Artificial Intelligence for Mental Health and Mental Illnesses: an Overview
Purpose of Review
Artificial intelligence (AI) technology holds both great promise to transform mental healthcare and potential pitfalls. This article provides an overview of AI and current applications in healthcare, a review of recent original research on AI specific to mental health, and a discussion of how AI can supplement clinical practice while considering its current limitations, areas needing additional research, and ethical implications regarding AI technology.
We reviewed 28 studies of AI and mental health that used electronic health records (EHRs), mood rating scales, brain imaging data, novel monitoring systems (e.g., smartphone, video), and social media platforms to predict, classify, or subgroup mental health illnesses including depression, schizophrenia or other psychiatric illnesses, and suicide ideation and attempts. Collectively, these studies revealed high accuracies and provided excellent examples of AI’s potential in mental healthcare, but most should be considered early proof-of-concept works demonstrating the potential of using machine learning (ML) algorithms to address mental health questions, and which types of algorithms yield the best performance.
As AI techniques continue to be refined and improved, it will be possible to help mental health practitioners re-define mental illnesses more objectively than currently done in the DSM-5, identify these illnesses at an earlier or prodromal stage when interventions may be more effective, and personalize treatments based on an individual’s unique characteristics. However, caution is necessary in order to avoid over-interpreting preliminary results, and more work is required to bridge the gap between AI in mental health research and clinical care.
KeywordsTechnology Machine learning Natural language processing Deep learning Schizophrenia Depression Suicide Bioethics Research ethics
This study was supported, in part, by the National Institute of Mental Health T32 Geriatric Mental Health Program (grant MH019934 to DVJ [PI]), the IBM Research AI through the AI Horizons Network IBM-UCSD AI for Healthy Living (AIHL) Center, by the Stein Institute for Research on Aging at the University of California San Diego, and by the National Institutes of Health, Grant UL1TR001442 of CTSA funding. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Compliance with Ethical Standards
Conflict of Interest
Sarah Graham, Xin Tu, and Ho-Cheol Kim each declare no potential conflicts of interest.
Colin Depp and Dilip V. Jeste are Co-Directors of UCSD-IBM Center on Artificial Intelligence for Healthy Living (2018–2022). This is a grant to UCSD from IBM. Drs. Depp and Jeste have no commercial interest in IBM or any other AI-related companies.
Ellen E. Lee has received grants from The National Institute of Mental Health, The National Institutes of Health, and The Stein Institute for Research on Aging.
Camille Nebeker is a co-investigator on a grant supported by IBM and her research on the ethics of emerging technologies is supported by the Robert Wood Johnson Foundation.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
- 2.Schwab K. The fourth Industrial Revolution. First. New York, NY: Currency; 2017. p. 192.Google Scholar
- 4.Metz C, Smith CS. “A.I. can be a boon to medicine that could easily go rogue’. The New York Times. 2019 Mar 25;B5.Google Scholar
- 6.John McCarthy. Artificial intelligence, logic and formalizing common sense. In Philosophical logic and artificial intelligence 1989 (pp. 161-190). Springer, Dordrecht.Google Scholar
- 7.Turing AM. Computing machinery and intelligence. Comput Mach Intell. 1950;49:433–60 Available from: https://linkinghub.elsevier.com/retrieve/pii/B978012386980750023X.Google Scholar
- 20.Topol E. Deep medicine: how artificial intelligence can make healthcare human again. 1st ed. New York, NY: Basic Books; 2019.Google Scholar
- 26.Mohr D, Zhang M, Schueller SM. Personal sensing: understanding mental health using ubiquitous sensors and machine learning. Annu Rev Clin Psychol. 2017;13:23–47. https://doi.org/10.1146/annurev-clinpsy-032816-044949.CrossRefPubMedGoogle Scholar
- 29.• Bzdok D, Meyer-Lindenberg A. Machine learning for precision psychiatry: opportunities and challenges. Biol Psychiatry Cogn Neurosci Neuroimaging. 2018;3(3):223–30. https://doi.org/10.1016/j.bpsc.2017.11.007 This review aquaints the reader with key terms related to artificial intelligence and psychiatry and gives an overview of the opportunities and challenges in bringing machine intelligence into psychiatric practice.CrossRefPubMedGoogle Scholar
- 30.Jeste DV, Glorioso D, Lee EE, Daly R, Graham S, Liu J, et al. Study of independent living residents of a continuing care senior housing community: sociodemographic and clinical associations of cognitive, physical, and mental health. Am J Geriatr Psychiatry [Internet]. 2019. https://doi.org/10.1016/j.jagp.2019.04.002.CrossRefGoogle Scholar
- 43.Kotsiantis SB, Zaharakis I, Pintelas P. Supervised machine learning: a review of classification techniques. Emerg Artif Intell Appl Comput Eng. 2007;160:3–24.Google Scholar
- 44.Dy JG, Brodley CE. Feature selection for unsupervised learning. J Mach Learn Res. 2004;5:845–89 Retrieved from: http://www.jmlr.org/papers/volume5/dy04a/dy04a.pdf.Google Scholar
- 49.Samek W, Wiegand T, Müller K-R. Explainable artificial intelligence: understanding, visualizing and interpreting deep learning models. arXiv Prepr arXiv. 2017;1708.08296. Available from: http://arxiv.org/abs/1708.08296
