Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations

BACKGROUND AND OBJECTIVESCentral cervical lymph node metastasis (CLNM) is considered a risk factor for recurrence in patients with papillary thyroid carcinoma (PTC). Traditional machine learning models suffered from "black-box" problems, which could not exactly explain the interactive effe...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computer methods and programs in biomedicine 2022-10, Vol.225, p.107038-107038, Article 107038
Hauptverfasser: Zou, Ying, Shi, Yan, Sun, Fang, Liu, Jihua, Guo, Yu, Zhang, Huanlei, Lu, Xiudi, Gong, Yan, Xia, Shuang
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 107038
container_issue
container_start_page 107038
container_title Computer methods and programs in biomedicine
container_volume 225
creator Zou, Ying
Shi, Yan
Sun, Fang
Liu, Jihua
Guo, Yu
Zhang, Huanlei
Lu, Xiudi
Gong, Yan
Xia, Shuang
description BACKGROUND AND OBJECTIVESCentral cervical lymph node metastasis (CLNM) is considered a risk factor for recurrence in patients with papillary thyroid carcinoma (PTC). Traditional machine learning models suffered from "black-box" problems, which could not exactly explain the interactive effects of the risk factors. We aimed to develop an eXtreme Gradient Boosting (XGBoost) model to assess CLNM, including positive and negative effects. METHODS1,122 patients with PTC admitted at Tianjin First Central Hospital from 2016 to 2020 were retrospectively selected. They were randomly divided into the training and test datasets with an 8:2 ratio. 108 patients with PTC admitted at Binzhou Medical University Hospital in 2020 served as the validation dataset. The XGBoost model was used to assess CLNM. The 10-fold cross-validation was utilized for model selection, and the metric used to evaluate classification performance was the average area under the curve (AUC) of 10-fold cross-validation. Interpretation and transparency of the "black-box" problem were performed. SHapley Additive exPlanations (SHAP) and local interpretable model-agnostic explanation (LIME) were used to ensure the stability and reliability of the model. RESULTSThe XGBoost model based on ultrasound and dual-energy computed tomography images of the solitary primary lesion had an excellent performance for assessing CLNM, with average AUCs of 0.918, 0.903, and 0.881 in the training, test, and validation datasets, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, diameter, iodine concentration in the venous phase, and calcification) and negative (i.e., sex and age) impacts. For all cases, the capsular invasion prediction weight was the highest; for individual cases, different predictors were assigned different weights. Moreover, the performance of the XGBoost model was better than classical machine-learning models. CONCLUSIONSThis study developed and validated an XGBoost model for assessing CLNM in patients with PTC. The ability to visually interpret the positive and negative effects made the XGBoost model an effective tool for guiding clinical treatment.
doi_str_mv 10.1016/j.cmpb.2022.107038
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2699702189</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2699702189</sourcerecordid><originalsourceid>FETCH-LOGICAL-c324t-af9264c7a2940aa414e9eb91c3138e049f68fab3179a9112b787ff98c5503dae3</originalsourceid><addsrcrecordid>eNotkd1q3DAQhU1podskL5CrueyNt5L8q96FkDaBQAtJr81YHme1tSVXo91k366PVpktDMwPH4c5nCy7lmIrhay_7LdmXvqtEkqlQyOK9l22kW2j8qaqq_fZJkE6V7VoPmafmPdCCFVV9Sb7e_cWA80ELwEHSy5C7z1H615g9gNNED0gMzFDsPwb_AgmUQGn1MPRmjRMp3nZgUs4zBSRU1kG62DBuEoyvNq4S9tipwnDCeLuFLwdwGAw1vkZv8KDG-zRDocktwQarInWOzjw-sjTPS4TneBmGGy0RwJ6-zmhwxXhy-zDiBPT1f9-kf36dvd8e58__vj-cHvzmJtClTHHUau6NA0qXQrEUpakqdfSFLJoSZR6rNsR-0I2GrWUqm_aZhx1a6pKFANScZF9Pusuwf85EMdutmwoGXLkD9ypWutGKNnqhKozaoJnDjR2S7BzMt5J0a1xdftujatb4-rOcRX_AF21j8s</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2699702189</pqid></control><display><type>article</type><title>Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations</title><source>ScienceDirect</source><creator>Zou, Ying ; Shi, Yan ; Sun, Fang ; Liu, Jihua ; Guo, Yu ; Zhang, Huanlei ; Lu, Xiudi ; Gong, Yan ; Xia, Shuang</creator><creatorcontrib>Zou, Ying ; Shi, Yan ; Sun, Fang ; Liu, Jihua ; Guo, Yu ; Zhang, Huanlei ; Lu, Xiudi ; Gong, Yan ; Xia, Shuang</creatorcontrib><description>BACKGROUND AND OBJECTIVESCentral cervical lymph node metastasis (CLNM) is considered a risk factor for recurrence in patients with papillary thyroid carcinoma (PTC). Traditional machine learning models suffered from "black-box" problems, which could not exactly explain the interactive effects of the risk factors. We aimed to develop an eXtreme Gradient Boosting (XGBoost) model to assess CLNM, including positive and negative effects. METHODS1,122 patients with PTC admitted at Tianjin First Central Hospital from 2016 to 2020 were retrospectively selected. They were randomly divided into the training and test datasets with an 8:2 ratio. 108 patients with PTC admitted at Binzhou Medical University Hospital in 2020 served as the validation dataset. The XGBoost model was used to assess CLNM. The 10-fold cross-validation was utilized for model selection, and the metric used to evaluate classification performance was the average area under the curve (AUC) of 10-fold cross-validation. Interpretation and transparency of the "black-box" problem were performed. SHapley Additive exPlanations (SHAP) and local interpretable model-agnostic explanation (LIME) were used to ensure the stability and reliability of the model. RESULTSThe XGBoost model based on ultrasound and dual-energy computed tomography images of the solitary primary lesion had an excellent performance for assessing CLNM, with average AUCs of 0.918, 0.903, and 0.881 in the training, test, and validation datasets, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, diameter, iodine concentration in the venous phase, and calcification) and negative (i.e., sex and age) impacts. For all cases, the capsular invasion prediction weight was the highest; for individual cases, different predictors were assigned different weights. Moreover, the performance of the XGBoost model was better than classical machine-learning models. CONCLUSIONSThis study developed and validated an XGBoost model for assessing CLNM in patients with PTC. The ability to visually interpret the positive and negative effects made the XGBoost model an effective tool for guiding clinical treatment.</description><identifier>ISSN: 0169-2607</identifier><identifier>EISSN: 1872-7565</identifier><identifier>DOI: 10.1016/j.cmpb.2022.107038</identifier><language>eng</language><ispartof>Computer methods and programs in biomedicine, 2022-10, Vol.225, p.107038-107038, Article 107038</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c324t-af9264c7a2940aa414e9eb91c3138e049f68fab3179a9112b787ff98c5503dae3</citedby><cites>FETCH-LOGICAL-c324t-af9264c7a2940aa414e9eb91c3138e049f68fab3179a9112b787ff98c5503dae3</cites><orcidid>0000-0002-7828-0474 ; 0000-0002-6450-6997 ; 0000-0002-9222-5316 ; 0000-0002-6191-4338</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Zou, Ying</creatorcontrib><creatorcontrib>Shi, Yan</creatorcontrib><creatorcontrib>Sun, Fang</creatorcontrib><creatorcontrib>Liu, Jihua</creatorcontrib><creatorcontrib>Guo, Yu</creatorcontrib><creatorcontrib>Zhang, Huanlei</creatorcontrib><creatorcontrib>Lu, Xiudi</creatorcontrib><creatorcontrib>Gong, Yan</creatorcontrib><creatorcontrib>Xia, Shuang</creatorcontrib><title>Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations</title><title>Computer methods and programs in biomedicine</title><description>BACKGROUND AND OBJECTIVESCentral cervical lymph node metastasis (CLNM) is considered a risk factor for recurrence in patients with papillary thyroid carcinoma (PTC). Traditional machine learning models suffered from "black-box" problems, which could not exactly explain the interactive effects of the risk factors. We aimed to develop an eXtreme Gradient Boosting (XGBoost) model to assess CLNM, including positive and negative effects. METHODS1,122 patients with PTC admitted at Tianjin First Central Hospital from 2016 to 2020 were retrospectively selected. They were randomly divided into the training and test datasets with an 8:2 ratio. 