Development of an early-warning system for high-risk patients for suicide attempt using deep learning and electronic health records

Using augmented intelligence to improve long term outcomes

PURPOSE OF REVIEW

For augmented intelligence (AI) tools to realize their potential, critical care clinicians must ensure they are designed to improve long-term outcomes. This overview is intended to align professionals with the state-of-the art of AI.

RECENT FINDINGS

Many AI tools are undergoing preliminary assessment of their ability to support the care of survivors and their caregivers at multiple time points after intensive care unit (ICU) discharge. The domains being studied include early identification of deterioration (physiological, mental), management of impaired physical functioning, pain, sleep and sexual dysfunction, improving nutrition and communication, and screening and treatment of cognitive impairment and mental health disorders.Several technologies are already being marketed and many more are in various stages of development. These technologies mostly still require clinical trials outcome testing. However, lacking a formal regulatory approval process, some are already in use.

SUMMARY

Plans for long-term management of ICU survivors must account for the development of a holistic follow-up system that incorporates AI across multiple platforms. A tiered post-ICU screening program may be established wherein AI tools managed by ICU follow-up clinics provide appropriate assistance without human intervention in cases with less pathology and refer severe cases to expert treatment.

Social determinants of health and sepsis: a case-control study

PURPOSE

We aimed to identify whether social determinants of health (SDoH) are associated with the development of sepsis and assess the differences between individuals living within systematically disadvantaged neighbourhoods compared with those living outside these neighbourhoods.

METHODS

We conducted a single-centre case-control study including 300 randomly selected adult patients (100 patients with sepsis and 200 patients without sepsis) admitted to the emergency department of a large academic tertiary care hospital in Hamilton, ON, Canada. We collected data on demographics and a limited set of SDoH variables, including neighbourhood household income, smoking history, social support, and history of alcohol disorder. We analyzed study data using multivariate logistic regression models.

RESULTS

The study included 100 patients with sepsis with a median [interquartile range (IQR)] age of 75 [58-84] yr and 200 patients without sepsis with a median [IQR] age of 72 [60-83] yr. Factors significantly associated with sepsis included arrival by ambulance, absence of a family physician, higher Hamilton Early Warning Score, and a recorded history of dyslipidemia. Important SDoH variables, such as individual or household income and race, were not available in the medical chart. In patients with SDoH available in their medical records, no SDoH was significantly associated with sepsis. Nevertheless, compared with their proportion of the Hamilton population, the rate of sepsis cases and sepsis deaths was approximately two times higher among patients living in systematically disadvantaged neighbourhoods.

CONCLUSIONS

This study revealed the lack of available SDoH data in electronic health records. Despite no association between the SDoH variables available and sepsis, we found a higher rate of sepsis cases and sepsis deaths among individuals living in systematically disadvantaged neighbourhoods. Including SDoH in electronic health records is crucial to study their effect on the risk of sepsis and to provide equitable care.

Validation of a Multivariable Model to Predict Suicide Attempt in a Mental Health Intake Sample

Importance

Given that suicide rates have been increasing over the past decade and the demand for mental health care is at an all-time high, targeted prevention efforts are needed to identify individuals seeking to initiate mental health outpatient services who are at high risk for suicide. Suicide prediction models have been developed using outpatient mental health encounters, but their performance among intake appointments has not been directly examined.

Objective

To assess the performance of a predictive model of suicide attempts among individuals seeking to initiate an episode of outpatient mental health care.

Design, Setting, and Participants

This prognostic study tested the performance of a previously developed machine learning model designed to predict suicide attempts within 90 days of any mental health outpatient visit. All mental health intake appointments scheduled between January 1, 2012, and April 1, 2022, at Kaiser Permanente Northern California, a large integrated health care delivery system serving over 4.5 million patients, were included. Data were extracted and analyzed from August 9, 2022, to July 31, 2023.

Main Outcome and Measures

Suicide attempts (including completed suicides) within 90 days of the appointment, determined by diagnostic codes and government databases. All predictors were extracted from electronic health records.

Results

The study included 1 623 232 scheduled appointments from 835 616 unique patients. There were 2800 scheduled appointments (0.17%) followed by a suicide attempt within 90 days. The mean (SD) age across appointments was 39.7 (15.8) years, and most appointments were for women (1 103 184 [68.0%]). The model had an area under the receiver operating characteristic curve of 0.77 (95% CI, 0.76-0.78), an area under the precision-recall curve of 0.02 (95% CI, 0.02-0.02), an expected calibration error of 0.0012 (95% CI, 0.0011-0.0013), and sensitivities of 37.2% (95% CI, 35.5%-38.9%) and 18.8% (95% CI, 17.3%-20.2%) at specificities of 95% and 99%, respectively. The 10% of appointments at the highest risk level accounted for 48.8% (95% CI, 47.0%-50.6%) of the appointments followed by a suicide attempt.

Conclusions and Relevance

In this prognostic study involving mental health intakes, a previously developed machine learning model of suicide attempts showed good overall classification performance. Implementation research is needed to determine appropriate thresholds and interventions for applying the model in an intake setting to target high-risk cases in a manner that is acceptable to patients and clinicians.

Role of machine learning algorithms in suicide risk prediction: a systematic review-meta analysis of clinical studies

OBJECTIVE

Suicide is a complex and multifactorial public health problem. Understanding and addressing the various factors associated with suicide is crucial for prevention and intervention efforts. Machine learning (ML) could enhance the prediction of suicide attempts.

METHOD

A systematic review was performed using PubMed, Scopus, Web of Science and SID databases. We aim to evaluate the performance of ML algorithms and summarize their effects, gather relevant and reliable information to synthesize existing evidence, identify knowledge gaps, and provide a comprehensive list of the suicide risk factors using mixed method approach.

