HEADS

HEADS

Human-in-the-loop Evaluation of Assisted Depression Screening

Lekhansh Shukla · National Institute of Mental Health and Neuro Sciences · India
What they're working onWe develop and evaluate HEADS (Human-in-the-loop Evaluation of Assisted Depression Screening), a multilingual, modular, interpretable, HIL, LLM-based automated pipeline for clinical depression diagnosis (DSM-5) and severity (PHQ-9) in diverse clinical populations. Psychiatrists will collect data via QuickSCID-5 and PHQ-9, with adaptive sampling across languages, genders, comorbidity. HEADS has individually trainable modules: (i) Automatic Speech Recognition (ASR) supporting five Indic languages; (ii) Domain-sensitive translation (Indic to English); (iii) Inference module providing summaries, DSM-5 diagnoses, PHQ-9 score-bands, reasoning and confidence scores. We will also generate deidentified datasets annotated by psychiatrists (audio interviews, native-language transcripts, English translations and psychiatric inference). The key goal is to create a safe (HIL), scalable, non-intrusive automated system to help non-specialists measure depression in culturally diverse populations, potentially addressing mental health gaps due to shortage of psychiatrists, underdiagnosis or misdiagnosis. Lived experience experts review all practical steps of the study, including consenting, interviewing, transcription and translation. With LEEs embedded at all stages of the study, we also co-design bias and safety analysis protocols collaboratively.
AI approach
LLMs · automatic speech recognition · translation · human-in-the-loop AI
Mental health focus
Depression · screening · diagnosis and severity
Links
  •  https://heads-ai.com/ 
Social media
  • LinkedIn:  https://www.linkedin.com/in/human-in-the-loop-evaluation-of-assisted-depression-screening-heads-0a552b389/ 
  • X:  https://x.com/HEADS_AI