Expanding the reach of psychosocial services for youth: Untapped potential of mentor‐delivered single session interventions

Hart, Mackenzie J.; Sung, Jenna; McQuillin, Samuel D.; Schleider, Jessica L.

doi:10.1002/jcop.22927

Cited by 3 publications

(4 citation statements)

References 84 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, neural network denoising has become the state-of-the-art approach to solving speech denoising. Therefore, one of the latest works [ 3 ] from Facebook was applied. It uses Demucs architecture, which was originally developed for the task of music source separation [ 12 ].…”

Section: Methodsmentioning

confidence: 99%

“…Conventional training methods, which typically involve long durations of apprenticeship and supervision, are often resource-intensive. The current system struggles to meet the growing demand for mental health services, especially for pediatric patients and youths [ 3 ]. Furthermore, evaluating the effectiveness of therapy remains challenging, as mental health conditions are often complex and multifaceted, and current evaluation methods primarily rely on subjective assessments [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Objectively Quantifying Pediatric Psychiatric Severity Using Artificial Intelligence, Voice Recognition Technology, and Universal Emotions: Pilot Study for Artificial Intelligence-Enabled Innovation to Address Youth Mental Health Crisis

Caulley,

Alemu,

Burson

et al. 2023

JMIR Res Protoc

View full text Add to dashboard Cite

Background Providing Psychotherapy, particularly for youth, is a pressing challenge in the health care system. Traditional methods are resource-intensive, and there is a need for objective benchmarks to guide therapeutic interventions. Automated emotion detection from speech, using artificial intelligence, presents an emerging approach to address these challenges. Speech can carry vital information about emotional states, which can be used to improve mental health care services, especially when the person is suffering. Objective This study aims to develop and evaluate automated methods for detecting the intensity of emotions (anger, fear, sadness, and happiness) in audio recordings of patients’ speech. We also demonstrate the viability of deploying the models. Our model was validated in a previous publication by Alemu et al with limited voice samples. This follow-up study used significantly more voice samples to validate the previous model. Methods We used audio recordings of patients, specifically children with high adverse childhood experience (ACE) scores; the average ACE score was 5 or higher, at the highest risk for chronic disease and social or emotional problems; only 1 in 6 have a score of 4 or above. The patients’ structured voice sample was collected by reading a fixed script. In total, 4 highly trained therapists classified audio segments based on a scoring process of 4 emotions and their intensity levels for each of the 4 different emotions. We experimented with various preprocessing methods, including denoising, voice-activity detection, and diarization. Additionally, we explored various model architectures, including convolutional neural networks (CNNs) and transformers. We trained emotion-specific transformer-based models and a generalized CNN-based model to predict emotion intensities. Results The emotion-specific transformer-based model achieved a test-set precision and recall of 86% and 79%, respectively, for binary emotional intensity classification (high or low). In contrast, the CNN-based model, generalized to predict the intensity of 4 different emotions, achieved test-set precision and recall of 83% for each. Conclusions Automated emotion detection from patients’ speech using artificial intelligence models is found to be feasible, leading to a high level of accuracy. The transformer-based model exhibited better performance in emotion-specific detection, while the CNN-based model showed promise in generalized emotion detection. These models can serve as valuable decision-support tools for pediatricians and mental health providers to triage youth to appropriate levels of mental health care services. International Registered Report Identifier (IRRID) RR1-10.2196/51912

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Objectively Quantifying Pediatric Psychiatric Severity Using Artificial Intelligence, Voice Recognition Technology, and Universal Emotions: Pilot Study for Artificial Intelligence-Enabled Innovation to Address Youth Mental Health Crisis

Caulley,

Alemu,

Burson

et al. 2023

JMIR Res Protoc

View full text Add to dashboard Cite

show abstract

“…Moreover, Schwalbe and colleagues ( 2014 ) identified that the following features enhance traditional MI workshops: First, chunking training into smaller segments and tailoring these units to specific contexts; second, using booster sessions in providing MI trainings; and third, employing computer-, video-, or web-based technologies to administer trainings. Each of these aspects aligns with just-in-time trainings (JITTs), which have been recognized for their fit with task-shifting, particularly in expanding psychosocial services to youth (Hart et al, 2022b ; McQuillin et al, 2019 ). Just-in-time training is described as “only the training necessary, when it is necessary, to produce competent service provision” (p. 3–4).…”

Section: Background and Mechanisms Of Motivational Interviewingmentioning

confidence: 99%

“…These workers may be called lay providers, non-professionals, or paraprofessionals interchangeably (Saxe-Custack & Weatherspoon, 2013 ), though they are referred to paraprofessionals herein. In addition to growing specialized workforces and thereby increasing access to care (Hart et al, 2022b ; Schneeberger & Mathai, 2015 ), task-shifting has been recognized for yielding a more efficient distribution of resources and services by measures of money and time, and for enhancing the role of community providers in service delivery without sacrificing the quality of care provided (WHO, 2008 ; Zachariah et al, 2009 ). Though originally developed and studied in medical contexts, task-shifting has spread to other fields, such as mental health and education.…”

mentioning

confidence: 99%

Expanding School-Based Motivational Interviewing Through Delivery by Paraprofessional Providers: A Preliminary Scoping Review

