output layer provides the probability of belonging to a given class.<\/p>\n\n<p style=\"text-align: justify;\">Trained with historical data, the network learns to recognize patterns and distinguish between categories, offering objective support for decision-making.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-098451f elementor-widget elementor-widget-image\" data-id=\"098451f\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"500\" height=\"683\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_network.webp\" class=\"attachment-large size-large wp-image-16589\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_network.webp 500w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_network-220x300.webp 220w\" sizes=\"(max-width: 500px) 100vw, 500px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-8247902 elementor-widget elementor-widget-text-editor\" data-id=\"8247902\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The network uses thirteen clinical and pathological variables to predict the probability of cancer relapse, with connections showing each variable&#8217;s contribution.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2fa54d6 elementor-widget elementor-widget-heading\" data-id=\"2fa54d6\" data-element_type=\"widget\" id=\"training_strategy\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">4. Training strategy<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3278812 elementor-widget elementor-widget-text-editor\" data-id=\"3278812\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">Training a neural network uses a loss function to measure errors and an optimization algorithm to adjust the model, ensuring it learns from data while avoiding overfitting for good performance on new cases.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-0afe3ff elementor-widget elementor-widget-image\" data-id=\"0afe3ff\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img loading=\"lazy\" decoding=\"async\" width=\"600\" height=\"400\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_error_init.webp\" class=\"attachment-large size-large wp-image-16588\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_error_init.webp 600w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_error_init-300x200.webp 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5373673 elementor-widget elementor-widget-text-editor\" data-id=\"5373673\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The model was trained for accuracy and stability on new data, with training and selection errors decreasing steadily (0.03 and 0.27 MSE), indicating effective learning and generalization.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-807c106 elementor-widget elementor-widget-heading\" data-id=\"807c106\" data-element_type=\"widget\" id=\"model_selection\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">5. Model selection<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1b56103 elementor-widget elementor-widget-text-editor\" data-id=\"1b56103\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The objective of <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/model-selection\">model selection<\/a> is to find the network architecture that minimizes the error, that is, with the best generalization properties for the <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/data-set#SelectionInstances\">selected instances<\/a> of the data set.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-cb8d253 elementor-widget elementor-widget-image\" data-id=\"cb8d253\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img loading=\"lazy\" decoding=\"async\" width=\"600\" height=\"400\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_error.webp\" class=\"attachment-large size-large wp-image-16587\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_error.webp 600w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_error-300x200.webp 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-75372b7 elementor-widget elementor-widget-text-editor\" data-id=\"75372b7\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">After multiple simulations, the optimal model achieved a <strong>training error<\/strong> of <strong>0.17 NSE<\/strong> and a <strong>selection error<\/strong> of <strong>0.20 NSE<\/strong>, showing improved performance on unseen samples.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-7f0555e elementor-widget elementor-widget-text-editor\" data-id=\"7f0555e\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">Also, we have reduced the number of inputs to only 11 features. Our network is now like this:<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-57d0930 elementor-widget elementor-widget-image\" data-id=\"57d0930\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img loading=\"lazy\" decoding=\"async\" width=\"500\" height=\"578\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_improved_network.webp\" class=\"attachment-large size-large wp-image-16586\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_improved_network.webp 500w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_improved_network-260x300.webp 260w\" sizes=\"(max-width: 500px) 100vw, 500px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d36c1ec elementor-widget elementor-widget-text-editor\" data-id=\"d36c1ec\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">Our final network has 11 inputs corresponding to <strong>month<\/strong>, <strong>pathological_nodes<\/strong>, <strong>pathological_tumour<\/strong>, <strong>RAD51<\/strong>, <strong>ADGRF5<\/strong>, <strong>COCH<\/strong>, <strong>SLC2A1<\/strong>, <strong>CLU<\/strong>, <strong>ZDHHC7<\/strong>, <strong>LRFN4<\/strong>, and <strong>AP2A2<\/strong>.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-6f3016d elementor-widget elementor-widget-heading\" data-id=\"6f3016d\" data-element_type=\"widget\" id=\"testing_analysis\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">6. Testing analysis<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ce2012f elementor-widget elementor-widget-text-editor\" data-id=\"ce2012f\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/testing-analysis\">testing analysis<\/a> aims to validate the performance of the generalization properties of the trained neural network.