1
|
Zhong Y, He J, Luo J, Zhao J, Cen Y, Song Y, Wu Y, Lin C, Pan L, Luo J. A machine learning algorithm-based model for predicting the risk of non-suicidal self-injury among adolescents in western China: A multicentre cross-sectional study. J Affect Disord 2024; 345:369-377. [PMID: 37898476 DOI: 10.1016/j.jad.2023.10.110] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/13/2023] [Revised: 10/06/2023] [Accepted: 10/15/2023] [Indexed: 10/30/2023]
Abstract
The prevalence of non-suicidal self-injurious (NSSI) in adolescents is high. However, few studies exist to predict NSSI in this population. This study employed a machine learning algorithm to develop a predictive model, aiming to more accurately assess the risk of NSSI in Chinese adolescents. Sociodemographic, psychological data were collected in 50 schools in western China. We constructed eXtreme Gradient Boosting (XGBoost) model and multivariate logistic regression model to predict the risk of NSSI and nomograms are plotted. Data from 13,304 adolescents were used for model development, with an average age of 13.00 ± 2.17 years; 617 individuals (4.6 %) reported non-suicidal self-injury (NSSI) behaviors. The results of the XGBoost model showed that depression and anxiety were the top two predictors of NSSI in adolescents. The results of the multivariate logistic regression model showed that the risk factors for adolescent NSSI behaviors include: gender (being female), Age, Living with whom (father), History of psychiatric consultation, Stress, Depression, Anxiety, Tolerance, Emotion abreaction. The XGBoost prediction and multivariate logistic regression model showed good predictive ability. Nomograms can serve as clinical tools to assist in intervention measures, helping adolescents reduce NSSI behaviors and improve their mental and physical well-being.
Collapse
Affiliation(s)
- Yunling Zhong
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Jinlong He
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Jing Luo
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Jiayu Zhao
- Department of Neurology, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Yu Cen
- School of Psychiatry, North Sichuan Medical College, 55 Dongshun Road, Gaoping District, Nanchong City, Sichuan Province, China
| | - Yuqin Song
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Yuhang Wu
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Cen Lin
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Lu Pan
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China
| | - Jiaming Luo
- Mental Health Center, Affiliated Hospital of North Sichuan Medical College, No.1 Maoyuan South Road, Shunqing District, Nanchong City, Sichuan Province, China; School of Psychiatry, North Sichuan Medical College, 55 Dongshun Road, Gaoping District, Nanchong City, Sichuan Province, China.
| |
Collapse
|
2
|
Arora A, Bojko L, Kumar S, Lillington J, Panesar S, Petrungaro B. Assessment of machine learning algorithms in national data to classify the risk of self-harm among young adults in hospital: A retrospective study. Int J Med Inform 2023; 177:105164. [PMID: 37516036 DOI: 10.1016/j.ijmedinf.2023.105164] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/22/2022] [Revised: 07/06/2023] [Accepted: 07/21/2023] [Indexed: 07/31/2023]
Abstract
BACKGROUND Self-harm is one of the most common presentations at accident and emergency departments in the UK and is a strong predictor of suicide risk. The UK Government has prioritised identifying risk factors and developing preventative strategies for self-harm. Machine learning offers a potential method to identify complex patterns with predictive value for the risk of self-harm. METHODS National data in the UK Mental Health Services Data Set were isolated for patients aged 18-30 years who started a mental health hospital admission between Aug 1, 2020 and Aug 1, 2021, and had been discharged by Jan 1, 2022. Data were obtained on age group, gender, ethnicity, employment status, marital status, accommodation status and source of admission to hospital and used to construct seven machine learning models that were used individually and as an ensemble to predict hospital stays that would be associated with a risk of self-harm. OUTCOMES The training dataset included 23 808 items (including 1081 episodes of self-harm) and the testing dataset 5951 items (including 270 episodes of self-harm). The best performing algorithms were the random forest model (AUC-ROC 0.70, 95%CI:0.66-0.74) and the ensemble model (AUC-ROC 0.77 95%CI:0.75-0.79). INTERPRETATION Machine learning algorithms could predict hospital stays with a high risk of self-harm based on readily available data that are routinely collected by health providers and recorded in the Mental Health Services Data Set. The findings should be validated externally with other real-world, prospective data. FUNDING This study was supported by the Midlands and Lancashire Commissioning Support Unit.
