Amar D. Mandavia

Dr. Amar D. Mandavia

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 June 2021 in Journal of Substance Abuse Treatment
Verified
Journal Article
Closed access
BHBrandy F. Henry
Brandy F. Henry
ACAimee Campbell
Aimee Campbell
THTimothy Hunt
Timothy Hunt
JJJessica K. Johnson
Jessica K. Johnson
AMAmar D Mandavia
Amar D Mandavia
NENabila El-Bassel
Nabila El-Bassel

5 September 2025 in Schizophrenia Bulletin
Verified
Journal Article
Closed access
EGEnrique Gutiérrez
Enrique Gutiérrez
CQCarlos Quesada
Carlos Quesada
EDEmily DeFraites
Emily DeFraites
DHDanielle J Harper
Danielle J Harper
AMAmar D Mandavia
Amar D Mandavia
Background and Hypothesis: Loose Associations (LA) in speech are key indicators of psychosis risk, notably in schizophrenia. Current detection methods are hampered by subjective evaluation, small samples, and poor generalizability. We hypothesize that combining Large Language Models (LLMs) with machine learning techniques could enhance objective identification of LA through improved semantic and probabilistic linguistic measures. Study Design: We propose a novel and reproducible workflow for generating synthetic conversational instances of LA using LLMs, guided by linguistic theory and validated through clinical expert review. This synthetic dataset forms the basis for model training and is complemented by an independently collected dataset for evaluation. Features extracted included traditional clause similarity measures alongside novel surprisal metrics quantifying semantic coherence and unexpected lexical shifts. A parsimonious and interpretable Light Gradient Boosting Machine model was trained using only four features. Study Results: The final model achieved high accuracy (83.46%; 95% CI: 82.96–83.95) on the synthetic dataset and robust performance on an independent set (82.36%; 95% CI: 81.94–82.78, AUC: 0.868). Our model outperformed baselines, including similarity-only models and prior thought disorder detection workflows. SHapley Additive exPlanations analysis confirmed the interpretability of the selected features, highlighting semantic coherence and word surprisal as key discriminators. Conclusions: Our approach demonstrates that LLM-derived linguistic features substantially enhance the objective, scalable detection of LA. The resulting model achieves high accuracy with minimal complexity, facilitating clinical applicability and interpretability. Future research should integrate additional lexical and contextual dimensions to further refine the identification of thought disorders, ultimately supporting early psychosis intervention.

19 March 2025 in Women's Health Issues
Verified
Journal Article
Closed access
AMAmar D Mandavia
Amar D Mandavia
Anne N BanducciAnne N Banducci
Anne N Banducci
TSTracy L. Simpson
Tracy L. Simpson
BMBrian P. Marx
Brian P. Marx
SHSage E. Hawn
Sage E. Hawn
NLNicholas Livingston
Nicholas Livingston

10 May 2025 in Social Science & Medicine
Verified
Journal Article
Closed access
ABAllison L. Baier
Allison L. Baier
YNYael I. Nillni
Yael I. Nillni
AMAmar D Mandavia
Amar D Mandavia
NLNicholas Livingston
Nicholas Livingston
TGTara Galovski
Tara Galovski

23 June 2025 in Drug and Alcohol Dependence
Verified
Journal Article
Closed access
AMAmar D Mandavia
Amar D Mandavia
JFJennifer R. Fonda
Jennifer R. Fonda
Anne N BanducciAnne N Banducci
Anne N Banducci
VAVictoria E. Ameral
Victoria E. Ameral
RHRebecca E. Sistad Hall
Rebecca E. Sistad Hall
NLNicholas Livingston
Nicholas Livingston

29 March 2024 in Social Science & Medicine
Verified
Journal Article
Closed access
JMJennifer H McQuaid
Jennifer H McQuaid
AMAmar D Mandavia
Amar D Mandavia
GCGalen Cassidy
Galen Cassidy
MSMichelle Alejandra Silva
Michelle Alejandra Silva
KEKaiz Esmail
Kaiz Esmail
KMKatherine McKenzie
Katherine McKenzie