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d14942f elementor-widget elementor-widget-heading\" data-id=\"d14942f\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">ROC curve<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5d1c371 elementor-widget elementor-widget-text-editor\" data-id=\"5d1c371\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p data-start=\"88\" data-end=\"320\" style=\"text-align: justify;\">The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/testing-analysis#RocCurve\">ROC curve<\/a> is a standard tool to evaluate a classification model, showing how well it distinguishes between two classes by comparing predicted results with actual outcomes, such as relapse or no relapse in lung cancer patients.<\/p>\n<p data-start=\"322\" data-end=\"390\" data-is-last-node=\"\" data-is-only-node=\"\" style=\"text-align: justify;\">A random classifier scores 0.5, while a perfect classifier scores 1.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-74d6914 elementor-widget elementor-widget-image\" data-id=\"74d6914\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img loading=\"lazy\" decoding=\"async\" width=\"600\" height=\"400\" src=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_roc.webp\" class=\"attachment-large size-large wp-image-16585\" alt=\"\" srcset=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_roc.webp 600w, https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/08\/lung_relapse_roc-300x200.webp 300w\" sizes=\"(max-width: 600px) 100vw, 600px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-e0cf2da elementor-widget elementor-widget-text-editor\" data-id=\"e0cf2da\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The model achieves an <strong>AUC<\/strong> of <strong>0.843<\/strong>, demonstrating excellent performance in distinguishing patients who will experience a relapse from those who will not.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-59a65a5 elementor-widget elementor-widget-heading\" data-id=\"59a65a5\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Confusion matrix<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1c428ef elementor-widget elementor-widget-text-editor\" data-id=\"1c428ef\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/testing-analysis#ConfusionMatrix\">confusion matrix<\/a> shows the model\u2019s performance by comparing predicted and actual relapse outcomes. It includes:<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3a7e840 elementor-widget elementor-widget-text-editor\" data-id=\"3a7e840\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<ul>\n \t<li data-start=\"191\" data-end=\"268\" style=\"text-align: justify;\">\n<p data-start=\"193\" data-end=\"268\"><strong data-start=\"193\" data-end=\"211\">True positives<\/strong> \u2013 patients correctly predicted to experience a relapse<\/p>\n<\/li>\n \t<li data-start=\"269\" data-end=\"349\" style=\"text-align: justify;\">\n<p data-start=\"271\" data-end=\"349\"><strong data-start=\"271\" data-end=\"290\">False positives<\/strong> \u2013 patients incorrectly predicted to experience a relapse<\/p>\n<\/li>\n \t<li data-start=\"350\" data-end=\"434\" style=\"text-align: justify;\">\n<p data-start=\"352\" data-end=\"434\"><strong data-start=\"352\" data-end=\"371\">False negatives<\/strong> \u2013 patients incorrectly predicted not to experience a relapse<\/p>\n<\/li>\n \t<li data-start=\"435\" data-end=\"514\" data-is-last-node=\"\" style=\"text-align: justify;\">\n<p data-start=\"437\" data-end=\"514\" data-is-last-node=\"\"><strong data-start=\"437\" data-end=\"455\">True negatives<\/strong> \u2013 patients correctly predicted not to experience a relapse<\/p>\n<\/li>\n<\/ul>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-a1e41ac elementor-widget elementor-widget-text-editor\" data-id=\"a1e41ac\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">For a decision threshold of 0.5, the confusion matrix was:<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-d608ad1 e-flex e-con-boxed e-con e-parent\" data-id=\"d608ad1\" data-element_type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-916b438 elementor-widget elementor-widget-text-editor\" data-id=\"916b438\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<table class=\"mi-tabla\" style=\"margin: auto; display: table; border-collapse: collapse;\"><tbody><tr><th>\u00a0<\/th><th>Predicted positive<\/th><th>Predicted negative<\/th><\/tr><tr><th style=\"text-align: left;\">Real positive<\/th><td style=\"text-align: right;\">18<\/td><td style=\"text-align: right;\">8<\/td><\/tr><tr><th style=\"text-align: left;\">Real negative<\/th><td style=\"text-align: right;\">8<\/td><td style=\"text-align: right;\">33<\/td><\/tr><\/tbody><\/table>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1131d43 elementor-widget elementor-widget-text-editor\" data-id=\"1131d43\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">In this case, 76% cases were <strong>correctly classified<\/strong> and 24% were <strong>misclassified<\/strong>.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-85c54d2 elementor-widget elementor-widget-heading\" data-id=\"85c54d2\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Binary classification<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-8f2f7ec elementor-widget elementor-widget-text-editor\" data-id=\"8f2f7ec\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">The performance of this <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/testing-analysis#BinaryClassificationTests\">binary classification<\/a> model is summarized with standard measures.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c8fa194 elementor-widget elementor-widget-text-editor\" data-id=\"c8fa194\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<ul><li style=\"text-align: justify;\" data-start=\"215\" data-end=\"295\"><p data-start=\"217\" data-end=\"295\"><strong data-start=\"217\" data-end=\"230\">Accuracy:<\/strong> 76.31% of patients were correctly classified regarding relapse.