Nylander Alyssa, Sisodia Nikki, Henderson Kyra, Wijangco Jaeleene, Koshal Kanishka, Poole Shane, Dias Marcelo, Linz Nicklas, Tröger Johannes, König Alexandra, Hayward-Koennecke Helen, Pedotti Rosetta, Brown Ethan, Halabi Cathra, Staffaroni Adam, Bove Riley
UCSF Weill Institute for Neurosciences, San Francisco, CA, USA.
ki elements GmbH, Saarbrücken, Germany.
Mult Scler. 2025 Feb;31(2):231-241. doi: 10.1177/13524585241303855. Epub 2024 Dec 17.
Fatigue is a major "invisible" symptom in people with multiple sclerosis (PwMS), which may affect speech. Automated speech analysis is an objective, rapid tool to capture digital speech biomarkers linked to functional outcomes.
To use automated speech analysis to assess multiple sclerosis (MS) fatigue metrics.
Eighty-four PwMS completed scripted and spontaneous speech tasks; fatigue was assessed with Modified Fatigue Impact Scale (MFIS). Speech was processed using an automated speech analysis pipeline (ki elements: SIGMA speech processing library) to transcribe speech and extract features. Regression models assessed associations between speech features and fatigue and validated in a separate set of 30 participants.
Cohort characteristics were as follows: mean age 49.8 (standard deviation () = 13.6), 71.4% female, 85% relapsing-onset, median Expanded Disability Status Scale (EDSS) 2.5 (range: 0-6.5), mean MFIS 27.6 ( = 19.4), and 30% with MFIS > 38. MFIS moderately correlated with pitch ( = 0.32, = 0.005), pause duration ( = 0.33, = 0.007), and utterance duration ( = 0.31, = 0.0111). A logistic model using speech features from multiple tasks accurately classified MFIS in training (area under the curve (AUC) = 0.95, = 0.59, < 0.001) and test sets (AUC = 0.93, = 0.54, = 0.0222). Adjusting for EDSS, processing speed, and depression in sensitivity analyses did not impact model accuracy.
Fatigue may be assessed using simple, low-burden speech tasks that correlate with gold-standard subjective fatigue measures.
疲劳是多发性硬化症患者（PwMS）的主要“无形”症状，可能会影响言语。自动语音分析是一种客观、快速的工具，可用于捕捉与功能结果相关的数字语音生物标志物。
使用自动语音分析来评估多发性硬化症（MS）的疲劳指标。
84名PwMS完成了有脚本和自发的言语任务；使用改良疲劳影响量表（MFIS）评估疲劳。使用自动语音分析管道（关键要素：SIGMA语音处理库）处理语音，以转录语音并提取特征。回归模型评估语音特征与疲劳之间的关联，并在另一组30名参与者中进行验证。
队列特征如下：平均年龄49.8岁（标准差（SD）=13.6），71.4%为女性，85%为复发型，扩展残疾状态量表（EDSS）中位数为2.5（范围：0 - 6.5），平均MFIS为27.6（SD = 19.4），30%的MFIS>38。MFIS与音高（r = 0.32，P = 0.005）、停顿持续时间（r = 0.33，P = 0.007）和话语持续时间（r = 0.3仁P = 0.0111）中度相关。使用来自多个任务的语音特征的逻辑模型在训练集（曲线下面积（AUC）= 0.95，灵敏度 = 0.59，P < 0.001）和测试集（AUC = 0.93，灵敏度 = 0.54，P = 0.0222）中准确分类MFIS。在敏感性分析中，调整EDSS、处理速度和抑郁情况不会影响模型准确性。
可以使用与金标准主观疲劳测量相关的简单、低负担言语任务来评估疲劳。
