Identifying multiple sclerosis subtypes using unsupervised machine learning and MRI data.
Identifying multiple sclerosis subtypes using unsupervised machine learning and MRI data.
复制标题
DOI：
10.1038/s41467-021-22265-2
复制
发表时间：
2021-04-06
影响因子：
16.6
通讯作者：
Ciccarelli O
中科院分区：
文献类型：
--
作者：
Eshaghi A;Young AL;Wijeratne PA;Prados F;Arnold DL;Narayanan S;Guttmann CRG;Barkhof F;Alexander DC;Thompson AJ;Chard D;Ciccarelli O
多发性硬化症（MS）可根据临床演变分为四种表型。这些表型的病理生理学界限尚不清楚，限制了治疗分层。机器学习可以使用多维数据识别具有相似特征的群体。在这里，为了根据病理特征对MS亚型进行分类，我们将无监督机器学习应用于先前发表的研究中获得的脑部MRI扫描。我们使用来自6322名MS患者的训练数据集来定义基于MRI的亚型，并使用3068名患者的独立队列进行验证。基于最早期的异常，我们将MS亚型定义为皮质主导型、外观正常的白色物质主导型和病变主导型。病变导致的亚型患者确诊残疾进展（CDP）的风险最高，复发率最高。病变导致的MS亚型患者在选定的临床试验中显示出积极的治疗反应。我们的研究结果表明，基于MRI的亚型可预测MS残疾进展和对治疗的反应，并可用于定义干预性试验中的患者组。多发性硬化症是一种异质性进行性疾病。在这里，作者使用一种无监督的机器学习算法来确定多发性硬化症的亚型、进展以及对基于神经影像学数据的潜在治疗的反应。
Multiple sclerosis (MS) can be divided into four phenotypes based on clinical evolution. The pathophysiological boundaries of these phenotypes are unclear, limiting treatment stratification. Machine learning can identify groups with similar features using multidimensional data. Here, to classify MS subtypes based on pathological features, we apply unsupervised machine learning to brain MRI scans acquired in previously published studies. We use a training dataset from 6322 MS patients to define MRI-based subtypes and an independent cohort of 3068 patients for validation. Based on the earliest abnormalities, we define MS subtypes as cortex-led, normal-appearing white matter-led, and lesion-led. People with the lesion-led subtype have the highest risk of confirmed disability progression (CDP) and the highest relapse rate. People with the lesion-led MS subtype show positive treatment response in selected clinical trials. Our findings suggest that MRI-based subtypes predict MS disability progression and response to treatment and may be used to define groups of patients in interventional trials. Multiple sclerosis is a heterogeneous progressive disease. Here, the authors use an unsupervised machine learning algorithm to determine multiple sclerosis subtypes, progression, and response to potential therapeutic treatments based on neuroimaging data.
登录
查看更多内容
影响因子：
9.9
作者：
Lublin FD;Reingold SC;Cohen JA;Cutter GR;Sørensen PS;Thompson AJ;Wolinsky JS;Balcer LJ;Banwell B;Barkhof F;Bebo B Jr;Calabresi PA;Clanet M;Comi G;Fox RJ;Freedman MS;Goodman AD;Inglese M;Kappos L;Kieseier BC;Lincoln JA;Lubetzki C;Miller AE;Montalban X;O'Connor PW;Petkau J;Pozzilli C;Rudick RA;Sormani MP;Stüve O;Waubant E;Polman CH
通讯作者：
Polman CH
DOI：
10.1093/brain/aws242
发表时间：
2012-10
期刊：
Brain : a journal of neurology
影响因子：
--
作者：
Kolasinski J;Stagg CJ;Chance SA;Deluca GC;Esiri MM;Chang EH;Palace JA;McNab JA;Jenkinson M;Miller KL;Johansen-Berg H
通讯作者：
Johansen-Berg H
影响因子：
11.2
作者：
Hawker, Kathleen;O'Connor, Paul;Smith, Craig H.
通讯作者：
Smith, Craig H.
影响因子：
168.9
作者：
Lublin, Fred;Miller, David H.;Kappos, Ludwig
通讯作者：
Kappos, Ludwig
影响因子：
9.9
作者：
Freedman, M. S.;Bar-Or, A.;Verco, T.
通讯作者：
Verco, T.