Data for: SPiQE: an automated analytical tool for detecting and characterising fasciculations in amyotrophic lateral sclerosis

Four files: 1. Mean noise bands and optimal amplitude inclusion thresholds for 599 one-minute recordings. 2. Sensitivity, specificity and classification accuracy scores for 80 one-minute recordings across multiple thresholds for each model (1 and 2). 3. Sensitivity, specificity and classification accuracy scores for 80 one-minute recordings for different thresholds of amplitude exclusion threshold 4. Comparison of manual and automated fasciculation counts using optimal model.本数据集包含如下4个文件： 1. 针对599份1分钟时长的录音，提供其平均噪声频段（noise bands）与最优振幅纳入阈值（optimal amplitude inclusion thresholds）。 2. 针对模型1与模型2，在多阈值条件下，80份1分钟时长的录音对应的灵敏度（Sensitivity）、特异度（Specificity）及分类准确率（classification accuracy）得分。 3. 针对不同取值的振幅排除阈值（amplitude exclusion threshold），80份1分钟时长的录音对应的灵敏度、特异度及分类准确率得分。 4. 采用最优模型时，人工与自动化肌束颤动（fasciculation）计数结果的对比。