Machine learning model predicts prognosis in SARS-CoV-2-infected autoimmune encephalitis patients

IF 3 3区医学 Q2 CLINICAL NEUROLOGY

Multiple sclerosis and related disorders Pub Date : 2026-05-01 Epub Date: 2026-02-06 DOI:10.1016/j.msard.2026.107028

Xingjie Li , Xueying Kong , Aiqing Li , Xue Gong , Kundian Guo , Xu Liu , Yue Liu , Tianyu Zhang , Xiaolin Deng , Jinmei Li , Dong Zhou , Zhen Hong

{"title":"Machine learning model predicts prognosis in SARS-CoV-2-infected autoimmune encephalitis patients","authors":"Xingjie Li , Xueying Kong , Aiqing Li , Xue Gong , Kundian Guo , Xu Liu , Yue Liu , Tianyu Zhang , Xiaolin Deng , Jinmei Li , Dong Zhou , Zhen Hong","doi":"10.1016/j.msard.2026.107028","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>To assess the neurological prognosis of patients with autoimmune encephalitis (AE) after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and construct risk prediction models for a worse prognosis.</div></div><div><h3>Methods</h3><div>This was a prospective, multicenter, observational cohort study. Patients with AE, with or without SARS-CoV-2, were followed up every 3 months. Multivariate logistic regression was used to identify factors influencing neurological prognosis. Prediction models were created using nine machine-learning strategies.</div></div><div><h3>Results</h3><div>SARS-CoV-2 infection was noted in 241 of 308 patients. Multivariate logistic regression showed inactivated immunizations plus recombinant or adenovirus vector vaccines protected against SARS-CoV-2 infection (odds ratio [OR] 0.05, 95% confidence interval [CI] 0.01–0.52, <em>p</em> = 0.01). Three months after infection, 12.4% of patients with stable AE had a worse neurological prognosis, 6.2% failed to recover their working conditions. Pre-infection modified Rankin Scale (mRS) = 1 (OR 4.06, 95% CI 1.11–14.91, <em>p</em> = 0.04) and mRS > 1 (OR 13.4, 95% CI 3.31–54.02, <em>p</em> < 0.01), immunotherapy during infection (OR 5.1, 95% CI 1.65–15.79, <em>p</em> = 0.01), and SARS-CoV-2-related drowsiness (OR 19.5, 95% CI 5.42–70.34, <em>p</em> < 0.01) and gastrointestinal symptoms (OR 4.4, 95% CI 1.58–11.96, <em>p</em> < 0.01) were identified as risk factors for worse prognosis. The ranger model(<span><span>https://xingjieli1999.shinyapps.io/clinical_prediction_app/</span><svg><path></path></svg></span>) using these four parameters showed a discrimination accuracy of 0.96 (95% CI 0.94–0.99).</div></div><div><h3>Conclusions</h3><div>Patients with AE could experience exacerbated neurological symptoms following SARS-CoV-2 infection. Machine-learning algorithms showed feasibility of predicting prognoses based on clinical information in patients with AE.</div></div>","PeriodicalId":18958,"journal":{"name":"Multiple sclerosis and related disorders","volume":"109 ","pages":"Article 107028"},"PeriodicalIF":3.0000,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Multiple sclerosis and related disorders","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2211034826000647","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/2/6 0:00:00","PubModel":"Epub","JCR":"Q2","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Objective

To assess the neurological prognosis of patients with autoimmune encephalitis (AE) after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and construct risk prediction models for a worse prognosis.

Methods

This was a prospective, multicenter, observational cohort study. Patients with AE, with or without SARS-CoV-2, were followed up every 3 months. Multivariate logistic regression was used to identify factors influencing neurological prognosis. Prediction models were created using nine machine-learning strategies.

