Predicting Postoperative Neurological Outcomes in Metastatic Spinal Tumor Surgery Using Machine Learning.

IF 2.6 2区 医学 Q2 CLINICAL NEUROLOGY
Spine Pub Date : 2025-03-03 DOI:10.1097/BRS.0000000000005322
Satoshi Maki, Yuki Shiratani, Sumihisa Orita, Akinobu Suzuki, Koji Tamai, Takaki Shimizu, Kenichiro Kakutani, Yutaro Kanda, Hiroyuki Tominaga, Ichiro Kawamura, Masayuki Ishihara, Masaaki Paku, Yohei Takahashi, Toru Funayama, Kousei Miura, Eiki Shirasawa, Hirokazu Inoue, Atsushi Kimura, Takuya Iimura, Hiroshi Moridaira, Hideaki Nakajima, Shuji Watanabe, Koji Akeda, Norihiko Takegami, Kazuo Nakanishi, Hirokatsu Sawada, Koji Matsumoto, Masahiro Funaba, Hidenori Suzuki, Haruki Funao, Tsutomu Oshigiri, Takashi Hirai, Bungo Otsuki, Kazu Kobayakawa, Koji Uotani, Hiroaki Manabe, Shinji Tanishima, Ko Hashimoto, Chizuo Iwai, Daisuke Yamabe, Akihiko Hiyama, Shoji Seki, Kenji Kato, Masashi Miyazaki, Kazuyuki Watanabe, Toshio Nakamae, Takashi Kaito, Hiroaki Nakashima, Narihito Nagoshi, Gen Inoue, Shiro Imagama, Kota Watanabe, Satoshi Kato, Seiji Ohtori, Takeo Furuya
{"title":"Predicting Postoperative Neurological Outcomes in Metastatic Spinal Tumor Surgery Using Machine Learning.","authors":"Satoshi Maki, Yuki Shiratani, Sumihisa Orita, Akinobu Suzuki, Koji Tamai, Takaki Shimizu, Kenichiro Kakutani, Yutaro Kanda, Hiroyuki Tominaga, Ichiro Kawamura, Masayuki Ishihara, Masaaki Paku, Yohei Takahashi, Toru Funayama, Kousei Miura, Eiki Shirasawa, Hirokazu Inoue, Atsushi Kimura, Takuya Iimura, Hiroshi Moridaira, Hideaki Nakajima, Shuji Watanabe, Koji Akeda, Norihiko Takegami, Kazuo Nakanishi, Hirokatsu Sawada, Koji Matsumoto, Masahiro Funaba, Hidenori Suzuki, Haruki Funao, Tsutomu Oshigiri, Takashi Hirai, Bungo Otsuki, Kazu Kobayakawa, Koji Uotani, Hiroaki Manabe, Shinji Tanishima, Ko Hashimoto, Chizuo Iwai, Daisuke Yamabe, Akihiko Hiyama, Shoji Seki, Kenji Kato, Masashi Miyazaki, Kazuyuki Watanabe, Toshio Nakamae, Takashi Kaito, Hiroaki Nakashima, Narihito Nagoshi, Gen Inoue, Shiro Imagama, Kota Watanabe, Satoshi Kato, Seiji Ohtori, Takeo Furuya","doi":"10.1097/BRS.0000000000005322","DOIUrl":null,"url":null,"abstract":"<p><strong>Study design: </strong>Retrospective analysis of data collected across multiple centers.</p><p><strong>Objective: </strong>To develop machine learning models for predicting neurological outcomes one month postoperatively in patients with metastatic spinal tumors undergoing surgery, and to identify key factors influencing neurological recovery.</p><p><strong>Summary of background data: </strong>The increasing prevalence of spinal metastases has led to a growing need for surgical intervention to address mechanical instability and neurological deficits. Predicting postoperative neurological status, as assessed by the Frankel classification, can provide valuable insights for surgical planning and patient counseling. Traditional prognostic models have shown limitations in capturing the complexity of neurological recovery patterns.</p><p><strong>Methods: </strong>We analyzed data from 244 patients who underwent spinal surgery for metastatic disease across 38 institutions. The primary outcome was functional ambulation, defined as Frankel grades D or E at one month postoperatively. Four machine learning algorithms (Random Forest, XGBoost, LightGBM, and CatBoost) were used to build predictive models. Feature selection employed the Boruta algorithm and Variance Inflation Factor analysis to reduce multicollinearity.</p><p><strong>Results: </strong>Among the 244 patients, the proportion of ambulatory patients (Frankel grades D or E) increased from 36.8% preoperatively to 63.1% at one month postoperatively. The Random Forest model achieved the highest area under the receiver operating characteristic curve (AUC-ROC) of 0.8516, followed by XGBoost (0.8351), CatBoost (0.8331), and LightGBM (0.8098). SHapley Additive exPlanations analysis identified preoperative Frankel classification, transfer ability, inflammatory markers (C-reactive protein, white blood cell-lymphocyte), and surgical timing as the most important predictors of postoperative outcomes.</p><p><strong>Conclusions: </strong>Machine learning models showed strong predictive performance in assessing postoperative neurological status for patients with metastatic spinal tumors. Key factors including preoperative neurological function, functional ability, and inflammation markers significantly influenced outcomes. These findings could inform surgical decision-making and help set realistic postoperative expectations while potentially improving patient care through more accurate outcome prediction.</p>","PeriodicalId":22193,"journal":{"name":"Spine","volume":" ","pages":""},"PeriodicalIF":2.6000,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Spine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/BRS.0000000000005322","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Study design: Retrospective analysis of data collected across multiple centers.