RESULTS

Forty-one studies published between 2011 and 2022, which matched inclusion criteria, were chosen as suitable. We included studies aimed at predicting the suicide risk by machine learning algorithms except natural language processing (NLP) and image processing. The neural network (NN) algorithm exhibited the lowest accuracy at 0.70, whereas the random forest demonstrated the highest accuracy, reaching 0.94. The study assessed the COX and random forest models and observed a minimum area under the curve (AUC) value of 0.54. In contrast, the XGBoost classifier yielded the highest AUC value, reaching 0.97. These specific AUC values emphasize the algorithm-specific performance in capturing the trade-off between sensitivity and specificity for suicide risk prediction. Furthermore, our investigation identified several common suicide risk factors, including age, gender, substance abuse, depression, anxiety, alcohol consumption, marital status, income, education, and occupation. This comprehensive analysis contributes valuable insights into the multifaceted nature of suicide risk, providing a foundation for targeted preventive strategies and intervention efforts.

CONCLUSIONS

The effectiveness of ML algorithms and their application in predicting suicide risk has been controversial. There is a need for more studies on these algorithms in clinical settings, and the related ethical concerns require further clarification.

Graph Attention-Based Curriculum Learning for Mental Healthcare Classification

Current research has examined the use of user-generated data from online media to identify and diagnose depression as a serious mental health issue that can significantly impact an individual's daily life. To this end, many studies examined words in personal statements to identify depression. In addition to aiding in the diagnosis and treatment of depression, this study uses and utilizes a Graph Attention Network (GAT) model for the classification of depression from online media. The model is based on masked self-attention layers, that assigns different weight to each node in a neighborhood without costly matrix operations. In addition, an emotion lexicon was extended using hypernyms to improve the model performance. Furthermore, embedding of the model was used to illustrate the contribution of the activated words to each symptom and to obtain qualitative agreement from psychiatrists. This technique uses previously learned embedding to illustrate the contribution of activated words to depressive symptoms in online forums. A significant improvement was observed in the model's performance through the use of the lexicon extension method, resulting in an increase in the ROC performance. The performance was also enhanced by an increase in vocabulary and the adoption of a graph-based curriculum. The lexicon expansion method involves the generation of additional words with similar semantic attributes, utilizing similarity metrics to reinforce lexical features. The graph-based curriculum learning also utilized to handle more challenging training samples, allowing the model to develop increasing expertise in learning complex correlations between input data and output labels.

The use of machine learning on administrative and survey data to predict suicidal thoughts and behaviors: a systematic review

Background

Machine learning is a promising tool in the area of suicide prevention due to its ability to combine the effects of multiple risk factors and complex interactions. The power of machine learning has led to an influx of studies on suicide prediction, as well as a few recent reviews. Our study distinguished between data sources and reported the most important predictors of suicide outcomes identified in the literature.

Objective

Our study aimed to identify studies that applied machine learning techniques to administrative and survey data, summarize performance metrics reported in those studies, and enumerate the important risk factors of suicidal thoughts and behaviors identified.

Methods

A systematic literature search of PubMed, Medline, Embase, PsycINFO, Web of Science, Cumulative Index to Nursing and Allied Health Literature (CINAHL), and Allied and Complementary Medicine Database (AMED) to identify all studies that have used machine learning to predict suicidal thoughts and behaviors using administrative and survey data was performed. The search was conducted for articles published between January 1, 2019 and May 11, 2022. In addition, all articles identified in three recently published systematic reviews (the last of which included studies up until January 1, 2019) were retained if they met our inclusion criteria. The predictive power of machine learning methods in predicting suicidal thoughts and behaviors was explored using box plots to summarize the distribution of the area under the receiver operating characteristic curve (AUC) values by machine learning method and suicide outcome (i.e., suicidal thoughts, suicide attempt, and death by suicide). Mean AUCs with 95% confidence intervals (CIs) were computed for each suicide outcome by study design, data source, total sample size, sample size of cases, and machine learning methods employed. The most important risk factors were listed.

Results

The search strategy identified 2,200 unique records, of which 104 articles met the inclusion criteria. Machine learning algorithms achieved good prediction of suicidal thoughts and behaviors (i.e., an AUC between 0.80 and 0.89); however, their predictive power appears to differ across suicide outcomes. The boosting algorithms achieved good prediction of suicidal thoughts, death by suicide, and all suicide outcomes combined, while neural network algorithms achieved good prediction of suicide attempts. The risk factors for suicidal thoughts and behaviors differed depending on the data source and the population under study.

Conclusion

The predictive utility of machine learning for suicidal thoughts and behaviors largely depends on the approach used. The findings of the current review should prove helpful in preparing future machine learning models using administrative and survey data.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022333454 identifier CRD42022333454.

A comparative study of cognitive function in young patients with bipolar disorder with and without non-suicidal self-injury

OBJECTIVE

Bipolar disorder (BD) is a chronic mental disorder characterized by alternating or mixed episodes of mania or hypomania and depression. Cognitive function impairment is a frequent associated feature of the disease. While many BD patients also engage in non-suicidal self-injury (NSSI), there is a lack of studies on the cognitive function of BD patients with NSSI. This study aimed to evaluate cognitive functioning of BD patients with NSSI and provide a clinical basis for the differential diagnosis and treatment of BD and NSSI.

METHODS

A total of 60 BD patients with NSSI, 60 BD patients without NSSI, and 60 healthy controls (HC) were selected for the study. All participants met the inclusion criteria and were not taking any medications, excluding the potential effects of medication on cognitive functions. The following neurocognitive tests were used to measure the cognitive functions in areas such as speed of processing, reasoning and problem solving, attention/vigilance, working memory, visual learning, and verbal learning: The Trail Making Test (TMT), Category Fluency, Digit Symbol Coding Test (DSCT), Brief Visuospatial Memory Test-Revised (BVMT-R), The Neuropsychological Assessment Battery Mazes (NABM), Wechsler Memory Scale Third Edition Spatial Span Test (WMS III-SST), Hopkins Verbal Learning Test-Revised (HVLTR) and Continuous Performance Test and Identical Prs (CPT-IP).

RESULTS

The findings indicated that BD patients with NSSI exhibited cognitive impairment in all measured cognitive domains. On the other hand, BD patients without NSSI showed less pronounced impairment in terms of speed of processing, but exhibited significant cognitive impairment in the remaining five areas compared to the HC group.

