Information technology in psychiatry: applications, development, types
About company
Our solutions
Resource center

Contact us to find out more and start working with us





    * Required field

    SberMedAI LLC uses Cookies (files with information about past visits to the website) for service personalization and user convenience. To continue browsing the website, you need to enable cookies and accept our Cookies policy. You may disable cookies, but the website will become unavailable. Cookies policy

    Accept

    Computer Technologies and Psychiatry

    February 17 2023

    |

    Reading time 14 minutes

    Contents

    Psychiatry uses conversations with patients as a key diagnostic technique. Artificial intelligence (AI) helps a doctor find hidden patterns in other data, such as medical images, electrophysiological signals or questionnaires. Algorithms also provide social support by helping patients maintain their mental health.

    Computer visualization in psychiatry

    Detecting psychiatric diseases based on diagnostic images, e.g. MRI images, is a challenging task. Changes in the brain structure captured in the image may be non-specific and similar to those seen in other disorders, including neurodegenerative diseases. A doctor analyzes clinical findings in order to reliably distinguish between these conditions.1

    Technology helps to make studies more informative. Computer models perform computations to distinguish between healthy individuals and those showing signs of a disorder. A doctor is provided with additional diagnostic information that helps to detect a number of disorders, such as:2

    • Major depressive disorder;

    • Schizophrenia;

    • Attention-deficit/hyperactivity disorder (ADHD);

    • Autism spectrum disorders.

    Researchers from the University of Alberta in Canada have studied heredity as a risk factor for schizophrenia. Their information model, EMPaSchiz, predicts the likelihood of the disease in first-degree relatives of a schizophrenia patient. The analysis was performed for functional magnetic resonance imaging (fMRI) data.3

    AI classified 14 out of 57 participants as potential schizophrenia patients. They were invited to complete a questionnaire, which revealed schizotypal personality features. The use of technology cannot substitute for an examination by a medical specialist, but it helps to identify at-risk patients.3

    AI speeds up fMRI data analysis as part of various clinical tasks. For instance, the Resting State fMRI model developed by SberMedAI maps functional areas of the brain. The technological solution analyzes fMRI data and highlights visual, motor and auditory areas in different colors. The findings produced by the service help to plan surgical interventions in cancer patients.

    Use of AI in EEG studies

    A doctor placing EEG electrodes

    An electroencephalogram (EEG) is a test used in psychiatry that measures electrical activity in the brain using electrodes attached to a patient’s scalp. Electrical impulses generated by nerve cells are shown as wavy lines on an EEG recording.4

    Schizophrenia is a severe psychiatric disorder accompanied by hallucinations, unusual beliefs (delusions) and a loss of interest in everyday activities.5 Organization (WHO), schizophrenia affects approximately 24 million people worldwide.6 Timely diagnosis and treatment help to control the symptoms.7

    Researchers hope to increase the relevance of EEG data for diagnosing schizophrenia by applying information algorithms. In one study, researchers represented EEG signals in a new format and trained AI to analyze them. The technique involves several steps:8

    1. EEG data are transformed into red-green-blue (RGB) images, with brain wave recordings displayed as RGB gradients.

    2. A neural network processes the images and performs the necessary computations.

    3. AI makes a conjecture as to whether EEG data indicate a healthy individual or an individual suffering from a disorder.

    The results were inspiring: the information method achieved an accuracy of 99.22%. However, so far the study has covered only 54 patients, and the researchers plan to expand its scope.8

    A research team from Iceland has used a computer algorithm to study risk factors for seasonal affective disorder (SAD). People with SAD may exhibit depressive symptoms in winter, with their condition improving in spring.9

    The study was conducted in summer. Participants were asked to undergo a number of tests, including an EEG, and complete a questionnaire in order to assess their current emotional state. Artificial intelligence analyzed the data to predict the severity of depressive symptoms in winter. The model achieved 86% specificity and 81% sensitivity.9

    Information models also help to study electrical activity in the brain associated with other disorders, such as epilepsy. The EEG Epilepsy model detects signs of focal epileptiform discharges (FEDs). FED characteristics can help locate the epileptic focus. The service enables the automated processing of EEG signals and displays the results within five minutes.

    Virtual reality technology

    A woman using virtual reality technology in a laboratory

    Virtual reality (VR) technology immerses users in a simulated world. This involves using special devices: a VR headset equipped with a display, headphones, as well as devices for manual manipulations and navigation in a three-dimensional world.10

    Interest in the use of VR in psychiatry stems from new opportunities for diagnosis and treatment offered by the technology:11

    • Removing psychological barriers, including stigmatization and a fear of conventional treatment methods;

    • An opportunity to provide care in various economic situations and geographical regions;

    • Tailoring informationmodels to the needs of a patient;

    • Improving interaction with the surroundings.

