How to Design a Relevant Corpus for Sleepiness Detection Through Voice?

Sleep deprivation detected by voice analysis

Sleep deprivation has an ever-increasing impact on individuals and societies. Yet, to date, there is no quick and objective test for sleep deprivation. Here, we used automated acoustic analyses of the voice to detect sleep deprivation. Building on current machine-learning approaches, we focused on interpretability by introducing two novel ideas: the use of a fully generic auditory representation as input feature space, combined with an interpretation technique based on reverse correlation. The auditory representation consisted of a spectro-temporal modulation analysis derived from neurophysiology. The interpretation method aimed to reveal the regions of the auditory representation that supported the classifiers' decisions. Results showed that generic auditory features could be used to detect sleep deprivation successfully, with an accuracy comparable to state-of-the-art speech features. Furthermore, the interpretation revealed two distinct effects of sleep deprivation on the voice: changes in slow temporal modulations related to prosody and changes in spectral features related to voice quality. Importantly, the relative balance of the two effects varied widely across individuals, even though the amount of sleep deprivation was controlled, thus confirming the need to characterize sleep deprivation at the individual level. Moreover, while the prosody factor correlated with subjective sleepiness reports, the voice quality factor did not, consistent with the presence of both explicit and implicit consequences of sleep deprivation. Overall, the findings show that individual effects of sleep deprivation may be observed in vocal biomarkers. Future investigations correlating such markers with objective physiological measures of sleep deprivation could enable "sleep stethoscopes" for the cost-effective diagnosis of the individual effects of sleep deprivation.

How Does Comparison With Artificial Intelligence Shed Light on the Way Clinicians Reason? A Cross-Talk Perspective

In order to create a dynamic for the psychiatry of the future, bringing together digital technology and clinical practice, we propose in this paper a cross-teaching translational roadmap comparing clinical reasoning with computational reasoning. Based on the relevant literature on clinical ways of thinking, we differentiate the process of clinical judgment into four main stages: collection of variables, theoretical background, construction of the model, and use of the model. We detail, for each step, parallels between: i) clinical reasoning; ii) the ML engineer methodology to build a ML model; iii) and the ML model itself. Such analysis supports the understanding of the empirical practice of each of the disciplines (psychiatry and ML engineering). Thus, ML does not only bring methods to the clinician, but also supports educational issues for clinical practice. Psychiatry can rely on developments in ML reasoning to shed light on its own practice in a clever way. In return, this analysis highlights the importance of subjectivity of the ML engineers and their methodologies.