108 patients with PTC admitted at Binzhou Medical University Hospital in 2020 served as the validation dataset. The XGBoost model was used to assess CLNM. The 10-fold cross-validation was utilized for model selection, and the metric used to evaluate classification performance was the average area under the curve (AUC) of 10-fold cross-validation. Interpretation and transparency of the "black-box" problem were performed. SHapley Additive exPlanations (SHAP) and local interpretable model-agnostic explanation (LIME) were used to ensure the stability and reliability of the model. RESULTSThe XGBoost model based on ultrasound and dual-energy computed tomography images of the solitary primary lesion had an excellent performance for assessing CLNM, with average AUCs of 0.918, 0.903, and 0.881 in the training, test, and validation datasets, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, diameter, iodine concentration in the venous phase, and calcification) and negative (i.e., sex and age) impacts. For all cases, the capsular invasion prediction weight was the highest; for individual cases, different predictors were assigned different weights. Moreover, the performance of the XGBoost model was better than classical machine-learning models. CONCLUSIONSThis study developed and validated an XGBoost model for assessing CLNM in patients with PTC. The ability to visually interpret the positive and negative effects made the XGBoost model an effective tool for guiding clinical treatment.</description><issn>0169-2607</issn><issn>1872-7565</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><recordid>eNotkd1q3DAQhU1podskL5CrueyNt5L8q96FkDaBQAtJr81YHme1tSVXo91k366PVpktDMwPH4c5nCy7lmIrhay_7LdmXvqtEkqlQyOK9l22kW2j8qaqq_fZJkE6V7VoPmafmPdCCFVV9Sb7e_cWA80ELwEHSy5C7z1H615g9gNNED0gMzFDsPwb_AgmUQGn1MPRmjRMp3nZgUs4zBSRU1kG62DBuEoyvNq4S9tipwnDCeLuFLwdwGAw1vkZv8KDG-zRDocktwQarInWOzjw-sjTPS4TneBmGGy0RwJ6-zmhwxXhy-zDiBPT1f9-kf36dvd8e58__vj-cHvzmJtClTHHUau6NA0qXQrEUpakqdfSFLJoSZR6rNsR-0I2GrWUqm_aZhx1a6pKFANScZF9Pusuwf85EMdutmwoGXLkD9ypWutGKNnqhKozaoJnDjR2S7BzMt5J0a1xdftujatb4-rOcRX_AF21j8s</recordid><startdate>202210</startdate><enddate>202210</enddate><creator>Zou, Ying</creator><creator>Shi, Yan</creator><creator>Sun, Fang</creator><creator>Liu, Jihua</creator><creator>Guo, Yu</creator><creator>Zhang, Huanlei</creator><creator>Lu, Xiudi</creator><creator>Gong, Yan</creator><creator>Xia, Shuang</creator><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-7828-0474</orcidid><orcidid>https://orcid.org/0000-0002-6450-6997</orcidid><orcidid>https://orcid.org/0000-0002-9222-5316</orcidid><orcidid>https://orcid.org/0000-0002-6191-4338</orcidid></search><sort><creationdate>202210</creationdate><title>Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations</title><author>Zou, Ying ; Shi, Yan ; Sun, Fang ; Liu, Jihua ; Guo, Yu ; Zhang, Huanlei ; Lu, Xiudi ; Gong, Yan ; Xia, Shuang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c324t-af9264c7a2940aa414e9eb91c3138e049f68fab3179a9112b787ff98c5503dae3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zou, Ying</creatorcontrib><creatorcontrib>Shi, Yan</creatorcontrib><creatorcontrib>Sun, Fang</creatorcontrib><creatorcontrib>Liu, Jihua</creatorcontrib><creatorcontrib>Guo, Yu</creatorcontrib><creatorcontrib>Zhang, Huanlei</creatorcontrib><creatorcontrib>Lu, Xiudi</creatorcontrib><creatorcontrib>Gong, Yan</creatorcontrib><creatorcontrib>Xia, Shuang</creatorcontrib><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Computer methods and programs in