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"296\" data-end=\"349\"><p data-start=\"298\" data-end=\"349\"><strong data-start=\"298\" data-end=\"313\">Error rate:<\/strong> 23.68% of cases were misclassified.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"350\" data-end=\"427\"><p data-start=\"352\" data-end=\"427\"><strong data-start=\"352\" data-end=\"368\">Sensitivity:<\/strong> 75.21% of patients who relapsed were correctly identified.<\/p><\/li><li style=\"text-align: justify;\" data-start=\"428\" data-end=\"511\"><p data-start=\"430\" data-end=\"511\"><strong data-start=\"430\" data-end=\"446\">Specificity:<\/strong> 77.78% of patients who did not relapse were correctly identified.<\/p><\/li><\/ul>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-7e2a0b0 elementor-widget elementor-widget-text-editor\" data-id=\"7e2a0b0\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p style=\"text-align: justify;\">These measures indicate that the model is highly effective at predicting lung cancer relapse.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-475f20f elementor-widget elementor-widget-heading\" data-id=\"475f20f\" data-element_type=\"widget\" id=\"model_deployment\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">7. Model deployment<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-575caf1 elementor-widget elementor-widget-text-editor\" data-id=\"575caf1\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p data-start=\"294\" data-end=\"446\" style=\"text-align: justify;\">After confirming the neural network\u2019s ability to generalize, the model can be saved for future use in deployment mode.<\/p>\n<p data-start=\"294\" data-end=\"446\" style=\"text-align: justify;\">This allows the trained network to be applied to new patients, using their clinical variables to calculate the probability of lung cancer relapse.<\/p>\n\n<div class=\"elementor-element elementor-element-76e3d7e elementor-widget elementor-widget-text-editor\" data-id=\"76e3d7e\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\n<p style=\"text-align: justify;\">In deployment mode, healthcare professionals can use the model as a reliable diagnostic support tool for classifying new patients.<\/p>\n\n<p style=\"text-align: justify;\">The\u00a0<a href=\"https:\/\/www.neuraldesigner.com\/my-account\/\">Neural Designer<\/a> software exports the trained model automatically, making it easy to integrate into clinical practice.\n\n<\/div>\t\t\t\t\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-0306368 e-grid e-con-boxed e-con e-parent\" data-id=\"0306368\" data-element_type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-0b59448 e-con-full e-flex e-con e-child\" data-id=\"0b59448\" data-element_type=\"container\">\n\t\t\t\t<div class=\"elementor-element elementor-element-8d7c879 elementor-widget__width-initial boton_descarga elementor-widget-mobile__width-initial elementor-widget elementor-widget-button\" data-id=\"8d7c879\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t\t\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"https:\/\/www.neuraldesigner.com\/downloads\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-icon\">\n\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-download\" viewBox=\"0 0 512 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M216 0h80c13.3 0 24 10.7 24 24v168h87.7c17.8 0 26.7 21.5 14.1 34.1L269.7 378.3c-7.5 7.5-19.8 7.5-27.3 0L90.1 226.1c-12.6-12.6-3.7-34.1 14.1-34.1H192V24c0-13.3 10.7-24 24-24zm296 376v112c0 13.3-10.7 24-24 24H24c-13.3 0-24-10.7-24-24V376c0-13.3 10.7-24 24-24h146.7l49 49c20.1 20.1 52.5 20.1 72.6 0l49-49H488c13.3 0 24 10.7 24 24zm-124 88c0-11-9-20-20-20s-20 9-20 20 9 20 20 20 20-9 20-20zm64 0c0-11-9-20-20-20s-20 9-20 20 9 20 20 20 20-9 20-20z\"><\/path><\/svg>\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">Download<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\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\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-a9d0990 elementor-widget elementor-widget-heading\" data-id=\"a9d0990\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Conclusions<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-db48c94 elementor-widget elementor-widget-text-editor\" data-id=\"db48c94\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t\t\t\t\t\t<p data-start=\"144\" data-end=\"277\" style=\"text-align: justify;\">The lung cancer recurrence prediction model (GEO data) achieved AUC = 0.84, accuracy = 76.3% for 5-year relapse prediction.<\/p>\n<p data-start=\"279\" data-end=\"415\" style=\"text-align: justify;\">Top predictors\u2014pathological nodes, tumor stage, and gene markers (SLC2A1, RAD51)\u2014match established clinical and molecular factors.<\/p>\n<p data-start=\"417\" data-end=\"521\" style=\"text-align: justify;\">The model can support risk stratification, follow-up, and treatment planning in clinical practice.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1963ed6 elementor-widget elementor-widget-heading\" data-id=\"1963ed6\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Related posts<\/h2>\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},"author":13,"featured_media":1944,"template":"","categories":[29],"tags":[38],"class_list":["post-3506","learning","type-learning","status-publish","has-post-thumbnail","hentry","category-examples","tag-healthcare"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Predicting cancer relapse with machine learning<\/title>\n<meta name=\"description\" content=\"Using machine learning and genomic data to build models for relapse risk assessment in lung cancer patients.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Lung cancer relapse machine learning example\" \/>\n<meta property=\"og:description\" content=\"This example aims to assess the probability to suffer relapse in lung cancer patients.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/\" \/>\n<meta property=\"og:site_name\" content=\"Neural Designer\" \/>\n<meta property=\"article:modified_time\" content=\"2026-02-19T08:58:58+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"628\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:title\" content=\"Lung cancer relapse machine learning example\" \/>\n<meta name=\"twitter:description\" content=\"This example aims to assess the probability to suffer relapse in lung cancer patients.