CONCLUSION

The study underscores the presence of cognitive impairment in BD, and the cognitive impairment is more severe in BD patients with NSSI compared to those without NSSI. In conclusion, both individuals with NSSI and those without NSSI in BD exhibit cognitive impairment, which provides ideas and strategies for using cognitive-behavioral therapy to treat BD and NSSI.

Deep sequential neural network models improve stratification of suicide attempt risk among US veterans

OBJECTIVE

To apply deep neural networks (DNNs) to longitudinal EHR data in order to predict suicide attempt risk among veterans. Local explainability techniques were used to provide explanations for each prediction with the goal of ultimately improving outreach and intervention efforts.

MATERIALS AND METHODS

The DNNs fused demographic information with diagnostic, prescription, and procedure codes. Models were trained and tested on EHR data of approximately 500 000 US veterans: all veterans with recorded suicide attempts from April 1, 2005, through January 1, 2016, each paired with 5 veterans of the same age who did not attempt suicide. Shapley Additive Explanation (SHAP) values were calculated to provide explanations of DNN predictions.

RESULTS

The DNNs outperformed logistic and linear regression models in predicting suicide attempts. After adjusting for the sampling technique, the convolutional neural network (CNN) model achieved a positive predictive value (PPV) of 0.54 for suicide attempts within 12 months by veterans in the top 0.1% risk tier. Explainability methods identified meaningful subgroups of high-risk veterans as well as key determinants of suicide attempt risk at both the group and individual level.

DISCUSSION AND CONCLUSION

The deep learning methods employed in the present study have the potential to significantly enhance existing suicide risk models for veterans. These methods can also provide important clues to explore the relative value of long-term and short-term intervention strategies. Furthermore, the explainability methods utilized here could also be used to communicate to clinicians the key features which increase specific veterans' risk for attempting suicide.

TransformEHR: transformer-based encoder-decoder generative model to enhance prediction of disease outcomes using electronic health records

Deep learning transformer-based models using longitudinal electronic health records (EHRs) have shown a great success in prediction of clinical diseases or outcomes. Pretraining on a large dataset can help such models map the input space better and boost their performance on relevant tasks through finetuning with limited data. In this study, we present TransformEHR, a generative encoder-decoder model with transformer that is pretrained using a new pretraining objective-predicting all diseases and outcomes of a patient at a future visit from previous visits. TransformEHR's encoder-decoder framework, paired with the novel pretraining objective, helps it achieve the new state-of-the-art performance on multiple clinical prediction tasks. Comparing with the previous model, TransformEHR improves area under the precision-recall curve by 2% (p < 0.001) for pancreatic cancer onset and by 24% (p = 0.007) for intentional self-harm in patients with post-traumatic stress disorder. The high performance in predicting intentional self-harm shows the potential of TransformEHR in building effective clinical intervention systems. TransformEHR is also generalizable and can be easily finetuned for clinical prediction tasks with limited data.

Ethical considerations for the use of artificial intelligence in medical decision-making capacity assessments

The rapid advancement of artificial intelligence (AI) and machine learning are providing new tools to clinicians. AI tools have the potential to process vast amounts of data in a short amount of time, providing new insights and changing how we approach complicated healthcare problems. AI has the potential to assist clinicians in medical decision-making capacity assessments by providing additional insights to an evaluation process that currently lacks universal objective standards. However, despite the promise of AI in this setting, there remain significant concerns making it unlikely to replace human evaluators anytime soon. AI remains highly susceptible to biased inputs and thus biased decisions, raises questions about autonomy, and creates uncertainty for who is accountable for the ultimate decision of capacity. In this paper we explore these ethical considerations of using AI for capacity assessments. While we acknowledge AI may not be ready to replace physicians in determining patient medical-decision making capacity, these new technologies have significant near-term potential as a tool to screen patients, uncover physician biases, and guide next steps after a capacity determination has been made.

Improving Diagnostics with Deep Forest Applied to Electronic Health Records

An electronic health record (EHR) is a vital high-dimensional part of medical concepts. Discovering implicit correlations in the information of this data set and the research and informative aspects can improve the treatment and management process. The challenge of concern is the data sources' limitations in finding a stable model to relate medical concepts and use these existing connections. This paper presents Patient Forest, a novel end-to-end approach for learning patient representations from tree-structured data for readmission and mortality prediction tasks. By leveraging statistical features, the proposed model is able to provide an accurate and reliable classifier for predicting readmission and mortality. Experiments on MIMIC-III and eICU datasets demonstrate Patient Forest outperforms existing machine learning models, especially when the training data are limited. Additionally, a qualitative evaluation of Patient Forest is conducted by visualising the learnt representations in 2D space using the t-SNE, which further confirms the effectiveness of the proposed model in learning EHR representations.

An efficient landmark model for prediction of suicide attempts in multiple clinical settings

Growing evidence has shown that applying machine learning models to large clinical data sources may exceed clinician performance in suicide risk stratification. However, many existing prediction models either suffer from "temporal bias" (a bias that stems from using case-control sampling) or require training on all available patient visit data. Here, we adopt a "landmark model" framework that aligns with clinical practice for prediction of suicide-related behaviors (SRBs) using a large electronic health record database. Using the landmark approach, we developed models for SRB prediction (regularized Cox regression and random survival forest) that establish a time-point (e.g., clinical visit) from which predictions are made over user-specified prediction windows using historical information up to that point. We applied this approach to cohorts from three clinical settings: general outpatient, psychiatric emergency department, and psychiatric inpatients, for varying prediction windows and lengths of historical data. Models achieved high discriminative performance (area under the Receiver Operating Characteristic curve 0.74-0.93 for the Cox model) across different prediction windows and settings, even with relatively short periods of historical data. In short, we developed accurate, dynamic SRB risk prediction models with the landmark approach that reduce bias and enhance the reliability and portability of suicide risk prediction models.

