Background: Assessing a patient’s suicide risk is challenging for health professionals because it depends on voluntary disclosure by the patient and often has limited resources. The application of novel machine learning approaches to determine suicide risk has clinical utility. Objective: This study aimed to investigate cross-sectional and longitudinal approaches to assess suicidality based on acoustic voice features of psychiatric patients using artificial intelligence. Methods: We collected 348 voice recordings during clinical interviews of 104 patients diagnosed with mood disorders at baseline and 2, 4, 8, and 12 months after recruitment. Suicidality was assessed using the Beck Scale for Suicidal Ideation and suicidal behavior using the Columbia Suicide Severity Rating Scale. The acoustic features of the voice, including temporal, formal, and spectral features, were extracted from the recordings. A between-person classification model that examines the vocal characteristics of individuals cross sectionally to detect individuals at high risk for suicide and a within-person classification model that detects considerable worsening of suicidality based on changes in acoustic features within an individual were developed and compared. Internal validation was performed using 10-fold cross validation of audio data from baseline to 2-month and external validation was performed using data from 2 to 4 months. Results: A combined set of 12 acoustic features and 3 demographic variables (age, sex, and past suicide attempts) were included in the single-layer artificial neural network for the between-person classification model. Furthermore, 13 acoustic features were included in the extreme gradient boosting machine learning algorithm for the within-person model. The between-person classifier was able to detect high suicidality with 69% accuracy (sensitivity 74%, specificity 62%, area under the receiver operating characteristic curve 0.62), whereas the within-person model was able to predict worsening suicidality over 2 months with 79% accuracy (sensitivity 68%, specificity 84%, area under receiver operating characteristic curve 0.67). The second model showed 62% accuracy in predicting increased suicidality in external sets. Conclusions: Within-person analysis using changes in acoustic features within an individual is a promising approach to detect increased suicidality. Automated analysis of voice can be used to support the real-time assessment of suicide risk in primary care or telemedicine.

Background: Assessing the complex and multifaceted symptoms of patients with acute psychiatric disorders proves to be significantly challenging for clinicians. Moreover, the staff in acute psychiatric wards face high work intensity and risk of burnout, yet research on the introduction of digital technologies in this field remains limited. The combination of continuous and objective wearable sensor data acquired from patients with deep learning techniques holds the potential to overcome the limitations of traditional psychiatric assessments and support clinical decision-making. Objective: This study aimed to develop and validate wearable-based deep learning models to comprehensively predict patient symptoms across various acute psychiatric wards in South Korea. Methods: Participants diagnosed with schizophrenia and mood disorders were recruited from 4 wards across 3 hospitals and prospectively observed using wrist-worn wearable devices during their admission period. Trained raters conducted periodic clinical assessments using the Brief Psychiatric Rating Scale, Hamilton Anxiety Rating Scale, Montgomery-Asberg Depression Rating Scale, and Young Mania Rating Scale. Wearable devices collected patients’ heart rate, accelerometer, and location data. Deep learning models were developed to predict psychiatric symptoms using 2 distinct approaches: single symptoms individually (Single) and multiple symptoms simultaneously via multitask learning (Multi). These models further addressed 2 problems: within-subject relative changes (Deterioration) and between-subject absolute severity (Score). Four configurations were consequently developed for each scale: Single-Deterioration, Single-Score, Multi-Deterioration, and Multi-Score. Data of participants recruited before May 1, 2024, underwent cross-validation, and the resulting fine-tuned models were then externally validated using data from the remaining participants. Results: Of the 244 enrolled participants, 191 (78.3%; 3954 person-days) were included in the final analysis after applying the exclusion criteria. The demographic and clinical characteristics of participants, as well as the distribution of sensor data, showed considerable variations across wards and hospitals. Data of 139 participants were used for cross-validation, while data of 52 participants were used for external validation. The Single-Deterioration and Multi-Deterioration models achieved similar overall accuracy values of 0.75 in cross-validation and 0.73 in external validation. The Single-Score and Multi-Score models attained overall R² values of 0.78 and 0.83 in cross-validation and 0.66 and 0.74 in external validation, respectively, with the Multi-Score model demonstrating superior performance. Conclusions: Deep learning models based on wearable sensor data effectively classified symptom deterioration and predicted symptom severity in participants in acute psychiatric wards. Despite lower computational costs, Multi models demonstrated equivalent or superior performance to Single models, suggesting that multitask learning is a promising approach for comprehensive symptom prediction. However, significant variations were observed across wards, which present a key challenge for developing clinical decision support systems in acute psychiatric wards. Future studies may benefit from recurring local validation or federated learning to address generalizability issues.

Yang JH, et al. J Korean Med Sci. 2024 Jul;39:e226. https://doi.org/10.3346/jkms.2024.39.e226

This study aimed to identify underlying demographic and clinical characteristics among individuals who had previously attempted suicide, utilizing the comprehensive Health Insurance Review and Assessment Service (HIRA) database. Data of patients aged 18 and above who had attempted suicide between January 1 and December 31, 2014, recorded in HIRA, were extracted. The index date was identified when a suicide attempt was made within the year 2014. The medical history of the three years before the index date and seven years of follow-up data after the index date were analyzed. Kaplan-Meier estimate was used to infer reattempt of the suicide attempters, and Cox-proportional hazard analysis was used to investigate risk factors associated with suicide reattempts. A total of 17,026 suicide attempters were identified, of which 1,853 (10.9%) reattempted suicide; 4,925 (28.9%) patients had been diagnosed with depressive disorder. Of the reattempters, 391 (21.1%) demonstrated a history of suicide attempts in the three years before the index date, and the mean number of prior attempts was higher compared to that of the non-reattempters (1.7 vs.1.3, p-value < 0.01). Prior psychiatric medication, polypharmacy, and an increase in the number of psychotropics were associated with suicide reattempt in overall suicide attempters. (Hazard ratio (HR) = 3.20, 95% confidence interval [CI] = 2.56–4.00; HR = 2.42, 95% CI = 1.87–3.14; HR = 19.66, 95% CI = 15.22–25.39 respectively). The risk of reattempt decreased in individuals receiving antidepressant prescriptions compared to those unmedicated, showing a reduction of 78% when prescribed by non-psychiatrists and 89% when prescribed by psychiatrists. Similar risk factors for suicide reattempts were observed in the depressive disorder subgroup, but the median time to reattempt was shorter (556.5 days) for this group compared to that for the overall attempters (578 days). Various risk factors including demographics, clinical characteristics, and medications should be considered to prevent suicide reattempts among suicide attempters, and patients with depressive disorder should be monitored more closely.

Yang JH, et al. J Korean Med Sci. 2021 Jan;36(5):e39. https://doi.org/10.3346/jkms.2021.36.e39