    Reducing fears through virtual exposure

    Specialists from Canada have combined VR and cognitive behavioral therapy techniques to help people suffering from social anxiety disorder. Patients were offered an opportunity to play out social scenarios causing fear and anxiety, e.g. giving a presentation to a big audience in a conference room, in a virtual world. A patient used a wireless computer mouse and a headset to interact with virtual interlocutors.12

    Every session was monitored by a therapist. The researchers pointed out that the use of information models of the surroundings had improved therapy outcomes. It is easier for a therapist to manage the scenarios by selecting locations and lines of dialogue. At the same time, this method helps to ensure the confidentiality of therapy.12

    Researchers from Oxford have used VR to treat persecutory delusions. In psychiatry, this term denotes a person’s mistaken belief that people around them are watching them and want to harm them.13

    Patients were placed in a simulated environment, e.g. the subway or an elevator. As the patient became accustomed to the computer-generated reality, new virtual individuals were added to it. Participants of the study were offered an opportunity to:13

    • Realize that their fears were unfounded;

    • Develop alternative behavior strategies;

    • Learn to manage stress.

    It is expected that the use of AI will help patients feel more confident in real-life social situations.13

    Insights provided by VR into the nature of various disorders

    Researchers from South Korea have used technology to assess risk factors that increase alcohol craving. The study covered 14 patients with alcohol dependence, who had abstained from drinking alcohol for at least three weeks. The control group consisted of 14 healthy volunteers.14

    The technology simulated a meeting in the street and in a pub:14

    1. The informationmodel displayed alcohol cues, e.g. a glass of alcoholic drink.

    2. A virtual avatar exerted social pressure by inviting the participants to have a drink and giving various arguments.

    3. The participants used a mouse to measure their levels of alcohol craving in various situations on a scale.

    The researchers found that alcohol cues could exacerbate existing dependence. This is possible even if an individual has learned to cope with social pressure. The researchers believe that the use of VR can help learn to overcome these challenges.14

    In psychiatry, the term “anorexia” denotes an eating disorder. Patients with anorexia have:15,16

    • An abnormally low body weight;

    • A fear of gaining weight;

    • A distorted image of their bodies: a person thinks that they are overweight even when they are underweight.

    In an experiment conducted in the Netherlands, participants with anorexia temporarily “inhabited” a virtual body. The information model had measurements (shoulder, abdomen and hip width) considered to be healthy according to the WHO.17

    In the computer simulation, the patients experienced additional tactile sensations. An experimenter touched their abdomen with a special device equipped with a movement sensor. The participants saw the touching movements displayed in a VR environment, which enhanced their perception of the virtual body as their own body.17

    The researchers compared body size estimation by the participants before and after the computer experiment. As a result, body size overestimation was reduced, even though it remained considerable.17

    Remote technology in psychiatry

    A woman consulting a doctor via a video call using a smartphone

    Telepsychiatry is the process of providing psychiatric care from a distance through technology. Various channels of communication, such as videoconferencing, can be used to perform a number of tasks:18

    • Psychiatric evaluations;

    • Remote individual, group and family therapy sessions;

    • Patient education and consultations;

    • Medication management.

    The American Psychiatric Association (APA) has recognized the effectiveness of telepsychiatry. Technology has helped to achieve significant progress in the treatment of post-traumatic stress disorder (PTSD), depression and ADHD.18

    This method helps to address a number of challenges in psychiatry:19

    • Enhance the monitoring of clinical symptoms and medication;

    • Overcome internalized stigma;

    • Ensure the anonymity of care and its availability at any time;

    • Provide care in a patient’s native language.

    The use of telepsychiatry increased during the COVID-19 pandemic and the associated lockdown. According to the National Health Service, between February and June 2020, the number of remote consultations in the UK saw a 3.5-fold to 6-fold increase. Patients continued to receive care, which ensured the continuity of treatment.20

    Barriers to the use of telepsychiatry include:21

    • Difficulties in establishing rapport between clinicians and patients;

    • The need for healthcare workers to attend to a patient in times of crisis;

    • Distrust of technology;

    • Data privacy concerns;

    • Technical issues related to poor sound and video quality;

    • Legal restrictions.

    Video consultations in psychiatry help to provide equal access to specialist care. However, there are a number of organizational issues that need to be addressed in order to make the process more convenient for everyone involved.

    Use of mobile apps to maintain mental health

    AI-powered mobile apps help to treat various disorders. For instance, AI Resp developed by SberMedAI detects symptoms of respiratory diseases based on audio recordings of breathing, coughing and speech. AI Skin makes a tentative diagnosis of seven types of skin rashes based on photos. The diagnosis made by AI then needs to be verified by a doctor.

    In psychiatry, mobile apps are used to provide emotional support to users. They encourage users to look after their mental health by:22

    • Helping to learn relaxation and emotion management techniques;

    • Helping to control negative thoughts;

    • Enabling users to keep a mood diary;

    • Enabling people faced with similar problems to share their experience;

    • Providing contact details of specialized services and organizations.

    The use of these apps helps to cope with anxiety and depression. An important feature of the services is that they are designed to ensure confidentiality and user anonymity.22

    There are apps targeted at people with schizophrenia. These services perform a number of functions:23

    • Suggesting exercises for psychological recovery;

    • Helping to monitor symptoms and manage treatment;

    • Providing access to background information on mental health.

    However, users need to be careful when choosing an app: the information may be out of date or stigmatizing.23

    How artificial intelligence can drive a transformation in psychiatry

    Algorithms used in psychiatry are able to process large amounts of data. Potential applications include:24

    • Patient monitoring, including activity tracking at night;

    • Developing personalized treatment regimens;

    • Monitoring risk factors for a number of disorders;

    • Enhancing communication between doctors and patients.

    Technology can distinguish between healthy individuals and those showing signs of disorders, achieving an impressive level of classification accuracy (up to 99%). Researchers try various computational approaches and test algorithms. However, some solutions are still at the laboratory stage and are awaiting clinical trials. A lot of studies cover only a small number of patients. To produce objective findings, technologies should be tested on a large sample.25

    AI has considerable potential for application in psychiatry. Possible ways to unlock this potential include addressing algorithm biases and the “black box” problem, and studying biological characteristics of the brain along with psychosocial factors.26 A doctor can establish a correlation between the characteristics of a patient’s consciousness and computation results to arrive at the correct conclusion.

    Sources

    1. Oh J, Oh BL, Lee KU, Chae JH, Yun K. Identifying Schizophrenia Using Structural MRI With a Deep Learning Algorithm. Front Psychiatry. 2020 Feb 3;11:16. doi: 10.3389/fpsyt.2020.00016. PMID: 32116837; PMCID: PMC7008229.

    2. Zhang Z, Li G, Xu Y, Tang X. Application of Artificial Intelligence in the MRI Classification Task of Human Brain Neurological and Psychiatric Diseases: A Scoping Review. Diagnostics (Basel). 2021 Aug 3;11(8):1402. doi: 10.3390/diagnostics11081402. PMID: 34441336; PMCID: PMC8392727.

    3. Kalmady SV, Paul AK, Greiner R, Agrawal R, Amaresha AC, Shivakumar V, Narayanaswamy JC, Greenshaw AJ, Dursun SM, Venkatasubramanian G. Extending schizophrenia diagnostic model to predict schizotypy in first-degree relatives. NPJ Schizophr. 2020 Nov 6;6(1):30. doi: 10.1038/s41537-020-00119-y. PMID: 33159092; PMCID: PMC7648110.

    4. EEG (electroencephalogram) [Online resource]: Mayo Clinic. URL:https://www.mayoclinic.org/tests-procedures/eeg/about/pac-20393875.

    5. Overview – Schizophrenia [Online resource]: NHS. URL:https://www.nhs.uk/mental-health/conditions/schizophrenia/overview/.

    6. Schizophrenia [Online resource]: World Health Organization. URL:https://www.who.int/news-room/fact-sheets/detail/schizophrenia.

    7. Schizophrenia [Online resource]: National Institute of Mental Health. URL:https://www.nimh.nih.gov/health/topics/schizophrenia.

    8. Sun J, Cao R, Zhou M, Hussain W, Wang B, Xue J, Xiang J. A hybrid deep neural network for classification of schizophrenia using EEG Data. Sci Rep. 2021 Feb 25;11(1):4706. doi: 10.1038/s41598-021-83350-6. PMID: 33633134; PMCID: PMC7907145.

    9. Höller Y, Urbschat MM, Kristófersson GK, Ólafsson RP. Predictability of Seasonal Mood Fluctuations Based on Self-Report Questionnaires and EEG Biomarkers in a Non-clinical Sample. Front Psychiatry. 2022 Apr 8;13:870079. doi: 10.3389/fpsyt.2022.870079. PMID: 35463521; PMCID: PMC9030950.

    10. Li L, Yu F, Shi D, Shi J, Tian Z, Yang J, Wang X, Jiang Q. Application of virtual reality technology in clinical medicine. Am J Transl Res. 2017 Sep 15;9(9):3867-3880. PMID: 28979666; PMCID: PMC5622235.

    11. Kim S, Kim E. The Use of Virtual Reality in Psychiatry: A Review. Soa Chongsonyon Chongsin Uihak. 2020 Jan 1;31(1):26-32. doi: 10.5765/jkacap.190037. PMID: 32612410; PMCID: PMC7324842.