Hart

McQuillin

Iachini

et al. 2023

School Mental Health

Self Cite

View full text Add to dashboard Cite

The supply of school mental health (SMH) providers and services cannot meet the demand of students in-need, and this gap is expected to widen in coming years. One way to increase the reach of helpful services for youth is to grow the SMH workforce through task-shifting to paraprofessionals. Task-shifting could be especially promising in expanding Motivational Interviewing (MI) interventions, as MI can be molded to target a number of academic and behavioral outcomes important to schools. However, no review of training exclusively paraprofessional samples in MI has yet been conducted. The current paper provides a scoping review of 19 studies of training paraprofessional providers to use MI to evaluate trainee characteristics, training content and format, and outcomes. Of these 19 studies, 15 reported that paraprofessionals improved in using MI following training. Nine studies reported that task-shifting MI was positively received by clients and/or providers. Six studies examined task-shifting MI in youth-serving contexts, and four examined the practice in traditional school contexts, suggesting its potential for use in SMH. Other findings and implications, such as client behavior change and provider fidelity, are shared, along with ideas for advancing research, practice, and policy in this subfield.

show abstract

Objectively Quantifying Pediatric Psychiatric Severity Using Artificial Intelligence, Voice Recognition Technology, and Universal Emotions: Pilot Study for Artificial Intelligence-Enabled Innovation to Address Youth Mental Health Crisis (Preprint)

Caulley,

Alemu,

Burson

et al. 2023

Preprint

View full text Add to dashboard Cite

BACKGROUND Providing Psychotherapy, particularly for youth, is a pressing challenge in the healthcare system. Traditional methods are resource-intensive, and there is a need for objective benchmarks to guide therapeutic interventions. Automated emotion detection from speech, using artificial intelligence (AI), presents an emerging approach to address these challenges. Speech can carry vital information about emotional states, which can be utilized to improve mental healthcare services, especially when the person is suffering. OBJECTIVE This study aims to develop and evaluate automated methods for detecting the intensity of emotions (anger, fear, sadness, and happiness) in audio recordings of patients’ speech. We also demonstrate the viability of deploying the models. Our model was validated in a previous publication by Alemu et al. (2023) with limited voice samples. This follow-up study used significantly more voice samples to validate the previous model. METHODS We used audio recordings of patients, specifically children with high Adverse Childhood Experience ACE scores; the average ACE score was five or higher, at the highest risk for chronic disease and social/emotional problems; only 1 in 6 have a score of 4 or above. The patients' structured voice sample was collected by reading a fixed script. Four highly trained therapists classified audio segments based on a scoring process of four emotions and their intensity levels for each of the four different emotions. We experimented with various preprocessing methods, including denoising, voice-activity detection, and diarization. Additionally, we explored various model architectures, including convolutional neural networks (CNNs) and transformers. We trained emotion-specific transformer-based models and a generalized CNN-based model to predict emotion intensities. RESULTS The emotion-specific transformer-based model achieved a test-set precision and recall of 86% and 79%, respectively, for binary emotional intensity classification (high or low). In contrast, the CNN-based model, generalized to predict the intensity of four different emotions, achieved test-set precision and recall of 83% for each. CONCLUSIONS Automated emotion detection from patients' speech using AI models is found to be feasible, leading to a high level of accuracy. The transformer-based model exhibited better performance in emotion-specific detection, while the CNN-based model showed promise in generalized emotion detection. These models can serve as valuable decision-support tools for pediatricians and mental health providers to triage youth to appropriate levels of mental healthcare services.

show abstract

Expanding the reach of psychosocial services for youth: Untapped potential of mentor‐delivered single session interventions

Cited by 3 publications

References 84 publications

Objectively Quantifying Pediatric Psychiatric Severity Using Artificial Intelligence, Voice Recognition Technology, and Universal Emotions: Pilot Study for Artificial Intelligence-Enabled Innovation to Address Youth Mental Health Crisis

Objectively Quantifying Pediatric Psychiatric Severity Using Artificial Intelligence, Voice Recognition Technology, and Universal Emotions: Pilot Study for Artificial Intelligence-Enabled Innovation to Address Youth Mental Health Crisis

Expanding School-Based Motivational Interviewing Through Delivery by Paraprofessional Providers: A Preliminary Scoping Review

Objectively Quantifying Pediatric Psychiatric Severity Using Artificial Intelligence, Voice Recognition Technology, and Universal Emotions: Pilot Study for Artificial Intelligence-Enabled Innovation to Address Youth Mental Health Crisis (Preprint)

Contact Info

Product

Resources

About