biomedicine</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zou, Ying</au><au>Shi, Yan</au><au>Sun, Fang</au><au>Liu, Jihua</au><au>Guo, Yu</au><au>Zhang, Huanlei</au><au>Lu, Xiudi</au><au>Gong, Yan</au><au>Xia, Shuang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations</atitle><jtitle>Computer methods and programs in biomedicine</jtitle><date>2022-10</date><risdate>2022</risdate><volume>225</volume><spage>107038</spage><epage>107038</epage><pages>107038-107038</pages><artnum>107038</artnum><issn>0169-2607</issn><eissn>1872-7565</eissn><abstract>BACKGROUND AND OBJECTIVESCentral cervical lymph node metastasis (CLNM) is considered a risk factor for recurrence in patients with papillary thyroid carcinoma (PTC). Traditional machine learning models suffered from "black-box" problems, which could not exactly explain the interactive effects of the risk factors. We aimed to develop an eXtreme Gradient Boosting (XGBoost) model to assess CLNM, including positive and negative effects. METHODS1,122 patients with PTC admitted at Tianjin First Central Hospital from 2016 to 2020 were retrospectively selected. They were randomly divided into the training and test datasets with an 8:2 ratio. 108 patients with PTC admitted at Binzhou Medical University Hospital in 2020 served as the validation dataset. The XGBoost model was used to assess CLNM. The 10-fold cross-validation was utilized for model selection, and the metric used to evaluate classification performance was the average area under the curve (AUC) of 10-fold cross-validation. Interpretation and transparency of the "black-box" problem were performed. SHapley Additive exPlanations (SHAP) and local interpretable model-agnostic explanation (LIME) were used to ensure the stability and reliability of the model. RESULTSThe XGBoost model based on ultrasound and dual-energy computed tomography images of the solitary primary lesion had an excellent performance for assessing CLNM, with average AUCs of 0.918, 0.903, and 0.881 in the training, test, and validation datasets, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, diameter, iodine concentration in the venous phase, and calcification) and negative (i.e., sex and age) impacts. For all cases, the capsular invasion prediction weight was the highest; for individual cases, different predictors were assigned different weights. Moreover, the performance of the XGBoost model was better than classical machine-learning models. CONCLUSIONSThis study developed and validated an XGBoost model for assessing CLNM in patients with PTC. The ability to visually interpret the positive and negative effects made the XGBoost model an effective tool for guiding clinical treatment.</abstract><doi>10.1016/j.cmpb.2022.107038</doi><tpages>1</tpages><orcidid>https://orcid.org/0000-0002-7828-0474</orcidid><orcidid>https://orcid.org/0000-0002-6450-6997</orcidid><orcidid>https://orcid.org/0000-0002-9222-5316</orcidid><orcidid>https://orcid.org/0000-0002-6191-4338</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0169-2607
ispartof Computer methods and programs in biomedicine, 2022-10, Vol.225, p.107038-107038, Article 107038
issn 0169-2607
1872-7565
language eng
recordid cdi_proquest_miscellaneous_2699702189
source ScienceDirect
title Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T09%3A59%3A05IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Extreme%20gradient%20boosting%20model%20to%20assess%20risk%20of%20central%20cervical%20lymph%20node%20metastasis%20in%20patients%20with%20papillary%20thyroid%20carcinoma:%20Individual%20prediction%20using%20SHapley%20Additive%20exPlanations&rft.jtitle=Computer%20methods%20and%20programs%20in%20biomedicine&rft.au=Zou,%20Ying&rft.date=2022-10&rft.volume=225&rft.spage=107038&rft.epage=107038&rft.pages=107038-107038&rft.artnum=107038&rft.issn=0169-2607&rft.eissn=1872-7565&rft_id=info:doi/10.1016/j.cmpb.2022.107038&rft_dat=%3Cproquest_cross%3E2699702189%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2699702189&rft_id=info:pmid/&rfr_iscdi=true