Complex modeling with detailed temporal predictors does not improve health records-based suicide risk prediction

Suicide risk prediction models can identify individuals for targeted intervention. Discussions of transparency, explainability, and transportability in machine learning presume complex prediction models with many variables outperform simpler models. We compared random forest, artificial neural network, and ensemble models with 1500 temporally defined predictors to logistic regression models. Data from 25,800,888 mental health visits made by 3,081,420 individuals in 7 health systems were used to train and evaluate suicidal behavior prediction models. Model performance was compared across several measures. All models performed well (area under the receiver operating curve [AUC]: 0.794-0.858). Ensemble models performed best, but improvements over a regression model with 100 predictors were minimal (AUC improvements: 0.006-0.020). Results are consistent across performance metrics and subgroups defined by race, ethnicity, and sex. Our results suggest simpler parametric models, which are easier to implement as part of routine clinical practice, perform comparably to more complex machine learning methods.

Identifying populations at ultra-high risk of suicide using a novel machine learning method

BACKGROUND

Targeted interventions for suicide prevention rely on adequate identification of groups at elevated risk. Several risk factors for suicide are known, but little is known about the interactions between risk factors. Interactions between risk factors may aid in detecting more specific sub-populations at higher risk.

METHODS

Here, we use a novel machine learning heuristic to detect sub-populations at ultra high-risk for suicide based on interacting risk factors. The data-driven and hypothesis-free model is applied to investigate data covering the entire population of the Netherlands.

FINDINGS

We found three sub-populations with extremely high suicide rates (i.e. >50 suicides per 100,000 person years, compared to 12/100,000 in the general population), namely: (1) people on unfit for work benefits that were never married, (2) males on unfit for work benefits, and (3) those aged 55-69 who live alone, were never married and have a relatively low household income. Additionally, we found two sub-populations where the rate was higher than expected based on individual risk factors alone: widowed males, and people aged 25-39 with a low level of education.

INTERPRETATION

Our model is effective at finding ultra-high risk groups which can be targeted using sub-population level interventions. Additionally, it is effective at identifying high-risk groups that would not be considered risk groups based on conventional risk factor analysis.

Single center blind testing of a US multi-center validated diagnostic algorithm for Kawasaki disease in Taiwan

Background

Kawasaki disease (KD) is the leading cause of acquired heart disease in children. The major challenge in KD diagnosis is that it shares clinical signs with other childhood febrile control (FC) subjects. We sought to determine if our algorithmic approach applied to a Taiwan cohort.

Methods

A single center (Chang Gung Memorial Hospital in Taiwan) cohort of patients suspected with acute KD were prospectively enrolled by local KD specialists for KD analysis. Our previously single-center developed computer-based two-step algorithm was further tested by a five-center validation in US. This first blinded multi-center trial validated our approach, with sufficient sensitivity and positive predictive value, to identify most patients with KD diagnosed at centers across the US. This study involved 418 KDs and 259 FCs from the Chang Gung Memorial Hospital in Taiwan.

Findings

Our diagnostic algorithm retained sensitivity (379 of 418; 90.7%), specificity (223 of 259; 86.1%), PPV (379 of 409; 92.7%), and NPV (223 of 247; 90.3%) comparable to previous US 2016 single center and US 2020 fiver center results. Only 4.7% (15 of 418) of KD and 2.3% (6 of 259) of FC patients were identified as indeterminate. The algorithm identified 18 of 50 (36%) KD patients who presented 2 or 3 principal criteria. Of 418 KD patients, 157 were infants younger than one year and 89.2% (140 of 157) were classified correctly. Of the 44 patients with KD who had coronary artery abnormalities, our diagnostic algorithm correctly identified 43 (97.7%) including all patients with dilated coronary artery but one who found to resolve in 8 weeks.

Interpretation

This work demonstrates the applicability of our algorithmic approach and diagnostic portability in Taiwan.

Can cognition help predict suicide risk in patients with major depressive disorder? A machine learning study

BACKGROUND

Previous studies suggest that deficits in cognition may increase the risk of suicide. Our study aims to develop a machine learning (ML) algorithm-based suicide risk prediction model using cognition in patients with major depressive disorder (MDD).

METHODS

Participants comprised 52 depressed suicide attempters (DSA) and 61 depressed non-suicide attempters (DNS), and 98 healthy controls (HC). All participants were required to complete a series of questionnaires, the Suicide Stroop Task (SST) and the Iowa Gambling Task (IGT). The performance in IGT was analyzed using repeated measures ANOVA. ML with extreme gradient boosting (XGBoost) classification algorithm and locally explanatory techniques assessed performance and relative importance of characteristics for predicting suicide attempts. Prediction performances were compared with the area under the curve (AUC), decision curve analysis (DCA), and net reclassification improvement (NRI).

RESULTS

DSA and DNS preferred to select the card from disadvantageous decks (decks "A" + "B") under risky situation (p = 0.023) and showed a significantly poorer learning effect during the IGT (F = 2.331, p = 0.019) compared with HC. Performance of XGBoost model based on demographic and clinical characteristics was compared with that of the model created after adding cognition data (AUC, 0.779 vs. 0.819, p > 0.05). The net benefit of model was improved and cognition resulted in continuous reclassification improvement with NRI of 5.3%. Several clinical dimensions were significant predictors in the XGBoost classification algorithm.

LIMITATIONS

A limited sample size and failure to include sufficient suicide risk factors in the predictive model.

CONCLUSION

This study demonstrate that cognitive deficits may serve as an important risk factor to predict suicide attempts in patients with MDD. Combined with other demographic characteristics and attributes drawn from clinical questionnaires, cognitive function can improve the predictive effectiveness of the ML model. Additionally, explanatory ML models can help clinicians detect specific risk factors for each suicide attempter within MDD patients. These findings may be helpful for clinicians to detect those at high risk of suicide attempts quickly and accurately, and help them make proactive treatment decisions.

