Contributions
Abstract: EP1274
Type: ePoster
Abstract Category: Clinical aspects of MS - 1 Diagnosis and differential diagnosis
Prevalence of depression in Multiple Sclerosis (MS) is high (from 27 to 54%) and depression is under-diagnosed. Because of symptoms overlap between depression and MS, diagnostic questionnaires might fail to diagnose depression in MS. Voice study (comprising the uses of voice) is expanding in several fields since the last 10 years: for example, Tsanas et al., (2012) showed that Parkinson disease could be detected precociously from voice thanks to acoustic analyses. To explore if voice analysis could help clinicians to diagnose depression in MS patients, we recorded 2 groups of female speakers: healthy women (28 non-depressed and 25 depressed); and MS patients without dysarthria (32 non-depressed and 30 depressed). Depression was diagnosed according to the Zigmond & Snaith HAD Scale. Our study consists of combining both acoustic measures and judgments made by non-expert listeners, from recorded voices. For each voice recorded, the listeners had to rate pleasantness and depression. We found significant differences in 4 voice parameters between the 2 non-MS groups and the 2 MS groups. For the non-MS groups, the values for the pitch (p≤0.01), the amplitude (p≤0.001), the speech rate (p≤0.05) and the variation of the pitch (p≤0.01) were higher than in the MS groups. In other words, the voices from the MS groups are lower (in intensity), deeper, slower and less expressive. We found differences between the depressed and the non-depressed women inside each group for 2 parameters: the voices from the depressed women were less expressive (p≤0.05) and lower (p≤0.001). The results from the listeners' judgments are similar to those from the voice analyses since non-MS/MS groups and non-depressed/depressed groups could be distinguished (Depression and MS make the voices sound less pleasant). In conclusion, our results highlight that it is possible 1) to detect depression in MS patients using voice analysis; and 2) to distinguish healthy people and MS patients, even if the MS patients with a dysarthria have been excluded.
Disclosure:
Laetitia Bruckert: this research has been supported by GENZYME Company.
Olivier Heinzlef: nothing to disclose
Sarah Jeannin: nothing to disclose
Gérard Leboucher: nothing to disclose
Marie-Claire Gay: nothing to disclose
Abstract: EP1274
Type: ePoster
Abstract Category: Clinical aspects of MS - 1 Diagnosis and differential diagnosis
Prevalence of depression in Multiple Sclerosis (MS) is high (from 27 to 54%) and depression is under-diagnosed. Because of symptoms overlap between depression and MS, diagnostic questionnaires might fail to diagnose depression in MS. Voice study (comprising the uses of voice) is expanding in several fields since the last 10 years: for example, Tsanas et al., (2012) showed that Parkinson disease could be detected precociously from voice thanks to acoustic analyses. To explore if voice analysis could help clinicians to diagnose depression in MS patients, we recorded 2 groups of female speakers: healthy women (28 non-depressed and 25 depressed); and MS patients without dysarthria (32 non-depressed and 30 depressed). Depression was diagnosed according to the Zigmond & Snaith HAD Scale. Our study consists of combining both acoustic measures and judgments made by non-expert listeners, from recorded voices. For each voice recorded, the listeners had to rate pleasantness and depression. We found significant differences in 4 voice parameters between the 2 non-MS groups and the 2 MS groups. For the non-MS groups, the values for the pitch (p≤0.01), the amplitude (p≤0.001), the speech rate (p≤0.05) and the variation of the pitch (p≤0.01) were higher than in the MS groups. In other words, the voices from the MS groups are lower (in intensity), deeper, slower and less expressive. We found differences between the depressed and the non-depressed women inside each group for 2 parameters: the voices from the depressed women were less expressive (p≤0.05) and lower (p≤0.001). The results from the listeners' judgments are similar to those from the voice analyses since non-MS/MS groups and non-depressed/depressed groups could be distinguished (Depression and MS make the voices sound less pleasant). In conclusion, our results highlight that it is possible 1) to detect depression in MS patients using voice analysis; and 2) to distinguish healthy people and MS patients, even if the MS patients with a dysarthria have been excluded.
Disclosure:
Laetitia Bruckert: this research has been supported by GENZYME Company.
Olivier Heinzlef: nothing to disclose
Sarah Jeannin: nothing to disclose
Gérard Leboucher: nothing to disclose
Marie-Claire Gay: nothing to disclose