Acoustic and prosodic information for home monitoring of bipolar disorder

Abstract

Epidemiological studies suggest that bipolar disorder has a prevalence of about 1% in European countries, becoming one of the most disabling illnesses in working age adults, and often long-term and persistent with complex management and treatment. Therefore, the capacity of home monitoring for patients with this disorder is crucial for their quality of life. The current paper introduces the use of speech-based information as an easy-to-record, ubiquitous and non-intrusive health sensor suitable for home monitoring, and its application in the framework on the NYMPHA-MD project. Some preliminary results also show the potential of acoustic and prosodic features to detect and classify bipolar disorder, by predicting the values of the Hamilton Depression Rating Scale (HDRS) and the Young Mania Rating Scale (YMRS) from speech.

Emerging technologies for supporting person-centred integrated home health care

“…This special issue also features a very interesting case study of using acoustic and prosodic information as an easy-to-record, ubiquitous and non-intrusive health sensor for home-based monitoring of bipolar disorder. 17 The study results show that the speech-based algorithm is a potential tool for predicting different mood statuses in patients with bipolar disorders. Moreover, an application, MoodRecord, is also featured in the solution to provide a new way to manage patient monitoring in addition to regular visits to clinics, leveraging the functions of estimating patient mood and a further monitoring and supervision of patients at home.…”

Mobile eHealth Platform for Home Monitoring of Bipolar Disorder

“…Mood detection from voice. Speech is a powerful identifier for the detection of bipolar disorder, as it has been shown in several works [21,19,11]. Moreover, voice is a non-intrusive and ubiquitous sensor.…”

“…This dataset is composed of 30 valid recordings with the corresponding evaluations by doctors. Although the dataset is small and most of the recordings were done with users in an euthymic state, the results obtained using support vector regression with a radial kernel showed that, while HDRS estimation achieved the best result by using low-level voice quality and formant features, the best result when estimating YMRS was achieved by using the whole combination of voice quality, formant, and prosodic features [11]. This can be explained by the fact that mania state is more expressive in terms of speech, and thus the need of prosodic information is much more relevant.…”

A multimodal fusion enabled ensemble approach for human activity recognition in smart homes

“…Home monitoring can be used as a possible solution to provide information on behavioral deviations and cognitive assistance. 3,4 Up to date, three types of sensors have been used to monitor ADL: vision-based, 5,6 noninvasive sensor-based, [7][8][9] and hybrid methods. 10 Vision-based approaches depend on video cameras or depth cameras for accurate activity recognition.…”

“…Home monitoring can be used as a possible solution to provide information on behavioral deviations and cognitive assistance. 3,4…”