Objective: To develop machine learning models for predicting neurological outcomes one month postoperatively in patients with metastatic spinal tumors undergoing surgery, and to identify key factors influencing neurological recovery.

Summary of background data: The increasing prevalence of spinal metastases has led to a growing need for surgical intervention to address mechanical instability and neurological deficits. Predicting postoperative neurological status, as assessed by the Frankel classification, can provide valuable insights for surgical planning and patient counseling. Traditional prognostic models have shown limitations in capturing the complexity of neurological recovery patterns.

Methods: We analyzed data from 244 patients who underwent spinal surgery for metastatic disease across 38 institutions. The primary outcome was functional ambulation, defined as Frankel grades D or E at one month postoperatively. Four machine learning algorithms (Random Forest, XGBoost, LightGBM, and CatBoost) were used to build predictive models. Feature selection employed the Boruta algorithm and Variance Inflation Factor analysis to reduce multicollinearity.

Results: Among the 244 patients, the proportion of ambulatory patients (Frankel grades D or E) increased from 36.8% preoperatively to 63.1% at one month postoperatively. The Random Forest model achieved the highest area under the receiver operating characteristic curve (AUC-ROC) of 0.8516, followed by XGBoost (0.8351), CatBoost (0.8331), and LightGBM (0.8098). SHapley Additive exPlanations analysis identified preoperative Frankel classification, transfer ability, inflammatory markers (C-reactive protein, white blood cell-lymphocyte), and surgical timing as the most important predictors of postoperative outcomes.

Conclusions: Machine learning models showed strong predictive performance in assessing postoperative neurological status for patients with metastatic spinal tumors. Key factors including preoperative neurological function, functional ability, and inflammation markers significantly influenced outcomes. These findings could inform surgical decision-making and help set realistic postoperative expectations while potentially improving patient care through more accurate outcome prediction.

利用机器学习预测转移性脊柱肿瘤手术后神经预后。
研究设计:对多个中心收集的数据进行回顾性分析。目的:建立机器学习模型,预测脊柱转移性肿瘤术后1个月的神经预后,并确定影响神经恢复的关键因素。背景资料摘要:脊柱转移的发病率越来越高,导致越来越需要手术干预来解决机械不稳定和神经功能障碍。通过Frankel分类来预测术后神经状态,可以为手术计划和患者咨询提供有价值的见解。传统的预后模型在捕捉神经恢复模式的复杂性方面显示出局限性。方法:我们分析了来自38家机构的244名因转移性疾病接受脊柱手术的患者的数据。主要终点是术后一个月的功能活动,定义为Frankel分级D或E。使用四种机器学习算法(Random Forest, XGBoost, LightGBM和CatBoost)构建预测模型。特征选择采用Boruta算法和方差膨胀因子分析来减少多重共线性。结果:244例患者中,门诊患者(Frankel D级或E级)比例由术前的36.8%上升至术后1个月的63.1%。随机森林模型的受试者工作特征曲线下面积(AUC-ROC)最高,为0.8516,其次是XGBoost(0.8351)、CatBoost(0.8331)和LightGBM(0.8098)。SHapley加法解释分析确定术前Frankel分类、转移能力、炎症标志物(c反应蛋白、白细胞-淋巴细胞)和手术时间是术后预后最重要的预测因素。结论:机器学习模型在评估转移性脊柱肿瘤患者术后神经系统状态方面表现出很强的预测能力。关键因素包括术前神经功能、功能能力和炎症标志物显著影响预后。这些发现可以为手术决策提供信息,并帮助设定现实的术后期望,同时通过更准确的结果预测可能改善患者护理。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Spine
Spine 医学-临床神经学
CiteScore
5.90
自引率
6.70%
发文量
361
审稿时长
6.0 months
期刊介绍: Lippincott Williams & Wilkins is a leading international publisher of professional health information for physicians, nurses, specialized clinicians and students. For a complete listing of titles currently published by Lippincott Williams & Wilkins and detailed information about print, online, and other offerings, please visit the LWW Online Store. Recognized internationally as the leading journal in its field, Spine is an international, peer-reviewed, bi-weekly periodical that considers for publication original articles in the field of Spine. It is the leading subspecialty journal for the treatment of spinal disorders. Only original papers are considered for publication with the understanding that they are contributed solely to Spine. The Journal does not publish articles reporting material that has been reported at length elsewhere.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信