\" \/>\n<meta name=\"twitter:image\" content=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp\" \/>\n<meta name=\"twitter:site\" content=\"@NeuralDesigner\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/\",\"url\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/\",\"name\":\"Predicting cancer relapse with machine learning\",\"isPartOf\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp\",\"datePublished\":\"2025-09-02T09:12:58+00:00\",\"dateModified\":\"2026-02-19T08:58:58+00:00\",\"description\":\"Using machine learning and genomic data to build models for relapse risk assessment in lung cancer patients.\",\"breadcrumb\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#primaryimage\",\"url\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp\",\"contentUrl\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp\",\"width\":1200,\"height\":628},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.neuraldesigner.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Learning\",\"item\":\"https:\/\/www.neuraldesigner.com\/learning\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Predicting cancer relapse with machine learning\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.neuraldesigner.com\/#website\",\"url\":\"https:\/\/www.neuraldesigner.com\/\",\"name\":\"Neural Designer\",\"description\":\"Explanable AI Platform\",\"publisher\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.neuraldesigner.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.neuraldesigner.com\/#organization\",\"name\":\"Neural Designer\",\"url\":\"https:\/\/www.neuraldesigner.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png\",\"contentUrl\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png\",\"width\":1024,\"height\":223,\"caption\":\"Neural Designer\"},\"image\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/NeuralDesigner\",\"https:\/\/es.linkedin.com\/showcase\/neuraldesigner\/\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Predicting cancer relapse with machine learning","description":"Using machine learning and genomic data to build models for relapse risk assessment in lung cancer patients.","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:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/","og_locale":"en_US","og_type":"article","og_title":"Lung cancer relapse machine learning example","og_description":"This example aims to assess the probability to suffer relapse in lung cancer patients.","og_url":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/","og_site_name":"Neural Designer","article_modified_time":"2026-02-19T08:58:58+00:00","og_image":[{"width":1200,"height":628,"url":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp","type":"image\/webp"}],"twitter_card":"summary_large_image","twitter_title":"Lung cancer relapse machine learning example","twitter_description":"This example aims to assess the probability to suffer relapse in lung cancer patients.","twitter_image":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp","twitter_site":"@NeuralDesigner","twitter_misc":{"Est. reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/","url":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/","name":"Predicting cancer relapse with machine learning","isPartOf":{"@id":"https:\/\/www.neuraldesigner.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#primaryimage"},"image":{"@id":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#primaryimage"},"thumbnailUrl":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp","datePublished":"2025-09-02T09:12:58+00:00","dateModified":"2026-02-19T08:58:58+00:00","description":"Using machine learning and genomic data to build models for relapse risk assessment in lung cancer patients.","breadcrumb":{"@id":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#primaryimage","url":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp","contentUrl":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/lung-recurrence.webp","width":1200,"height":628},{"@type":"BreadcrumbList","@id":"https:\/\/www.neuraldesigner.com\/learning\/examples\/lung-cancer-recurrence\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.neuraldesigner.com\/"},{"@type":"ListItem","position":2,"name":"Learning","item":"https:\/\/www.neuraldesigner.com\/learning\/"},{"@type":"ListItem","position":3,"name":"Predicting cancer relapse with machine learning"}]},{"@type":"WebSite","@id":"https:\/\/www.neuraldesigner.com\/#website","url":"https:\/\/www.neuraldesigner.com\/","name":"Neural Designer","description":"Explanable AI Platform","publisher":{"@id":"https:\/\/www.neuraldesigner.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.neuraldesigner.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.neuraldesigner.com\/#organization","name":"Neural Designer","url":"https:\/\/www.neuraldesigner.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/","url":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png","contentUrl":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png","width":1024,"height":223,"caption":"Neural Designer"},"image":{"@id":"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/NeuralDesigner","https:\/\/es.linkedin.com\/showcase\/neuraldesigner\/"]}]}},"_links":{"self":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/learning\/3506","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/learning"}],"about":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/types\/learning"}],"author":[{"embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/users\/13"}],"version-history":[{"count":10,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/learning\/3506\/revisions"}],"predecessor-version":[{"id":21954,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/learning\/3506\/revisions\/21954"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/media\/1944"}],"wp:attachment":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/media?parent=3506"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/categories?post=3506"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/tags?post=3506"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}