Predicting suicidal ideation from irregular and incomplete time series of questionnaires in a smartphone-based suicide prevention platform: a pilot study - Institut Brestois Santé Agro Matière
Article Dans Une Revue Scientific Reports Année : 2024

Predicting suicidal ideation from irregular and incomplete time series of questionnaires in a smartphone-based suicide prevention platform: a pilot study

Résumé

Over 700,000 people die by suicide annually. Collecting longitudinal fine-grained data about at-risk individuals, as they occur in the real world, can enhance our understanding of the temporal dynamics of suicide risk, leading to better identification of those in need of immediate intervention. Selfassessment questionnaires were collected over time from 89 at-risk individuals using the EMMA smartphone application. An artificial intelligence (AI) model was trained to assess current level of suicidal ideation (SI), an early indicator of the suicide risk, and to predict its progression in the following days. A key challenge was the unevenly spaced and incomplete nature of the time series data. To address this, the AI was built on a missing value imputation algorithm. The AI successfully distinguished high SI levels from low SI levels both on the current day (AUC = 0.804, F1 = 0.625, MCC = 0.459) and three days in advance (AUC = 0.769, F1 = 0.576, MCC = 0.386). Besides past SI levels, the most significant questions were related to psychological pain, well-being, agitation, emotional tension, and protective factors such as contacts with relatives and leisure activities. This represents a promising step towards early AI-based suicide risk prediction using a smartphone application.

With over 700,000 suicides annually and 20 times that number in suicide attempts, suicidal behavior remains a significant global health issue 1 , which resists efforts in prevention and treatment. Clinicians are still facing the current impossibility to predict the occurrence of suicidal thoughts and behavior in at risk patients 2 . Recently, smartphone-based solutions have emerged to monitor suicide risk 3,4 , with the aim to detect in real time the potential for suicidal gesture within a short period of time. These technologies are widely available and easily leveraged to collect real-time ecological momentary assessment (EMA) data, which refers to actively asking questions via smartphone. When an imminent risk is detected, the patient can be offered an immediate preventive intervention called Just-in-time adaptive interventions 5 . During the suicidal crisis, JITAIs would deliver an intervention such as a safety planning intervention, which has been largely proven to prevent suicidal behavior 6 .

By providing an accurate depiction of the patient's symptoms 7 , EMA enhances understanding of the temporal dynamics of suicide risk 8 . Indeed, in recent years, a growing number of studies provided relevant new findings about the nature and short-term predictors of suicidal thoughts and behavior using smartphone-based EMA 9 . Recent reviews of papers focusing on intensive longitudinal data and suicidal ideation (SI) revealed that suicidal OPEN.

Fichier principal
Vignette du fichier
2024_Quellec.pdf (2.22 Mo) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-04696762 , version 1 (13-09-2024)

Identifiants

Citer

Gwenolé Quellec, Sofian Berrouiguet, Margot Morgiève, Jonathan Dubois, Marion Leboyer, et al.. Predicting suicidal ideation from irregular and incomplete time series of questionnaires in a smartphone-based suicide prevention platform: a pilot study. Scientific Reports, 2024, 14 (1), pp.20870. ⟨10.1038/s41598-024-71760-1⟩. ⟨hal-04696762⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

More