Investigating the Utility of Multimodal Conversational Technology and Audiovisual Analytic Measures for the Assessment and Monitoring of Amyotrophic Lateral Sclerosis at Scale

A store-and-forward cloud-based telemonitoring system for automatic assessing dysarthria evolution in neurological diseases from video-recording analysis

“…Currently the preprint paper of Neumann et al [49] proposes the closest solution with respect to ours. Indeed, with NEMSI the authors present an internet-based multimodal dialogue system designed to gather necessary evidence for the identification or ongoing monitoring of patients with ALS.…”

Remote Inference of Cognitive Scores in ALS Patients Using a Picture Description

“…Every participant documented informed consent prior to participation. By using the modality.ai platform [10] -a dialog platform for the remote assessment and monitoring of ALS at scale,-the participants performed a series of verbal tasks every week: sustained vowel phonation, reading tasks, a measure of diadochokinetic (DDK) rate, and free speech via a picture description task. The latter task was found to be very useful for assessing cognitive impairment [7]- [9], and this is the one task we focus on for this study.…”

“…The ECAS test was administered remotely using a digitized version of the original ECAS test. By being part of the Speech Study, the participants record themselves at home using the same cloudbased platform employed for performing speech tasks [10]. Our inclusion criteria for selecting participants for this study is to have at least one recording of the participants performing the picture description task within +/-60 days of having administered the ECAS test.…”

“…Methodology used to infer ECAS sub-scores and total score. Participants performed the picture description task and ECAS test using modality.ai platform[10]. Linguistic and acoustic features are extracted and used to train linear regression models to infer the ECAS subs-cores and total score.…”