Results

SARS-CoV-2 infection was noted in 241 of 308 patients. Multivariate logistic regression showed inactivated immunizations plus recombinant or adenovirus vector vaccines protected against SARS-CoV-2 infection (odds ratio [OR] 0.05, 95% confidence interval [CI] 0.01–0.52, p = 0.01). Three months after infection, 12.4% of patients with stable AE had a worse neurological prognosis, 6.2% failed to recover their working conditions. Pre-infection modified Rankin Scale (mRS) = 1 (OR 4.06, 95% CI 1.11–14.91, p = 0.04) and mRS > 1 (OR 13.4, 95% CI 3.31–54.02, p < 0.01), immunotherapy during infection (OR 5.1, 95% CI 1.65–15.79, p = 0.01), and SARS-CoV-2-related drowsiness (OR 19.5, 95% CI 5.42–70.34, p < 0.01) and gastrointestinal symptoms (OR 4.4, 95% CI 1.58–11.96, p < 0.01) were identified as risk factors for worse prognosis. The ranger model(https://xingjieli1999.shinyapps.io/clinical_prediction_app/) using these four parameters showed a discrimination accuracy of 0.96 (95% CI 0.94–0.99).

Conclusions

Patients with AE could experience exacerbated neurological symptoms following SARS-CoV-2 infection. Machine-learning algorithms showed feasibility of predicting prognoses based on clinical information in patients with AE.

查看原文本刊更多论文

机器学习模型预测sars - cov -2感染自身免疫性脑炎患者的预后。

目的：评价自身免疫性脑炎（AE）患者感染严重急性呼吸综合征冠状病毒2 （SARS-CoV-2）后的神经系统预后，并建立预后不良的风险预测模型。方法：这是一项前瞻性、多中心、观察性队列研究。AE患者，不论有无SARS-CoV-2，每3个月随访一次。采用多因素logistic回归分析确定影响神经系统预后的因素。使用九种机器学习策略创建了预测模型。结果：308例患者中有241例出现SARS-CoV-2感染。多因素logistic回归显示，灭活免疫加重组或腺病毒载体疫苗可预防SARS-CoV-2感染（优势比[or] 0.05, 95%可信区间[CI] 0.01 ~ 0.52, p = 0.01）。感染3个月后，12.4%的稳定AE患者神经系统预后较差，6.2%的患者无法恢复工作状态。感染前改良Rankin量表(mRS) = 1 （OR 4.06, 95% CI 1.11-14.91, p = 0.04）和mRS bbb1 （OR 13.4, 95% CI 3.31-54.02, p < 0.01）、感染期间免疫治疗（OR 5.1, 95% CI 1.65-15.79, p = 0.01）、sars - cov -2相关嗜睡（OR 19.5, 95% CI 5.42-70.34, p < 0.01）和胃肠道症状（OR 4.4, 95% CI 1.58-11.96, p < 0.01）被确定为预后不良的危险因素。游侠模型（https://xingjieli1999.shinyapps）。io/clinical_prediction_app/)使用这四个参数的识别准确率为0.96 （95% CI 0.94-0.99）。结论：AE患者感染SARS-CoV-2后可出现神经系统症状加重。机器学习算法显示了基于AE患者临床信息预测预后的可行性。

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

Multiple sclerosis and related disorders CLINICAL NEUROLOGY-

CiteScore

5.80

自引率

20.00%

发文量

814

审稿时长

66 days

期刊介绍： Multiple Sclerosis is an area of ever expanding research and escalating publications. Multiple Sclerosis and Related Disorders is a wide ranging international journal supported by key researchers from all neuroscience domains that focus on MS and associated disease of the central nervous system. The primary aim of this new journal is the rapid publication of high quality original research in the field. Important secondary aims will be timely updates and editorials on important scientific and clinical care advances, controversies in the field, and invited opinion articles from current thought leaders on topical issues. One section of the journal will focus on teaching, written to enhance the practice of community and academic neurologists involved in the care of MS patients. Summaries of key articles written for a lay audience will be provided as an on-line resource. A team of four chief editors is supported by leading section editors who will commission and appraise original and review articles concerning: clinical neurology, neuroimaging, neuropathology, neuroepidemiology, therapeutics, genetics / transcriptomics, experimental models, neuroimmunology, biomarkers, neuropsychology, neurorehabilitation, measurement scales, teaching, neuroethics and lay communication.