Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun 13:5:621-626.
doi: 10.1109/OJEMB.2023.3284798. eCollection 2024.

An Emotion-Driven Vocal Biomarker-Based PTSD Screening Tool

Thomas F Quatieri  1   2 , Jing Wang  1 , James R Williamson  1 , Richard DeLaura  1 , Tanya Talkar  1   2 , Nancy P Solomon  3   4 , Stefanie E Kuchinsky  3   4 , Megan Eitel  3 , Tracey Brickell  5   6   7   4 , Sara Lippa  6   4 , Kristin J Heaton  8 , Douglas S Brungart  3   4 , Louis French  5   6   4 , Rael Lange  5   6   7   4 , Jeffrey Palmer  1 , Hayley Reynolds  1
Affiliations

An Emotion-Driven Vocal Biomarker-Based PTSD Screening Tool

Thomas F Quatieri et al. IEEE Open J Eng Med Biol. .

Abstract

Goal: This paper introduces an automated post-traumatic stress disorder (PTSD) screening tool that could potentially be used as a self-assessment or inserted into routine medical visits to aid in PTSD diagnosis and treatment. Methods: With an emotion estimation algorithm providing arousal (excited to calm) and valence (pleasure to displeasure) levels through discourse, we select regions of the acoustic signal that are most salient for PTSD detection. Our algorithm was tested on a subset of data from the DVBIC-TBICoE TBI Study, which contains PTSD Check List Civilian (PCL-C) assessment scores. Results: Speech from low-arousal and positive-valence regions provide the highest discrimination for PTSD. Our model achieved an AUC (area under the curve) of 0.80 in detecting PCL-C ratings, outperforming models with no emotion filtering (AUC = 0.68). Conclusions: This result suggests that emotion drives the selection of the most salient temporal regions of an audio recording for PTSD detection.

Keywords: Emotional digital twin; PTSD; emotion sensing; neuromotor coordination; vocal biomarkers.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.
Schematic of PTSD prediction system overview. Continuous estimates of arousal and valence provide the basis of an ‘emotion filter’ that selects regions of the acoustic signal most salient for PTSD detection.
Fig. 2.
Fig. 2.
Prediction/ground truth scatter plot corresponding to the fusion of all eight feature sets. A linear fit is shown as a visual aid while the Spearman correlation yields R = 0.47. The corresponding AUC (Fig. 3) is 0.80.
Fig. 3.
Fig. 3.
ROC curves corresponding to the comparison of PTSD detection performance with positive emotion filtering (low arousal & positive valence) (AUC = 0.80), negative emotion filtering (high arousal & negative valence) (AUC = 0.70), and no emotion filtering (AUC = 0.68).
See this image and copyright information in PMC

References

    1. American Psychiatric Association, American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM V), 4th ed. WA, DC, USA: Amer. Psychiatric Assoc., 2013.
    1. Bylsma L. M., Morris B. H., and Rottenberg J., “A meta-analysis of emotional reactivity in major depressive disorder,” Clin. Psychol. Rev., vol. 28, no. 4, pp. 676–691, 2008. - PubMed
    1. Boersma P., “Accurate short-term analysis of the fundamental frequency and the harmonics-to-noise ratio of a sampled sound,” Proc. Inst. Phonetic Sci., vol. 17, no. 1193, pp. 97–110, 1993.
    1. Boersma P. and Weenink D., “Praat: Doing phonetics by computer [computer program],” Version 6.3.01, 1992–2022. Accessed: Jan. 23, 2022. [Online]. Available: http://www.praat.org/
    1. Borgström B. J. and Brandstein M. S., “Speech enhancement via attention masking network (SEAMNET): An end-to-end system for joint suppression of noise and reverberation,” IEEE/ACM Trans. Audio, Speech, Lang. Process., vol. 29, pp. 515–526, 2021.