- 56.Sargent DJ. Comparison of artificial neural networks with other statistical approaches. Cancer. 2002;91(S8):1636–42. https://doi.org/10.1002/1097-0142(20010415)91:8+<1636::AID-CNCR1176>3.0.CO;2-D.CrossRefGoogle Scholar
- 60.Fernandes AC, Dutta R, Velupillai S, Sanyal J, Stewart R, Chandran D. Identifying suicide ideation and suicidal attempts in a psychiatric clinical research database using natural language processing. Sci Rep. 2018;8(1):7426. https://doi.org/10.1038/s41598-018-25773-2.CrossRefPubMedPubMedCentralGoogle Scholar
- 61.Jackson RG, Patel R, Jayatilleke N, Kolliakou A, Ball M, Gorrell G, et al. Natural language processing to extract symptoms of severe mental illness from clinical text: the Clinical Record Interactive Search Comprehensive Data Extraction (CRIS-CODE) project. BMJ Open. 2017;7(1):e012012. https://doi.org/10.1136/bmjopen-2016-012012.CrossRefPubMedPubMedCentralGoogle Scholar
- 62.• Kessler RC, Hwang I, Hoffmire CA, Mccarthy JF, Maria V, Rosellini AJ, et al. Developing a practical suicide risk prediction model for targeting high-risk patients in the Veterans health Administration. Int J Methods Psychiatr Res. 2017;26(3):1–14. https://doi.org/10.1002/mpr.1575 This study from the US Veterans Health Administration (VHA) compared machine learning approaches within and out of sample with traditional statistics to identify veterans at high suicide risk for more targeted care.CrossRefGoogle Scholar
- 64.• Chekroud AM, Zotti RJ, Shehzad Z, Gueorguieva R, Johnson MK, Trivedi MH, et al. Cross-trial prediction of treatment outcome in depression: a machine learning approach. Lancet Psychiatry. 2016;3(3):243–50. https://doi.org/10.1016/S2215-0366(15)00471-X This study used machine learning to identify 25 variables from the STAR*D clinical trial that were most predictive of treatment outcome following a 12-week course of the antidepressant citalopram and externally validated their models in an indepdent sample from the CO-MED clinical trial undergoing escitalopram treatment.CrossRefGoogle Scholar
- 65.• Chekroud AM, Gueorguieva R, Krumholz HM, Trivedi MH, Krystal JH, McCarthy G. Reevaluating the efficacy and predictability of antidepressant treatments a symptom clustering approach. JAMA Psychiatry. 2017;74(4):370–8. https://doi.org/10.1001/jamapsychiatry.2017.0025 This study demonstrated that clusters of symptoms are detectable in 2 common depression rating scales (QIDS-SR and HAM-D), and these symptom clusters vary in their responsiveness to different antidepressant treatments.CrossRefPubMedPubMedCentralGoogle Scholar
- 67.• Drysdale AT, Grosenick L, Downar J, Dunlop K, Mansouri F, Meng Y, et al. Resting-state connectivity biomarkers define neurophysiological subtypes of depression. Nat Med. 1878;23(1):28–38. DOI: https://doi.org/10.1038/nm.4246. This study used unsupervised and supervised machine learning with fMRI data and demonstrated that patients with depression can be subdivided into four neurophysiological subtypes defined by distinct patterns of dysfunctional connectivity in limbic and frontostriatal networks and further that these subtypes predicted which patients responded to repetitive transcranial magnetic stimulation (TMS) therapy.CrossRefGoogle Scholar
- 68.Kalmady SV, Greiner R, Agrawal R, Shivakumar V, Narayanaswamy JC, Brown MRG, et al. Towards artificial intelligence in mental health by improving schizophrenia prediction with multiple brain parcellation ensemble-learning. NPJ Schizophr. 2019;5(1):2. https://doi.org/10.1038/s41537-018-0070-8.CrossRefPubMedPubMedCentralGoogle Scholar
- 69.• Dwyer DB, Cabral C, Kambeitz-Ilankovic L, Sanfelici R, Kambeitz J, Calhoun V, et al. Brain subtyping enhances the neuroanatomical discrimination of schizophrenia. Schizophr Bull. 2018;44(5):1060–9. https://doi.org/10.1093/schbul/sby008 This study used both unsupervised and supervised machine learning with structural MRI data and suggested that sMRI-based subtyping enhances neuroanatomical discrimination of schizophrenia by identifying generalizable brain patterns that align with a clinical staging model of the disorder.CrossRefPubMedPubMedCentralGoogle Scholar
- 71.Patel MJ, Andreescu C, Price JC, Edelman KL, Reynolds CF, Aizenstein HJ. Machine learning approaches for integrating clinical and imaging features in late-life depression classification and response prediction. Int J Geriatr Psychiatry. 2015;30(10):1056–67. https://doi.org/10.1002/gps.4262.CrossRefPubMedPubMedCentralGoogle Scholar
- 74.Bain EE, Shafner L, Walling DP, Othman AA, Chuang-Stein C, Hinkle J, et al. Use of a novel artificial intelligence platform on mobile devices to assess dosing compliance in a phase 2 clinical trial in subjects with schizophrenia. JMIR mHealth uHealth. 2017;5(2):e18. https://doi.org/10.2196/mhealth.7030.CrossRefPubMedPubMedCentralGoogle Scholar
- 78.Cook BL, Progovac AM, Chen P, Mullin B, Hou S, Baca-Garcia E. Novel use of natural language processing (NLP) to predict suicidal ideation and psychiatric symptoms in a text-based mental health intervention in Madrid. Comput Math Methods Med. 2016;2016:1–8. https://doi.org/10.1155/2016/8708434.CrossRefGoogle Scholar
- 79.Aldarwish MM, Ahmad HF. Predicting depression levels using social media posts. Proc - 2017 IEEE 13th Int Symp Auton Decentralized Syst ISADS 2017. 2017;277–80. DOI: https://doi.org/10.1109/ISADS.2017.41.
