[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118027-en":3,"doc-seo-118027-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},118027,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Machine Learning-Guided Adjuvant Treatment of Head and Neck Cancer","Supplemental tables summarize modeling setup and evaluation for machine learning–guided adjuvant treatment in head and neck cancer. Included materials provide optimal hyperparameters for DeepSurv and N-MTLR models and for Random Survival Forest, along with missing-value patterns before imputation. Additional content quantifies permutation feature importance via concordance-index concordance changes and reports hazard ratios for patient subgroups, linking model recommendations to survival outcomes in the test cohort.","Supplemental Online Content  \nHoward FM, Kochanny S, Koshy M, Spiotto M, Pearson AT. Machine learning–guided adjuvant treatment of head and neck cancer. JAMA Netw Open. 2020;3(11):e2025881 .  \ndoi:10.1001/jamanetworkopen.2020.25881  \neTable 1. Optimal Hyperparameters, DeepSurv and N-MTLR Models  \neTable 2. Optimal Hyperparameters, Random Survival Forest  \neTable 3. Missing Values for Features in the Final Data Set, Prior to Imputation eTable 4. Permutation Feature Importance  \neTable 5. Hazard Ratios for Subgroups  \neFigure. Survival Outcomes for Treatment According to Machine Learning Model Recommendations, Test Cohort  \nThis supplemental material has been provided by the authors to give readers additional information about their work.  \n© 2020 Howard FM et al. JAMA Network Open.  \neTable 1. Optimal Hyperparameters, DeepSurv and N-MTLR Models  \n\n| Parameter | DeepSurv Model | N-MTLR Model |\n| --- | --- | --- |\n| Neural Network Structure\u003Cbr>(Layer: activation; nodes) | 1: LeCun Tanh; 64\u003Cbr>2: Sinc; 128\u003Cbr>3: Log Sigmoid; 32\u003Cbr>4: Inverse Square Root; 128\u003Cbr>5: Swish; 16 | 1: Sinc; 32\u003Cbr>2: Sigmoid; 32\u003Cbr>3: Log Log; 8\u003Cbr>4: Arc Tan; 8 |\n| Learning Rate | 0.00001 | 0.0001 |\n| Number of Epochs | 6000 | 4000 |\n| Dropout | 0.1 | 0.1 |\n| L2 Regularization | 0.0001 | 0.001 |\n| Batch Normalization | False | False |\n| L2 Smoothing Regularization |  | 0.001 |\n| Bins |  | 100 |\n\n© 2020 Howard FM et al. JAMA Network Open.  \neTable 2. Optimal Hyperparameters, Random Survival Forest  \n\n| Parameter | Random Survival Forest Model |\n| --- | --- |\n| Trees | 80 |\n| Max Features | 0.1 |\n| Max Depth | 40 |\n| Min Node Size | 80 |\n| Sample Size Percent | 60% |\n| Importance Mode | Permutation |\n\n© 2020 Howard FM et al. JAMA Network Open.  \neTable 3. Missing Values for Features in the Final Data Set, Prior to Imputation  \n\n|  | Number Missing (% Total) |\n| --- | --- |\n| Gender | 0 (0) |\n| Age | 0 (0) |\n| Year of Diagnosis | 0 (0) |\n| Ethnicity | 279 (0 .8) |\n| Academic Center | 1,199 (3.6) |\n| Charlson/Deyo Score | 0 (0) |\n| Status At Last Contact | 0 (0) |\n| Primary Site | 0 (0) |\n| Tumor stage | 0 (0) |\n| Nodal Stage | 0 (0) |\n| Measured Tumor Size | 5,027 (15 .0) |\n| Tumor Thickness ( n,% of oral cavity cancers)a | 9,670 (61.1) |\n| Differentiation | 2,518 (7.5) |\n| LVI | 15,969 (47 .6) |\n| Lymph Nodes Positive | 4,465 (13 . 3) |\n| Lymph Node Levels Involved |  |\n| I, II, III | 2,211 (6 .6) |\n| IV, V, Retropharyngeal | 2,269 (6 .8) |\n| Parapharyngeal | 2,227 (6 .6) |\n| HPV Positivityb |  |\n| Oropharynx | 5,972 (53 . 5) |\n\n© 2020 Howard FM et al. JAMA Network Open.  \n\n| Non-oropharynx | 18,182 (81 .3) |\n| --- | --- |\n| Multiagent Chemotherapy | 1,581 (0) |\n| Time from Surgery to Completion of RT | 1,508 (4.7) |\n| RT Dose | 0 (0) |\n| Adequate Lymph Node Dissection\u003Cbr>(18+ Nodes Examined) | 79 (0.2) |\n\na For tumor thickness, percent missing is calculated as a percentage of oral cavity cancers  \nb For HPV positivity, percent missing values are calculated as a percentage of total oropharynx cancers, and as a percentage of total non-oropharynx cancers.  \n© 2020 Howard FM et al. JAMA Network Open.  \neTable 4. Permutation Feature Importance  \nGiven as % decrease in concordance index with permutation. A high reduction in concordance index with variable permutation indicates that feature is important in prediction of prognosis.  \n\n| Feature | DeepSurv | N-MTLR | RSF | Average |\n| --- | --- | --- | --- | --- |\n| Life Expectancy | 4.0922 | 4.4706 | 4.5369 | 4.3666 |\n| Year of Diagnosis 2013-2016 | 2.4761 | 1.5635 | 2.8574 | 2.299 |\n| Tumor Stage T4 | 2.3638 | 2.0986 | 0.6507 | 1.7044 |\n| HPV+ (Oropharynx) | 1.157 | 1.0594 | 1.4945 | 1.237 |\n| Tonsillar Subsite | 0.8495 | 1.2884 | 0.921 | 1.0196 |\n| 5-9 Lymph Nodes Positive | 1.4671 | 0.5078 | 0.5448 | 0.8399 |\n| Tumor Size | 0.4326 | 0.1226 | 1.644 | 0.7331 |\n| 2-4 Lymph Nodes Positive | 0.656 | 0.729 | 0.2707 | 0.5519 |\n| 10+ Lymph Nodes Positive | 0.8419 | 0.6268 | 0.126 | 0.5316 |\n| Tongue Subsite | 0.6829 ","cbCaipxIqJEOLQSP","https://ap.wps.com/l/cbCaipxIqJEOLQSP","pdf",1206716,1,10,"English","en",105,"# eTable 1 - Optimal Hyperparameters for DeepSurv and N-MTLR\n## Model structure and training settings\n# eTable 2 - Optimal Hyperparameters for Random Survival Forest\n## Tree and feature sampling parameters\n# eTable 3 - Missing Values Prior to Imputation\n## Missingness by clinical and tumor features\n# eTable 4 - Permutation Feature Importance\n## Variables ranked by concordance-index impact\n# eFigure - Survival Outcomes by Model Recommendations\n## Outcomes in the test cohort","[{\"question\":\"What hyperparameter settings are reported for the DeepSurv and N-MTLR models?\",\"answer\":\"The document lists optimal hyperparameters including neural network layer structures, learning rates, epoch counts, dropout, and regularization settings for DeepSurv and N-MTLR.\"},{\"question\":\"How does the supplemental material address missing data before imputation?\",\"answer\":\"It provides eTable 3 with the number and percentage of missing values for each feature in the final dataset, prior to imputation.\"},{\"question\":\"How is feature importance measured in the study?\",\"answer\":\"It uses permutation feature importance reported as the % decrease in concordance index when a variable is permuted, with higher reductions indicating stronger prognostic importance.\"}]","Machine Learning-Guided Adjuvant Treatment of Head and Neck Cancer | PDF",1785680832,25,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-guided-adjuvant-treatment-of-head-and-neck-cancer","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-guided-adjuvant-treatment-of-head-and-neck-cancer/118027/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-02",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What hyperparameter settings are reported for the DeepSurv and N-MTLR models?","Question",{"text":76,"@type":77},"The document lists optimal hyperparameters including neural network layer structures, learning rates, epoch counts, dropout, and regularization settings for DeepSurv and N-MTLR.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the supplemental material address missing data before imputation?",{"text":81,"@type":77},"It provides eTable 3 with the number and percentage of missing values for each feature in the final dataset, prior to imputation.",{"name":83,"@type":74,"acceptedAnswer":84},"How is feature importance measured in the study?",{"text":85,"@type":77},"It uses permutation feature importance reported as the % decrease in concordance index when a variable is permuted, with higher reductions indicating stronger prognostic importance.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":21,"slug":134},"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]