Structured data vs. unstructured data in machine learning prediction models for suicidal behaviors: A systematic review and meta-analysis

Suicide remains a leading cause of preventable death worldwide, despite advances in research and decreases in mental health stigma through government health campaigns. Machine learning (ML), a type of artificial intelligence (AI), is the use of algorithms to simulate and imitate human cognition. Given the lack of improvement in clinician-based suicide prediction over time, advancements in technology have allowed for novel approaches to predicting suicide risk. This systematic review and meta-analysis aimed to synthesize current research regarding data sources in ML prediction of suicide risk, incorporating and comparing outcomes between structured data (human interpretable such as psychometric instruments) and unstructured data (only machine interpretable such as electronic health records). Online databases and gray literature were searched for studies relating to ML and suicide risk prediction. There were 31 eligible studies. The outcome for all studies combined was AUC = 0.860, structured data showed AUC = 0.873, and unstructured data was calculated at AUC = 0.866. There was substantial heterogeneity between the studies, the sources of which were unable to be defined. The studies showed good accuracy levels in the prediction of suicide risk behavior overall. Structured data and unstructured data also showed similar outcome accuracy according to meta-analysis, despite different volumes and types of input data.

Achieving health equity in US suicides: a narrative review and commentary

Suicide rates in the United States (US) reached a peak in 2018 and declined in 2019 and 2020, with substantial and often growing disparities by age, sex, race/ethnicity, geography, veteran status, sexual minority status, socioeconomic status, and method employed (means disparity). In this narrative review and commentary, we highlight these many disparities in US suicide deaths, then examine the possible causes and potential solutions, with the overarching goal of reducing suicide death disparities to achieve health equity.The data implicate untreated, undertreated, or unidentified depression or other mental illness, and access to firearms, as two modifiable risk factors for suicide across all groups. The data also reveal firearm suicides increasing sharply and linearly with increasing county rurality, while suicide rates by falls (e.g., from tall structures) decrease linearly by increasing rurality, and suicide rates by other means remain fairly constant regardless of relative county urbanization. In addition, for all geographies, gun suicides are significantly higher in males than females, and highest in ages 51-85 + years old for both sexes. Of all US suicides from 1999-2019, 55% of male suicides and 29% of female suicides were by gun in metropolitan (metro) areas, versus 65% (Male) and 42% (Female) suicides by gun in non-metro areas. Guns accounted for 89% of suicides in non-metro males aged 71-85 + years old. Guns (i.e., employment of more lethal means) are also thought to be a major reason why males have, on average, 2-4 times higher suicide rates than women, despite having only 1/4-1/2 as many suicide attempts as women. Overall the literature and data strongly implicate firearm access as a risk factor for suicide across all populations, and even more so for male, rural, and older populations.To achieve the most significant results in suicide prevention across all groups, we need 1) more emphasis on policies and universal programs to reduce suicidal behaviors, and 2) enhanced population-based strategies for ameliorating the two most prominent modifiable targets for suicide prevention: depression and firearms.

Individualized prediction of psychiatric readmissions for patients with major depressive disorder: a 10-year retrospective cohort study

Patients with major depressive disorder (MDD) are at high risk of psychiatric readmission while the factors associated with such adverse illness trajectories and the impact of the same factor at different follow-up times remain unclear. Based on machine learning (ML) approaches and real-world electronic medical records (EMR), we aimed to predict individual psychiatric readmission within 30, 60, 90, 180, and 365 days of an initial major depression hospitalization. In addition, we examined to what extent our prediction model could be made interpretable by quantifying and visualizing the features that drive the predictions at different follow-up times. By identifying 13,177 individuals discharged from a hospital located in western China between 2009 and 2018 with a recorded diagnosis of MDD, we established five prediction-modeling cohorts with different follow-up times. Four different ML models were trained with features extracted from the EMR, and explainable methods (SHAP and Break Down) were utilized to analyze the contribution of each of the features at both population-level and individual-level. The model showed a performance on the holdout testing dataset that decreased over follow-up time after discharge: AUC 0.814 (0.758-0.87) within 30 days, AUC 0.780 (0.728-0.833) within 60 days, AUC 0.798 (0.75-0.846) within 90 days, AUC 0.740 (0.687-0.794) within 180 days, and AUC 0.711 (0.676-0.747) within 365 days. Results add evidence that markers of depression severity and symptoms (recurrence of the symptoms, combination of key symptoms, the number of core symptoms and physical symptoms), along with age, gender, type of payment, length of stay, comorbidity, treatment patterns such as the use of anxiolytics, antipsychotics, antidepressants (especially Fluoxetine, Clonazepam, Olanzapine, and Alprazolam), physiotherapy, and psychotherapy, and vital signs like pulse and SBP, may improve prediction of psychiatric readmission. Some features can drive the prediction towards readmission at one follow-up time and towards non-readmission at another. Using such a model for decision support gives the clinician dynamic information of the patient's risk of psychiatric readmission and the specific features pulling towards readmission. This finding points to the potential of establishing personalized interventions that change with follow-up time.

Testing the Effectiveness of a Blended Intervention to Reduce Suicidal Ideation among School Adolescents in Chile: A Protocol for a Cluster Randomized Controlled Trial

Suicidal ideation is prevalent in adolescents and is a marker for subsequent psychiatric vulnerability and symptom severity. Literature shows that blended care (integrating online and offline components in a treatment process) could improve the effectiveness and adherence of interventions targeting suicidal ideation in adolescents, but the evidence is inconclusive. Thus, we will test the effectiveness of a blended intervention to reduce suicidal ideation (primary outcome) in school settings using a single-blind two-armed cluster randomized controlled trial (cRCT). The internet-based component corresponds to the Reframe-IT, a program encompassing eight online sessions based on cognitive-behavioral therapy (CBT) principles. The face-to-face intervention will be delivered through four CBT sessions. Additionally, we will assess the effect of the intervention on the following secondary outcomes: suicidal attempts, depressive symptoms, hopelessness, emotional regulation, and problem-solving skills. Primary and secondary outcomes will be assessed at post-intervention, 3-month, 6-month, and 12-month follow-up. Finally, we will explore the mediation role of cognitive, emotional, and behavioral correlates of suicide on the effect of the intervention. Results will inform whether the intervention can reduce suicide among school adolescents and be implemented on a large scale in Chile.

Translating promise into practice: a review of machine learning in suicide research and prevention

In ever more pressured health-care systems, technological solutions offering scalability of care and better resource targeting are appealing. Research on machine learning as a technique for identifying individuals at risk of suicidal ideation, suicide attempts, and death has grown rapidly. This research often places great emphasis on the promise of machine learning for preventing suicide, but overlooks the practical, clinical implementation issues that might preclude delivering on such a promise. In this Review, we synthesise the broad empirical and review literature on electronic health record-based machine learning in suicide research, and focus on matters of crucial importance for implementation of machine learning in clinical practice. The challenge of preventing statistically rare outcomes is well known; progress requires tackling data quality, transparency, and ethical issues. In the future, machine learning models might be explored as methods to enable targeting of interventions to specific individuals depending upon their level of need-ie, for precision medicine. Primarily, however, the promise of machine learning for suicide prevention is limited by the scarcity of high-quality scalable interventions available to individuals identified by machine learning as being at risk of suicide.

Artificial intelligence and suicide prevention: a systematic review

Background

Suicide is one of the main preventable causes of death. Artificial intelligence (AI) could improve methods for assessing suicide risk. The objective of this review is to assess the potential of AI 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.

Conclusions

AI appears to have a 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.

Application of Artificial Intelligence on Psychological Interventions and Diagnosis: An Overview

BACKGROUND

Innovative technologies, such as machine learning, big data, and artificial intelligence (AI) are approaches adopted for personalized medicine, and psychological interventions and diagnosis are facing huge paradigm shifts. In this literature review, we aim to highlight potential applications of AI on psychological interventions and diagnosis.

METHODS

This literature review manifest studies that discuss how innovative technology as deep learning (DL) and AI is affecting psychological assessment and psychotherapy, we performed a search on PUBMED, and Web of Science using the terms "psychological interventions," "diagnosis on mental health disorders," "artificial intelligence," and "deep learning." Only studies considering patients' datasets are considered.

RESULTS

Nine studies met the inclusion criteria. Beneficial effects on clinical symptoms or prediction were shown in these studies, but future study is needed to determine the long-term effects.

LIMITATIONS

The major limitation for the current study is the small sample size, and lies in the lack of long-term follow-up-controlled studies for a certain symptom.

CONCLUSIONS

AI such as DL applications showed promising results on clinical practice, which could lead to profound impact on personalized medicine for mental health conditions. Future studies can improve furthermore by increasing sample sizes and focusing on ethical approvals and adherence for online-therapy.

Development of Autonomic Nervous System Assays as Point-of-Care Tests to Supplement Clinical Judgment in Risk Assessment for Suicidal Behavior: A Review

PURPOSE OF REVIEW

A biomarker point-of-care (POC) test that supplements the psychiatric interview and improves detection of patients at risk for suicide would be of value, and assays of autonomic nervous system (ANS) activity would satisfy the logistical requirements for a POC test. We performed a selective review of the available literature of ANS assays related to risk for suicide.

RECENT FINDINGS

We searched PubMed and Web of Science with the strategy: "suicide OR suicidal" AND "electrodermal OR heart rate variability OR pupillometry OR pupillography." The search produced 119 items, 21 of which provided original data regarding ANS methods and suicide. These 21 studies included 6 for electrodermal activity, 14 for heart rate variability, and 1 for the pupillary light reflex. The 21 papers showed associations between ANS assays and suicide risk in a direction suggesting underlying hyperarousal in patients at risk for suicide. ANS assays show promise for future development as POC tests to supplement clinical decision making in estimating risk for suicide.

5-year incidence of suicide-risk in youth: A gradient tree boosting and SHAP study

BACKGROUND

Machine learning methods for suicidal behavior so far have failed to be implemented as a prediction tool. In order to use the capabilities of machine learning to model complex phenomenon, we assessed the predictors of suicide risk using state-of-the-art model explanation methods.

METHODS

Prospective cohort study including a community sample of 1,560 young adults aged between 18 and 24. The first wave took place between 2007 and 2009, and the second wave took place between 2012 and 2014. Sociodemographic and clinical characteristics were assessed at baseline. Incidence of suicide risk at five-years of follow-up was the main outcome. The outcome was assessed using the Mini Neuropsychiatric Interview (MINI) at both waves.

RESULTS

The risk factors for the incidence of suicide risk at follow-up were: female sex, lower socioeconomic status, older age, not studying, presence of common mental disorder symptoms, and poor quality of life. The interaction between overall health and socioeconomic status in relation to suicide risk was also captured and shows a shift from protection to risk by socioeconomic status as overall health increases.

LIMITATIONS

Proximal factors associated with the incidence of suicide risk were not assessed.

CONCLUSIONS

Our findings indicate that factors related to poor quality of life, not studying, and common mental disorder symptoms of young adults are already in place prior to suicide risk. Most factors present critical non-linear patterns that were identified. These findings are clinically relevant because they can help clinicians to early detect suicide risk.

Identifying socio-demographic risk factors for suicide using data on an individual level

Background

Suicide is a complex issue. Due to the relative rarity of the event, studies into risk factors are regularly limited by sample size or biased samples. The aims of the study were to find risk factors for suicide that are robust to intercorrelation, and which were based on a large and unbiased sample.

Methods

Using a training set of 5854 suicides and 596,416 control cases, we fit a logistic regression model and then evaluate the performance on a test set of 1425 suicides and 594,893 control cases. The data used was micro-data of Statistics Netherlands (CBS) with data on each inhabitant of the Netherlands.

Results

Taking the effect of possible correlating risk factors into account, those with a higher risk for suicide are men, middle-aged people, people with low income, those living alone, the unemployed, and those with mental or physical health problems. People with a lower risk are the highly educated, those with a non-western immigration background, and those living with a partner.

Conclusion

We confirmed previously known risk factors such as male gender, middle-age, and low income and found that they are risk factors that are robust to intercorrelation. We found that debt and urbanicity were mostly insignificant and found that the regional differences found in raw frequencies are mostly explained away after correction of correlating risk factors, indicating that these differences were primarily caused due to the differences in the demographic makeup of the regions. We found an AUC of 0.77, which is high for a model predicting suicide death and comparable to the performance of deep learning models but with the benefit of remaining explainable.

Public Health and Epidemiology Informatics: Recent Research Trends

OBJECTIVES

To introduce and analyse current trends in Public Health and Epidemiology Informatics.

METHODS

PubMed search of 2020 literature on public health and epidemiology informatics was conducted and all retrieved references were reviewed by the two section editors. Then, 15 candidate best papers were selected among the 920 references. These papers were then peer-reviewed by the two section editors, two chief editors, and external reviewers, including at least two senior faculty, to allow the Editorial Committee of the 2021 International Medical Informatics Association (IMIA) Yearbook to make an informed decision regarding the selection of the best papers.

RESULTS

Among the 920 references retrieved from PubMed, four were suggested as best papers and the first three were finally selected. The fourth paper was excluded because of reproducibility issues. The first best paper is a very public health focused paper with health informatics and biostatistics methods applied to stratify patients within a cohort in order to identify those at risk of suicide; the second paper describes the use of a randomized design to test the likely impact of fear-based messages, with and without empowering self-management elements, on patient consultations or antibiotic requests for influenza-like illnesses. The third selected paper evaluates the perception among communities of routine use of Whole Genome Sequencing and Big Data technologies to capture more detailed and specific personal information.

CONCLUSIONS

The findings from the three studies suggest that using Public Health and Epidemiology Informatics methods could leverage, when combined with Deep Learning, early interventions and appropriate treatments to mitigate suicide risk. Further, they also demonstrate that well informing and empowering patients could help them to be involved more in their care process.

Social and Behavioral Determinants of Health in the Era of Artificial Intelligence with Electronic Health Records: A Scoping Review

Background. There is growing evidence that social and behavioral determinants of health (SBDH) play a substantial effect in a wide range of health outcomes. Electronic health records (EHRs) have been widely employed to conduct observational studies in the age of artificial intelligence (AI). However, there has been limited review into how to make the most of SBDH information from EHRs using AI approaches.Methods. A systematic search was conducted in six databases to find relevant peer-reviewed publications that had recently been published. Relevance was determined by screening and evaluating the articles. Based on selected relevant studies, a methodological analysis of AI algorithms leveraging SBDH information in EHR data was provided.Results. Our synthesis was driven by an analysis of SBDH categories, the relationship between SBDH and healthcare-related statuses, natural language processing (NLP) approaches for extracting SBDH from clinical notes, and predictive models using SBDH for health outcomes.Discussion. The associations between SBDH and health outcomes are complicated and diverse; several pathways may be involved. Using NLP technology to support the extraction of SBDH and other clinical ideas simplifies the identification and extraction of essential concepts from clinical data, efficiently unlocks unstructured data, and aids in the resolution of unstructured data-related issues.Conclusion. Despite known associations between SBDH and diseases, SBDH factors are rarely investigated as interventions to improve patient outcomes. Gaining knowledge about SBDH and how SBDH data can be collected from EHRs using NLP approaches and predictive models improves the chances of influencing health policy change for patient wellness, ultimately promoting health and health equity.

Deep graph neural network-based prediction of acute suicidal ideation in young adults

Precise remote evaluation of both suicide risk and psychiatric disorders is critical for suicide prevention as well as for psychiatric well-being. Using questionnaires is an alternative to labor-intensive diagnostic interviews in a large general population, but previous models for predicting suicide attempts suffered from low sensitivity. We developed and validated a deep graph neural network model that increased the prediction sensitivity of suicide risk in young adults (n = 17,482 for training; n = 14,238 for testing) using multi-dimensional questionnaires and suicidal ideation within 2 weeks as the prediction target. The best model achieved a sensitivity of 76.3%, specificity of 83.4%, and an area under curve of 0.878 (95% confidence interval, 0.855-0.899). We demonstrated that multi-dimensional deep features covering depression, anxiety, resilience, self-esteem, and clinico-demographic information contribute to the prediction of suicidal ideation. Our model might be useful for the remote evaluation of suicide risk in the general population of young adults for specific situations such as the COVID-19 pandemic.

MIMIC-SBDH: A Dataset for Social and Behavioral Determinants of Health

Social and Behavioral Determinants of Health (SBDHs) are environmental and behavioral factors that have a profound impact on health and related outcomes. Given their importance, physicians document SBDHs of their patients in Electronic Health Records (EHRs). However, SBDHs are mostly documented in unstructured EHR notes. Determining the status of the SBDHs requires manually reviewing the notes which can be a tedious process. Therefore, there is a need to automate identifying the patients' SBDH status in EHR notes. In this work, we created MIMIC-SBDH, the first publicly available dataset of EHR notes annotated for patients' SBDH status. Specifically, we annotated 7,025 discharge summary notes for the status of 7 SBDHs as well as marked SBDH-related keywords. Using this annotated data for training and evaluation, we evaluated the performance of three machine learning models (Random Forest, XGBoost, and Bio-ClinicalBERT) on the task of identifying SBDH status in EHR notes. The performance ranged from the lowest 0.69 F1 score for Drug Use to the highest 0.96 F1 score for Community-Present. In addition to standard evaluation metrics such as the F1 score, we evaluated four capabilities that a model must possess to perform well on the task using the CheckList tool (Ribeiro et al., 2020). The results revealed several shortcomings of the models. Our results highlighted the need to perform more capability-centric evaluations in addition to standard metric comparisons.

Smart Healthcare System Based on Cloud-Internet of Things and Deep Learning

Introduction

Health monitoring and remote diagnosis can be realized through Smart Healthcare. In view of the existing problems such as simple measurement parameters of wearable devices, huge computing pressure of cloud servers, and lack of individualization of diagnosis, a novel Cloud-Internet of Things (C-IOT) framework for medical monitoring is put forward.

Methods

Smart phones are adopted as gateway devices to achieve data standardization and preprocess to generate health gray-scale map uploaded to the cloud server. The cloud server realizes the business logic processing and uses the deep learning model to carry out the gray-scale map calculation of health parameters. A deep learning model based on the convolution neural network (CNN) is constructed, in which six volunteers are selected to participate in the experiment, and their health data are marked by private doctors to generate initial data set.

Results

Experimental results show the feasibility of the proposed framework. The test data set is used to test the CNN model after training; the forecast accuracy is over 77.6%.

Conclusion

The CNN model performs well in the recognition of health status. Collectively, this Smart Healthcare System is expected to assist doctors by improving the diagnosis of health status in clinical practice.

Social determinants of health in electronic health records and their impact on analysis and risk prediction: A systematic review

OBJECTIVE

This integrative review identifies and analyzes the extant literature to examine the integration of social determinants of health (SDoH) domains into electronic health records (EHRs), their impact on risk prediction, and the specific outcomes and SDoH domains that have been tracked.

MATERIALS AND METHODS

In accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we conducted a literature search in the PubMed, CINAHL, Cochrane, EMBASE, and PsycINFO databases for English language studies published until March 2020 that examined SDoH domains in the context of EHRs.

RESULTS

Our search strategy identified 71 unique studies that are directly related to the research questions. 75% of the included studies were published since 2017, and 68% were U.S.-based. 79% of the reviewed articles integrated SDoH information from external data sources into EHRs, and the rest of them extracted SDoH information from unstructured clinical notes in the EHRs. We found that all but 1 study using external area-level SDoH data reported minimum contribution to performance improvement in the predictive models. In contrast, studies that incorporated individual-level SDoH data reported improved predictive performance of various outcomes such as service referrals, medication adherence, and risk of 30-day readmission. We also found little consensus on the SDoH measures used in the literature and current screening tools.

CONCLUSIONS

The literature provides early and rapidly growing evidence that integrating individual-level SDoH into EHRs can assist in risk assessment and predicting healthcare utilization and health outcomes, which further motivates efforts to collect and standardize patient-level SDoH information.

A Comprehensive Review of Computer-Aided Diagnosis of Major Mental and Neurological Disorders and Suicide: A Biostatistical Perspective on Data Mining

The World Health Organization (WHO) suggests that mental disorders, neurological disorders, and suicide are growing causes of morbidity. Depressive disorders, schizophrenia, bipolar disorder, Alzheimer's disease, and other dementias account for 1.84%, 0.60%, 0.33%, and 1.00% of total Disability Adjusted Life Years (DALYs). Furthermore, suicide, the 15th leading cause of death worldwide, could be linked to mental disorders. More than 68 computer-aided diagnosis (CAD) methods published in peer-reviewed journals from 2016 to 2021 were analyzed, among which 75% were published in the year 2018 or later. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol was adopted to select the relevant studies. In addition to the gold standard, the sample size, neuroimaging techniques or biomarkers, validation frameworks, the classifiers, and the performance indices were analyzed. We further discussed how various performance indices are essential based on the biostatistical and data mining perspective. Moreover, critical information related to the Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guidelines was analyzed. We discussed how balancing the dataset and not using external validation could hinder the generalization of the CAD methods. We provided the list of the critical issues to consider in such studies.

Demographic Predictors of Dropping Out of Treatment (DOT) in Substance Use Disorder Treatment

BACKGROUND

Researchers have not studied or used novel methods for identifying potential disparities for sexual minorities, those with criminal pasts, and veterans in (DOT).

METHODS

We used Bayesian logistic regression to identify factors associated with DOT, tested interaction effects, and used machine learning to classify qualitative responses.

FINDINGS

With 2,772 clients from two inpatient clinics in the Southwest United States, we found sexual minorities and females had 52% and 61%, increases and African Americans had 54% decreases in the odds of DOT. Additionally, those with a criminal past and 34.5 and older were less likely to DOT by 5% relative to clients with no prior involvement in the criminal justice system.

CONCLUSIONS

This study illustrated the disparities for women and sexual minorities in DOT as well as demonstrated novel methodological approaches to addressing previously unanswered questions.

Suicide Risk Assessment Using Machine Learning and Social Networks: a Scoping Review

According to the World Health Organization (WHO) report in 2016, around 800,000 of individuals have committed suicide. Moreover, suicide is the second cause of unnatural death in people between 15 and 29 years. This paper reviews state of the art on the literature concerning the use of machine learning methods for suicide detection on social networks. Consequently, the objectives, data collection techniques, development process and the validation metrics used for suicide detection on social networks are analyzed. The authors conducted a scoping review using the methodology proposed by Arksey and O'Malley et al. and the PRISMA protocol was adopted to select the relevant studies. This scoping review aims to identify the machine learning techniques used to predict suicide risk based on information posted on social networks. The databases used are PubMed, Science Direct, IEEE Xplore and Web of Science. In total, 50% of the included studies (8/16) report explicitly the use of data mining techniques for feature extraction, feature detection or entity identification. The most commonly reported method was the Linguistic Inquiry and Word Count (4/8, 50%), followed by Latent Dirichlet Analysis, Latent Semantic Analysis, and Word2vec (2/8, 25%). Non-negative Matrix Factorization and Principal Component Analysis were used only in one of the included studies (12.5%). In total, 3 out of 8 research papers (37.5%) combined more than one of those techniques. Supported Vector Machine was implemented in 10 out of the 16 included studies (62.5%). Finally, 75% of the analyzed studies implement machine learning-based models using Python.

Deep neural networks detect suicide risk from textual facebook posts

Detection of suicide risk is a highly prioritized, yet complicated task. Five decades of research have produced predictions slightly better than chance (AUCs = 0.56–0.58). In this study, Artificial Neural Network (ANN) models were constructed to predict suicide risk from everyday language of social media users. The dataset included 83,292 postings authored by 1002 authenticated Facebook users, alongside valid psychosocial information about the users. Using Deep Contextualized Word Embeddings for text representation, two models were constructed: A Single Task Model (STM), to predict suicide risk from Facebook postings directly (Facebook texts → suicide) and a Multi-Task Model (MTM), which included hierarchical, multilayered sets of theory-driven risk factors (Facebook texts → personality traits → psychosocial risks → psychiatric disorders → suicide). Compared with the STM predictions (0.621 ≤ AUC ≤ 0.629), the MTM produced significantly improved prediction accuracy (0.697 ≤ AUC ≤ 0.746), with substantially larger effect sizes (0.729 ≤ d ≤ 0.936). Subsequent content analyses suggested that predictions did not rely on explicit suicide-related themes, but on a range of text features. The findings suggest that machine learning based analyses of everyday social media activity can improve suicide risk predictions and contribute to the development of practical detection tools.