    12. Bouchard S, Dumoulin S, Robillard G, Guitard T, Klinger É, Forget H, Loranger C, Roucaut FX. Virtual reality compared with in vivo exposure in the treatment of social anxiety disorder: a three-arm randomised controlled trial. Br J Psychiatry. 2017 Apr;210(4):276-283. doi: 10.1192/bjp.bp.116.184234. Epub 2016 Dec 15. PMID: 27979818.

    13. Freeman D, Bradley J, Antley A, Bourke E, DeWeever N, Evans N, Černis E, Sheaves B, Waite F, Dunn G, Slater M, Clark DM. Virtual reality in the treatment of persecutory delusions: randomised controlled experimental study testing how to reduce delusional conviction. Br J Psychiatry. 2016 Jul;209(1):62-7. doi: 10.1192/bjp.bp.115.176438. Epub 2016 May 5. PMID: 27151071; PMCID: PMC4929408.

    14. Lee JS, Namkoong K, Ku J, Cho S, Park JY, Choi YK, Kim JJ, Kim IY, Kim SI, Jung YC. Social pressure-induced craving in patients with alcohol dependence: application of virtual reality to coping skill training. Psychiatry Investig. 2008 Dec;5(4):239-43. doi: 10.4306/pi.2008.5.4.239. Epub 2008 Dec 31. PMID: 20046344; PMCID: PMC2796010.

    15. Anorexia nervosa [Online resource]: Mayo Clinic. URL:https://www.mayoclinic.org/diseases-conditions/anorexia-nervosa/symptoms-causes/syc-20353591.

    16. Anorexia [Online resource]: NHS. URL:https://www.nhs.uk/mental-health/conditions/anorexia/overview/.

    17. Keizer A, van Elburg A, Helms R, Dijkerman HC. A Virtual Reality Full Body Illusion Improves Body Image Disturbance in Anorexia Nervosa. PLoS One. 2016 Oct 6;11(10):e0163921. doi: 10.1371/journal.pone.0163921. PMID: 27711234; PMCID: PMC5053411.

    18. What is Telepsychiatry? [Online resource] American Psychiatric Association. URL:https://www.psychiatry.org/patients-families/telepsychiatry.

    19. Naik SS, Manjunatha N, Kumar CN, Math SB, Moirangthem S. Patient’s Perspectives of Telepsychiatry: The Past, Present and Future. Indian J Psychol Med. 2020 Nov 1;42(5 Suppl):102S-107S. doi: 10.1177/0253717620963341. PMID: 33354053; PMCID: PMC7736740.

    20. Hong JS, Sheriff R, Smith K, Tomlinson A, Saad F, Smith T, Engelthaler T, Phiri P, Henshall C, Ede R, Denis M, Mitter P, D’Agostino A, Cerveri G, Tomassi S, Rathod S, Broughton N, Marlowe K, Geddes J, Cipriani A. Impact of COVID-19 on telepsychiatry at the service and individual patient level across two UK NHS mental health Trusts. Evid Based Ment Health. 2021 Nov;24(4):161-166. doi: 10.1136/ebmental-2021-300287. Epub 2021 Sep 28. PMID: 34583940; PMCID: PMC8483920.

    21. Cowan KE, McKean AJ, Gentry MT, Hilty DM. Barriers to Use of Telepsychiatry: Clinicians as Gatekeepers. Mayo Clin Proc. 2019 Dec;94(12):2510-2523. doi: 10.1016/j.mayocp.2019.04.018. PMID: 31806104.

    22. NHS Mental Health Apps [Online resource]: University Hospitals Sussex NHS Foundation Trust. URL:https://www.bsuh.nhs.uk/library/wp-content/uploads/sites/8/2020/09/NHS-Mental-Health-Apps.pdf.

    23. Kwon S, Firth J, Joshi D, Torous J. Accessibility and availability of smartphone apps for schizophrenia. Schizophrenia (Heidelb). 2022 Nov 16;8(1):98. doi: 10.1038/s41537-022-00313-0. PMID: 36385116; PMCID: PMC9668219.

    24. Rocheteau E. On the role of artificial intelligence in psychiatry. Br J Psychiatry. 2022 Sep 12:1-4. doi: 10.1192/bjp.2022.132. Epub ahead of print. PMID: 36093950.

    25. Liu Y, Pu C, Xia S, Deng D, Wang X, Li M. Machine learning approaches for diagnosing depression using EEG: A review. Transl Neurosci. 2022 Aug 12;13(1):224-235. doi: 10.1515/tnsci-2022-0234. PMID: 36045698; PMCID: PMC9375981.

    26. Wiese W, Friston KJ. AI ethics in computational psychiatry: From the neuroscience of consciousness to the ethics of consciousness. Behav Brain Res. 2022 Feb 26;420:113704. doi: 10.1016/j.bbr.2021.113704. Epub 2021 Dec 4. PMID: 34871706; PMCID: PMC9125160.

    Другие новости