Collapse
Affiliation(s)
- Anmol Arora
- School of Clinical Medicine, University of Cambridge, Cambridge, UK; Health Economics Unit, NHS Midlands and Lancashire Commissioning Support Unit, Leyland, UK.
| | - Louis Bojko
- Health Economics Unit, NHS Midlands and Lancashire Commissioning Support Unit, Leyland, UK
| | - Santosh Kumar
- Health Economics Unit, NHS Midlands and Lancashire Commissioning Support Unit, Leyland, UK
| | - Joseph Lillington
- Health Economics Unit, NHS Midlands and Lancashire Commissioning Support Unit, Leyland, UK
| | - Sukhmeet Panesar
- Senior Adviser, Office of Chief Data and Analytics Officer, NHS England and NHS Improvement, UK
| | - Bruno Petrungaro
- Health Economics Unit, NHS Midlands and Lancashire Commissioning Support Unit, Leyland, UK
| |
Collapse
|
3
|
Breitmayer M, Stach M, Kraft R, Allgaier J, Reichert M, Schlee W, Probst T, Langguth B, Pryss R. Predicting the presence of tinnitus using ecological momentary assessments. Sci Rep 2023; 13:8989. [PMID: 37268689 DOI: 10.1038/s41598-023-36172-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/19/2022] [Accepted: 05/24/2023] [Indexed: 06/04/2023] Open
Abstract
Mobile applications have gained popularity in healthcare in recent years. These applications are an increasingly important pillar of public health care, as they open up new possibilities for data collection and can lead to new insights into various diseases and disorders thanks to modern data analysis approaches. In this context, Ecological Momentary Assessment (EMA) is a commonly used research method that aims to assess phenomena with a focus on ecological validity and to help both the user and the researcher observe these phenomena over time. One phenomenon that benefits from this capability is the chronic condition tinnitus. TrackYourTinnitus (TYT) is an EMA-based mobile crowdsensing platform designed to provide more insight into tinnitus by repeatedly assessing various dimensions of tinnitus, including perception (i.e., perceived presence). Because the presence of tinnitus is the dimension that is of great importance to chronic tinnitus patients and changes over time in many tinnitus patients, we seek to predict the presence of tinnitus based on the not directly related dimensions of mood, stress level, arousal, and concentration level that are captured in TYT. In this work, we analyzed a dataset of 45,935 responses to a harmonized EMA questionnaire using different machine learning techniques. In addition, we considered five different subgroups after consultation with clinicians to further validate our results. Finally, we were able to predict the presence of tinnitus with an accuracy of up to 78% and an AUC of up to 85.7%.
Collapse
Affiliation(s)
- Marius Breitmayer
- Institute of Databases and Information Systems, Ulm University, Ulm, Germany.
| | - Michael Stach
- Institute of Clinical Epidemiology and Biometry, University of Würzburg, Würzburg, Germany
| | - Robin Kraft
- Institute of Databases and Information Systems, Ulm University, Ulm, Germany
- Department of Clinical Psychology and Psychotherapy, Ulm University, Ulm, Germany
| | - Johannes Allgaier
- Institute of Clinical Epidemiology and Biometry, University of Würzburg, Würzburg, Germany
| | - Manfred Reichert
- Institute of Databases and Information Systems, Ulm University, Ulm, Germany
| | - Winfried Schlee
- Institute for Information and Process Management, Eastern Switzerland University of Applied Sciences, St. Gallen, Switzerland
- Clinic and Policlinic for Psychiatry and Psychotherapy, University of Regensburg, Regensburg, Germany
| | - Thomas Probst
- Department for Psychotherapy and Biopsychosocial Health, Danube University Krems, Krems, Austria
| | - Berthold Langguth
- Clinic and Policlinic for Psychiatry and Psychotherapy, University of Regensburg, Regensburg, Germany
| | - Rüdiger Pryss
- Institute of Clinical Epidemiology and Biometry, University of Würzburg, Würzburg, Germany
| |
Collapse
|
4
|
Krüger J, Siegert I, Junne F. Künstliche Intelligenz für die Sprachanalyse in der
Psychotherapie – Chancen und Risiken. Psychother Psychosom Med Psychol 2022; 72:395-396. [DOI: 10.1055/a-1915-2589] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
Affiliation(s)
- Julia Krüger
- Universitätsklinik für Psychosomatische Medizin und
Psychotherapie, Medizinische Fakultät,
Otto-von-Guericke-Universität Magdeburg
| | - Ingo Siegert
- Fachgebiet Mobile Dialogsysteme, Institut für Informations- und
Kommunikationstechnik, Fakultät für Elektrotechnik,
Otto-von-Guericke-Universität Magdeburg
| | - Florian Junne
- Universitätsklinik für Psychosomatische Medizin und
Psychotherapie, Medizinische Fakultät,
Otto-von-Guericke-Universität Magdeburg
| |
Collapse
|