Alissa Valentine

Hi! I’m Alissa (she/they), a researcher studying mental health with the help of computational tools like LLMs and machine learning. Here you can find my projects, CV, and ways to get in touch.

GitHub Email LinkedIn CV Publications

Projects

The Point of View of a Sentiment: Towards Clinician Bias Detection in Clinical Notes – The aim of this work is to leverage large language models (LLMs) to characterize harmful language usage in psychiatry by identifying the sentiment expressed in sentences describing psychiatric patients based on the reader’s point of view (physician or non-physician).
- View on GitHub
- View preprint on LLM approach to quantifying clinician bias
Determining the Diagnostic Disparities of Schizophrenia in the Mount Sinai Health System, a Large Diverse Health System in NYC – The aim of this work is to characterize the diagnostic disparities within schizophrenia (SCZ) in a diverse set of patients using EHR from MSHS. In these analyses, we aimed to explore how measures of clinician bias and intersectional patient sociodemographic variables impact diagnosis.
- View preprint on SCZ diagnosis in NYC population
- View preprint on SCZ diagnosis in psychiatric patients in the emergency department taking into account measures of clinician bias