Your browser doesn't support javascript.
loading
Predicting suicidal and self-injurious events in a correctional setting using AI algorithms on unstructured medical notes and structured data.
Lu, Hongxia; Barrett, Alex; Pierce, Albert; Zheng, Jianwei; Wang, Yun; Chiang, Chun; Rakovski, Cyril.
Affiliation
  • Lu H; Schmid College of Science and Technology, Chapman University, 1 University Drive, Orange, CA, USA. Electronic address: holu@chapman.edu.
  • Barrett A; Orange County Health Care Agency, 405 W. 5th Street, Santa Ana, CA, USA. Electronic address: abarrett.chapman@icloud.com.
  • Pierce A; Schmid College of Science and Technology, Chapman University, 1 University Drive, Orange, CA, USA. Electronic address: alpierce@chapman.edu.
  • Zheng J; School of Pharmacy, Chapman University, 9401 Jeronimo Road, Irvine, CA, USA. Electronic address: jzheng@chapman.edu.
  • Wang Y; School of Pharmacy, Chapman University, 9401 Jeronimo Road, Irvine, CA, USA. Electronic address: yunwang@chapman.edu.
  • Chiang C; Orange County Health Care Agency, 405 W. 5th Street, Santa Ana, CA, USA. Electronic address: cchiang@ochca.com.
  • Rakovski C; Schmid College of Science and Technology, Chapman University, 1 University Drive, Orange, CA, USA. Electronic address: rakovski@chapman.edu.
J Psychiatr Res ; 160: 19-27, 2023 04.
Article in En | MEDLINE | ID: mdl-36773344
Suicidal and self-injurious incidents in correctional settings deplete the institutional and healthcare resources, create disorder and stress for staff and other inmates. Traditional statistical analyses provide some guidance, but they can only be applied to structured data that are often difficult to collect and their recommendations are often expensive to act upon. This study aims to extract information from medical and mental health progress notes using AI algorithms to make actionable predictions of suicidal and self-injurious events to improve the efficiency of triage for health care services and prevent suicidal and injurious events from happening at California's Orange County Jails. The results showed that the notes data contain more information with respect to suicidal or injurious behaviors than the structured data available in the EHR database at the Orange County Jails. Using the notes data alone (under-sampled to 50%) in a Transformer Encoder model produced an AUC-ROC of 0.862, a Sensitivity of 0.816, and a Specificity of 0.738. Incorporating the information extracted from the notes data into traditional Machine Learning models as a feature alongside structured data (under-sampled to 50%) yielded better performance in terms of Sensitivity (AUC-ROC: 0.77, Sensitivity: 0.89, Specificity: 0.65). In addition, under-sampling is an effective approach to mitigating the impact of the extremely imbalanced classes.
Subject(s)
Key words

Full text: 1 Collection: 01-internacional Database: MEDLINE Main subject: Prisons / Suicidal Ideation Type of study: Prognostic_studies / Risk_factors_studies Limits: Humans Language: En Journal: J Psychiatr Res Year: 2023 Document type: Article Country of publication:

Full text: 1 Collection: 01-internacional Database: MEDLINE Main subject: Prisons / Suicidal Ideation Type of study: Prognostic_studies / Risk_factors_studies Limits: Humans Language: En Journal: J Psychiatr Res Year: 2023 Document type: Article Country of publication: