.png)
Ongoing Projects
Click on the boxes below to learn more about our main projects!
Project STARS: Short-term Assessment of Risk for Suicide
The major goal of this study is to evaluate candidate neurocognitive markers of short-term risk for suicidal behavior in clinically acute adolescents using novel advances in computational psychiatry.
Inpatient
Smartphone
Wearable
​PI: Richard Liu
Funding: National Institute of Mental Health R01 MH115905
Suicidal behavior often occurs during periods of acute arousal, and these periods of acute stress can compromise cognitive functioning and decision-making. This may be even more so for individuals struggling with their mental health, particularly individuals at risk for suicide. This study will evaluate whether executive control and other aspects of neurocognitive functioning during acute stress may be short-term risk indicators for suicidal behavior in adolescents. The eventual goal of this project is to develop a youth suicide risk prediction algorithm that may inform prevention and clinical decision-making in the ER and inpatient visits, points of clinical contact with high levels of acute stress.
Project MOON
The major goal of this study is to adopt ambulatory assessments of stress, sleep disturbance, and physiological arousal, and a measure of stress-related executive control in evaluating the interrelation of these risk indices in characterizing short-term risk for suicidal behavior in clinically acute adolescents.
Inpatient
Smartphone
Wearable
PI: Richard Liu
Funding: National Institute of Mental Health RF1 MH120830
The major goal of this study is to leverage recent developments in mobile technology to conduct mobile assessments of psychosocial stress (i.e., through smartphone apps and social media activity and content), physiological arousal, sleep disturbance in relation to stress-related executive control to characterize short-term risk for suicidal behavior in clinically acute adolescents continuously in the month after discharge, the period of greatest risk for suicidal behavior. The translational goal of this project is to develop an algorithm that can alert clinicians and caregivers to acute suicide risk in teens in outpatient care and other settings with regular repeated contact.
Project MARS: Multi-site Assessment of Risk for Suicide​
The goal of this project is to explore proximal mechanisms of the relationship between negative interpersonal events in family and suicidal thoughts and behaviors.
Inpatient
Smartphone
Wearable
​MPIs: Evan Kleiman, Cassie Glenn, Richard Liu
Funding: National Institute of Mental Health R01 MH12489
In our prior work, we have found that for a substantial proportion of teens admitted to inpatient care, suicidal thoughts and behaviors that result in psychiatric admission are precipitated by interpersonal conflict, often within family relationships. The goal of this project is to leverage mobile technology with both caregivers and teens to evaluate real-time interpersonal family dynamics in relation to suicidal thoughts and behaviors in the month after discharge from inpatient care. Caregivers have an integral role in psychotherapy for suicidal youth. In addition to improving our ability to predict acute suicidal risk, the goal of this project is to inform the development of new psychotherapeutic strategies for treating suicidal youth.
Project CARE: Computer Assisted Risk Evaluation
The goal of our study is to better understand teens' self-harm and risk for suicide. We hope that by collecting pictures and information about scars, we will be able to build a tool to help doctors and other healthcare workers better understand self-injury and improve teens' mental health.
Inpatient
Community
Smartphone
Social Media
Machine Learning
MPIs: Taylor Burke & Thomas Serre
Funding: National Institute of Mental Health R21 MH127231
This project aims to utilize cutting-edge computer vision techniques to automate the visual assessment of self-injury severity and determine the utility of these visual signals in predicting prospective suicide attempt risk. This proof-of-concept study will set the stage to pursue our long-term goal of integrating this technology into psychiatric care entry-points (e.g., EDs) to assess whether it can serve as a clinical decision-support tool.
​
Evaluating the predictive validity of computational markers of self-injury among high-risk youth
PI: Taylor Burke
Funding: American Foundation for Suicide Prevention YIG-1-030-20
The objective of this study is to extend ongoing research supported by a National Institute of Mental Health-funded R21 aimed at utilizing computer vision techniques to automate the assessment of self-injury visual severity indicators and to determine the utility of these visual signals in predicting suicide risk. This study will fund the recruitment of a sample of psychiatrically hospitalized adolescents followed prospectively. One month after psychiatric hospitalization discharge, adolescents will be assessed for prospective engagement in suicide attempts. Deep convolutional neural networks will be applied to the images to detect severity indices of self-injury and to examine their accuracy in predicting prospective suicide attempt risk in this high-risk clinical sample.
​
To learn more about Project CARE, visit its website: mghprojectcare.org
Project INSIGHT: Identifying Novel Signals of Idiographic Health-risk among Teens
The goal of this project is to employ mobile sensing and actigraphy to assess whether acute behavioral changes from typical patterns of social engagement, sleep, and physical activity indicate proximal risk for increases in suicidal ideation in high-risk adolescents.
Inpatient
Smartphone
Wearable
Social Media
PI: Taylor Burke
Funding: National Institute of Mental Health K23 MH126168
The goal of this project is to employ mobile sensing and actigraphy to assess whether objectively and passively measured acute behavioral changes from typical patterns of social engagement, sleep, and physical activity indicate proximal risk for increases in suicidal ideation using idiographic n-of-1 models in high-risk adolescents. This study aims to set the stage to develop and test the efficacy of personalized brief treatments.
Project Anchor
The goal of this project is to recruit parents and guardians who use Bark to learn more about their experiences receiving alerts that their child may be at risk of self-harm and suicide. We hope to use results from this study to design interventions to help parents and guardians navigate these alerts in the future.
Community
Parents/Guardians
Smartphone
Social Media
Co-Investigator: Taylor Burke (PI: Kathryn Fox)
Funding: Mental Research Institute
The purpose of this grant is to better understand guardians’ experiences learning that their child may be at risk for suicide or self-harm. Results will inform the development of a digital intervention for guardians aimed at helping guardians keep their children safe while managing their own emotional reactions, maintaining relationships, and maximizing child autonomy. After intervention development, a randomized control trial will be conducted to assess its efficacy. If efficacious, this intervention will be rolled out in a scalable manner through partnership with industry.
​
To learn more about Project Anchor, visit its website: projectanchor.org
.png)
Project SPACE: Study of Parents' And Children's Experiences with Suicide
This project aims to examine how parent-child stress interactions may be associated with or protect against suicide risk in adolescents using automatic sensing of acoustic and visual behaviors.
Inpatient
Parents/Guardians
Machine Learning
MPIs: Richard Liu & Taylor Burke
Funding: National Institute of Mental Health R21 MH130767
This project aims to examine how parent-child stress interactions may be associated with or protect against suicide risk in adolescents using automatic sensing of acoustic and visual behaviors.
Project PLUTO
This project aims to understand how teens' daily social media interactions with parents and peers are related to risk for
suicidal thoughts and behaviors.
Inpatient
Smartphone
Social Media
Machine Learning
PI: Richard Liu
Funding: National Institute of Mental Health K24 MH136418
Understanding how social media may affect teens' mental health has been an increasing priority in recent years. In 2023, the US Senate held a congressional hearing on this issue and the Surgeon General issued a general warning of the effects of social media on youth mental health. Although a lot of studies have been published on frequency of social media use, little is known about how the content of teens' social media affects their mental health. In our work in clinical care settings, many teens that are hospitalized for suicidal thoughts and behaviors experience stress with family and peers shortly before hospitalization. It is possible that the interpersonal stress teens experience in their social media interactions may precipitate suicidal thoughts and behaviors. The aim of this project is to use advanced analytical techniques (machine learning) to develop a program to automate detection of interpersonal stress in teens' social media content and to evaluate them in relation to suicidal thoughts and behaviors in the subsequent hours and days.
Project PUPS
This project aims to understand risk for suicidal thoughts and behaviors among preteens and develop new tools to help
parents, preadolescents, and doctors to help prevent suicidal thoughts and acts and related outcomes in the future.
Inpatient
Parents/Guardians
​MPIs: Evan Kleiman, Cassie Glenn, Richard Liu
Funding: National Institute of Mental Health 1R01MH137793-01
Suicide risk is increasing in preteens, i.e. 8-12 years old: in clinical samples (e.g., inpatient psychiatry), more than 40% have seriously considered suicide and nearly 20% have attempted suicide. Our knowledge of suicide risk has been limited by the paucity of developmentally appropriate tools to assess suicidal thoughts, and relevant risk factors, in preteens. The goal of this study is to adapt a developmentally and culturally appropriate measures of suicide risk for preteens (via community-engaged research) and to examine different phenotypes (i.e., presentations) of suicide risk among at-risk preteens following discharge from acute psychiatric care settings.