- 81.Landeiro Dos Reis V, Culotta A. Using matched samples to estimate the effects of exercise on mental health from twitter. Proc Twenty-Ninth AAAI Conf Artif Intell. 2015:182–8 Retrieved from: https://www.aaai.org/ocs/index.php/AAAI/AAAI15/paper/viewPaper/9960.
- 83.Mowery D, Park A, Conway M, Bryan C. Towards automatically classifying depressive symptoms from twitter data for population health. Proc Work Comput Model People’s Opin Personal Emot Soc Media. 2016:182–91 Available from: https://www.aclweb.org/anthology/W16-4320.
- 93.Lee EE, Depp C, Palmer BW, Glorioso D, Daly R, Liu J, et al. High prevalence and adverse health effects of loneliness in community-dwelling adults across the lifespan: role of wisdom as a protective factor. Int Psychogeriatr. 2018;(May):1–16. https://doi.org/10.1017/S1041610218002120.CrossRefGoogle Scholar
- 95.Lemaitre G, Nogueira F, Aridas CK. Imbalanced-learn: a Python toolbox to tackle the curse of imbalanced datasets in machine learning. J Mach Learn Res. 2017;18(1):559–63 Available from: http://www.jmlr.org/papers/volume18/16-365/16-365.pdf.Google Scholar
- 96.World Health Organization. Frequently asked questions. 2019. Available from: https://www.who.int/about/who-we-are/frequently-asked-questions
- 97.American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Publication; 2013.Google Scholar
- 99.Torrey L, Shavlik J. Transfer learning. In: Handbook of research on machine learning applications and trends: algorithms, methods, and techniques. IGI Global. 2009:242–64.Google Scholar
- 101.Razzak MI, Naz S, Zaib A. Deep learning for medical image processing: overview, challenges and future. In: Classification in BioApps. Springer Cham.; p. 323–50.Google Scholar
- 102.Kemker R, McClure M, Abitino A, Hayes T, Kanan C. Measuring catastrophic forgetting in neural networks. Thirty-second AAAI Conf Artif Intell. 2018:3390–8 Available from: http://arxiv.org/abs/1708.02072.
- 106.Nebeker C, Harlow J, Giacinto RE, Orozco- r, Bloss CS, Weibel N, et al. Ethical and regulatory challenges of research using pervasive sensing and other emerging technologies: IRB perspectives. AJOB Empir Bioeth 2017;8(4):266–276. DOI: https://doi.org/10.1080/23294515.2017.1403980.CrossRefGoogle Scholar
- 107.Sears M. AI Bias and the “people factor” in AI development. 2018 [cited 2019 Feb 26]. Available from: https://www.forbes.com/sites/colehaan/2019/04/30/from-the-bedroom-to-the-boardroom-how-a-sleepwear-company-is-empowering-women/#7717796a2df3
- 109.Huang H, Cao B, Yu PS, Wang C-D, Leow AD. dpMood: exploiting local and periodic typing dynamics for personalized mood prediction. 2018 IEEE Conf Data Min. 2018:157–66. https://doi.org/10.1109/ICDM.2018.00031.
- 111.De Choudhury M, Kiciman E. Integrating artificial and human intelligence in complex, sensitive problem domains: experiences from mental health. AI Mag. 2018;39(3):69–80 Retrieved from: http://kiciman.org/wp-content/uploads/2018/10/AIMag_IntegratingAIandHumanIntelligence_Fall2018.pdf.CrossRefGoogle Scholar