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
. 2022 Jan 23;22(3):846.
doi: 10.3390/s22030846.

Mental Health Intent Recognition for Arabic-Speaking Patients Using the Mini International Neuropsychiatric Interview (MINI) and BERT Model

Ridha Mezzi  1 , Aymen Yahyaoui  1   2 , Mohamed Wassim Krir  3 , Wadii Boulila  4   5 , Anis Koubaa  4
Affiliations

Mental Health Intent Recognition for Arabic-Speaking Patients Using the Mini International Neuropsychiatric Interview (MINI) and BERT Model

Ridha Mezzi et al. Sensors (Basel). .

Abstract

For many years, mental health has been hidden behind a veil of shame and prejudice. In 2017, studies claimed that 10.7% of the global population suffered from mental health disorders. Recently, people started seeking relaxing treatment through technology, which enhanced and expanded mental health care, especially during the COVID-19 pandemic, where the use of mental health forums, websites, and applications has increased by 95%. However, these solutions still have many limits, as existing mental health technologies are not meant for everyone. In this work, an up-to-date literature review on state-of-the-art of mental health and healthcare solutions is provided. Then, we focus on Arab-speaking patients and propose an intelligent tool for mental health intent recognition. The proposed system uses the concepts of intent recognition to make mental health diagnoses based on a bidirectional encoder representations from transformers (BERT) model and the International Neuropsychiatric Interview (MINI). Experiments are conducted using a dataset collected at the Military Hospital of Tunis in Tunisia. Results show excellent performance of the proposed system (the accuracy is over 92%, the precision, recall, and F1 scores are over 94%) in mental health patient diagnosis for five aspects (depression, suicidality, panic disorder, social phobia, and adjustment disorder). In addition, the tool was tested and evaluated by medical staff at the Military Hospital of Tunis, who found it very interesting to help decision-making and prioritizing patient appointment scheduling, especially with a high number of treated patients every day.

Keywords: BERT model; MINI; intent recognition; machine learning; mental health; natural language processing; psychiatry.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure A1
Figure A1
Packages and libraries import.
Figure A2
Figure A2
Dataset and model load with Python on Google Colab.
Figure A3
Figure A3
The tokenization code.
Figure A4
Figure A4
Model creation.
Figure A5
Figure A5
Compiling the model.
Figure A6
Figure A6
Fitting code.
Figure 1
Figure 1
Most common scenarios of depression using MINI in our system.
Figure 2
Figure 2
BERT architecture.
Figure 3
Figure 3
System’s global architecture.
Figure 4
Figure 4
3D human avatar–patient interaction.
Figure 5
Figure 5
Speech-to-Text process.
Figure 6
Figure 6
A sample from the suicidality dataset in Tunisian Darija and in English.
Figure 7
Figure 7
Class distribution of the datasets. (a) Depression dataset. (b) Suicidality dataset. (c) Panic disorder dataset. (d) Social phobia dataset. (e) Adjustement disorder dataset.
Figure 7
Figure 7
Class distribution of the datasets. (a) Depression dataset. (b) Suicidality dataset. (c) Panic disorder dataset. (d) Social phobia dataset. (e) Adjustement disorder dataset.
Figure 8
Figure 8
Text classification with BERT.
Figure 9
Figure 9
Text classification with BERT in detail.
Figure 10
Figure 10
BERT model confusion matrix when using the depression model.
Figure 11
Figure 11
BERT model confusion matrix when using the suicidality model.
Figure 12
Figure 12
BERT model confusion matrix when the using the panic disorder model.
Figure 13
Figure 13
BERT model confusion matrix when using the social phobia model.
Figure 14
Figure 14
BERT model confusion matrix when using the adjustment disorder model.
Figure 15
Figure 15
ROC curves when using the depression model.
Figure 16
Figure 16
ROC curves when using the suicidality model.
Figure 17
Figure 17
ROC curves when using the panic disorder model.
Figure 18
Figure 18
ROC curves when using the social phobia model.
Figure 19
Figure 19
ROC curves when using the adjustment disorder model.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Carpenter-Song E., Noel V.A., Acquilano S.C., Drake R.E. Real-world technology use among people with mental illnesses: Qualitative study. JMIR Ment. Health. 2018;5:e10652. doi: 10.2196/10652. - DOI - PMC - PubMed
    1. Sheehan D.V., Lecrubier Y., Sheehan K.H., Amorim P., Janavs J., Weiller E., Hergueta T., Baker R., Dunbar G.C. The Mini-International Neuropsychiatric Interview (MINI): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry. 1998;59:22–33. - PubMed
    1. Le Glaz A., Haralambous Y., Kim-Dufor D.H., Lenca P., Billot R., Ryan T.C., Marsh J., Devylder J., Walter M., Berrouiguet S., et al. Machine learning and natural language processing in mental health: Systematic review. J. Med. Internet Res. 2021;23:e15708. doi: 10.2196/15708. - DOI - PMC - PubMed
    1. Devlin J., Chang M.W. Open sourcing BERT: State-of-the-art pre-training for natural language processing. [(accessed on 30 March 2021)];Google AI Blog. 2018 Available online: https://ai.googleblog.com/2018/11/open-sourcing-bert-state-of-art-pre.html.
    1. Acheampong F.A., Nunoo-Mensah H., Chen W. Transformer models for text-based emotion detection: A review of BERT-based approaches. Artif. Intell. Rev. 2021;54:5789–5829. doi: 10.1007/s10462-021-09958-2. - DOI