Detection of Suicide Attempters among Suicide Ideators Using Machine Learning

Ryu, Seunghyong; Lee, Hyeongrae; Lee, Dong-Kyun; Kim, Sung‐Wan; Kim, Chul-Eung

doi:10.30773/pi.2019.06.19

Cited by 19 publications

(13 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The number of variables available for splitting at each tree node (mtry) was determined by the best performing mtry option in the training cohort. Due to the imbalanced outcome in the training data, (Cohort 1: 52 Deaths vs. 16 Survivals), a data balancing algorithm, SMOTE (Ryu et al, 2002 ), was applied in the training process to artificially balance the training data.…”

Section: Methodsmentioning

confidence: 99%

Prognostic Value of Transfer Learning Based Features in Resectable Pancreatic Ductal Adenocarcinoma

Zhang

Lobo-Mueller

Karanicolas

et al. 2020

Front. Artif. Intell.

View full text Add to dashboard Cite

Background: Pancreatic Ductal Adenocarcinoma (PDAC) is one of the most aggressive cancers with an extremely poor prognosis. Radiomics has shown prognostic ability in multiple types of cancer including PDAC. However, the prognostic value of traditional radiomics pipelines, which are based on hand-crafted radiomic features alone is limited. Methods: Convolutional neural networks (CNNs) have been shown to outperform radiomics models in computer vision tasks. However, training a CNN from scratch requires a large sample size which is not feasible in most medical imaging studies. As an alternative solution, CNN-based transfer learning models have shown the potential for achieving reasonable performance using small datasets. In this work, we developed and validated a CNN-based transfer learning model for prognostication of overall survival in PDAC patients using two independent resectable PDAC cohorts. Results: The proposed transfer learning-based prognostication model for overall survival achieved the area under the receiver operating characteristic curve of 0.81 on the test cohort, which was significantly higher than that of the traditional radiomics model (0.54). To further assess the prognostic value of the models, the predicted probabilities of death generated from the two models were used as risk scores in a univariate Cox Proportional Hazard model and while the risk score from the traditional radiomics model was not associated with overall survival, the proposed transfer learning-based risk score had significant prognostic value with hazard ratio of 1.86 (95% Confidence Interval: 1.15-3.53, p-value: 0.04). Conclusions: This result suggests that transfer learning-based models may significantly improve prognostic performance in typical small sample size medical imaging studies.

show abstract

Section: Methodsmentioning

confidence: 99%

Prognostic Value of Transfer Learning Based Features in Resectable Pancreatic Ductal Adenocarcinoma

Zhang

Lobo-Mueller

Karanicolas

et al. 2020

Front. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Data collected during the medical visit provided a statistically significant improvement in the prediction of suicidal risk, but with a low effect size. A Korean team [30] sought to identify patients at risk of suicide among those who expressed suicidal ideations in a self-administered questionnaire, by analyzing retrospective data from a national database. This team reported good overall performance, with an accuracy of 88.9% and an AUC of 0.947.…”

Section: Data Usedmentioning

confidence: 99%

Artificial intelligence and suicide prevention: A systematic review

et al. 2022

View full text Add to dashboard Cite

Background and objectives: Suicide is one of the main preventable causes of death. Artificial intelligence could improve methods for assessing suicide risk. The objective of this review is to assess the potential of artificial intelligence in identifying patients who are at risk of attempting suicide.Methods: A systematic review of the literature was conducted on PubMed, EMBASE and SCOPUS databases, using relevant keywords.Results: Thanks to this research, 296 studies were identified. Seventeen studies, published between 2014 and 2020 and matching inclusion criteria, were selected as relevant. Included studies aimed at predicting individual suicide risk or identifying at-risk individuals in a specific population. The AI performance was overall good, although variable across different algorithms and application settings.Conclusion: AI appears to have high potential for identifying patients at risk of suicide. The precise use of these algorithms in clinical situations, as well as the ethical issues it raises, remain to be clarified.

show abstract

“…However, given the low base rate of suicide in the general population, the results based on our data are valuable as they allow us to gain insights into explicit and implicit suicide cognition in an otherwise difficult to obtain group of respondents. Various studies that aimed to build prediction models to detect suicide attempters apply a commonly used method of oversampling suicide attempters 84 .…”

Section: Limitations and Future Directionsmentioning

confidence: 99%

Suicidality and Mood: The Impact of Trends, Seasons, Day of the Week, and Time of Day on Implicit and Explicit Cognitions

Freichel¹,

O’Shea²

2021

Preprint

View full text Add to dashboard Cite

Decades of research have established seasonality effects on completed and attempted suicides, with an increase in rates during spring and early summer. Using more than six years of data (April 2012 – November 2018), we used new Prophet models to forecast mood and explicit and implicit measures of self-harm among an online community sample residing in the US and UK (N > 7,975). We decomposed the time series into trends across the years, within years (seasons), weekly, and daily seasonal patterns. Across all outcomes, the long-term changes across the years and the seasonal patterns show the strongest variation on explicit and implicit cognitions, followed by the time of day (negative cognitions peaking around 4 am – 5 am), with the day of the week showing the weakest effects. The data show a general increase of negative cognitions across the six years, paralleling trends in suicide rates and depression prevalence in the US and UK. Autoregressive-integrated moving average (ARIMA) models showed seasonality effects for mood and desire to die among US, UK, and Canadian respondents (N > 10,445), particularly in the group of respondents who previously made a suicide attempt. Negative cognitions were generally the lowest in summer (June) and peaked in winter (December). These negative cognitions precede the rise in suicidal behaviors during spring and early summer. We discuss potential reasons for lagged effects of negative cognitions on suicidal behavior and implicit cognitions, which may be crucial for theoretical advancements. Our findings have implications for the clinical risk assessment of patients witha history of suicide attempts and public policies regarding the availability of health services.

show abstract

Detection of Suicide Attempters among Suicide Ideators Using Machine Learning

Cited by 19 publications

References 22 publications

Prognostic Value of Transfer Learning Based Features in Resectable Pancreatic Ductal Adenocarcinoma

Prognostic Value of Transfer Learning Based Features in Resectable Pancreatic Ductal Adenocarcinoma

Artificial intelligence and suicide prevention: A systematic review

Suicidality and Mood: The Impact of Trends, Seasons, Day of the Week, and Time of Day on Implicit and Explicit Cognitions

Contact Info

Product

Resources

About