Vedansh Thakkar

Mr. Vedansh Thakkar

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
15 May 2025 in PLOS ONE
Verified
Journal Article
Open access
PDF file
VTVedansh Thakkar
Vedansh Thakkar
GSGreg M. Silverman
Greg M. Silverman
AKAbhinab Kc
Abhinab Kc
NINicholas Elliott Ingraham
Nicholas Elliott Ingraham
EJEmma K. Jones
Emma K. Jones
CTChristopher J. Tignanelli
Christopher J. Tignanelli
Background: Patient symptoms, crucial for disease progression and diagnosis, are often captured in unstructured clinical notes. Large language models (LLMs) offer potential advantages in extracting patient symptoms compared to traditional rule-based information extraction (IE) systems. Methods: This study compared fine-tuned LLMs (LLaMA2-13B and LLaMA3-8B) against BioMedICUS, a rule-based IE system, for extracting symptoms related to acute and post-acute sequelae of SARS-CoV-2 from clinical notes. The study utilized three corpora: UMN-COVID, UMN-PASC, and N3C-COVID. Prevalence, keyword and fairness analyses were conducted to assess symptom distribution and model equity across demographics. Results: BioMedICUS outperformed fine-tuned LLMs in most cases. On the UMN PASC dataset, BioMedICUS achieved a macro-averaged F1-score of 0.70 for positive mention detection, compared to 0.66 for LLaMA2-13B and 0.62 for LLaMA3-8B. For the N3C COVID dataset, BioMedICUS scored 0.75, while LLaMA2-13B and LLaMA3-8B scored 0.53 and 0.68, respectively for positive mention detection. However, LLMs performed better in specific instances, such as detecting positive mentions of change in sleep in the UMN PASC dataset, where LLaMA2-13B (0.79) and LLaMA3-8B (0.65) outperformed BioMedICUS (0.60). For fairness analysis, BioMedICUS generally showed stronger performance across patient demographics. Keyword analysis using ANOVA on symptom distributions across all three corpora showed that both corpus (df = 2, p < 0.001) and symptom (df = 79, p < 0.001) have a statistically significant effect on log-transformed term frequency-inverse document frequency (TF-IDF) values such that corpus accounts for 52% of the variance in log_tfidf values and symptom accounts for 35%. Conclusion: While BioMedICUS generally outperformed the LLMs, the latter showed promising results in specific areas, particularly LLaMA3-8B, in identifying negative symptom mentions. However, both LLaMA models faced challenges in demographic fairness and generalizability. These findings underscore the need for diverse, high-quality training datasets and robust annotation processes to enhance LLMs’ performance and reliability in clinical applications.

20 September 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
VTVedansh Thakkar
Vedansh Thakkar
GSGreg Silverman
Greg Silverman
AKAbhinab Kc
Abhinab Kc
NINicholas Elliott Ingraham
Nicholas Elliott Ingraham
EJEmma Kate Jones
Emma Kate Jones
CTChristopher J. Tignanelli
Christopher J. Tignanelli
Patient symptoms play a critical role in disease progression and diagnosis, yet they are most often captured in unstructured clinical notes. This study explores use of large language models (LLMs) for extracting patient symptoms from clinical notes, comparing their performance against rule-based information extraction (IE) systems, like BioMedICUS. By fine-tuning an LLM on diverse corpora from multiple healthcare institutions, we aimed to improve symptom extraction accuracy and efficiency with symptoms related to acute and post-acute sequelae of SARS-CoV-2. We also conducted prevalence analysis to highlight significant differences in symptom prevalence across corpora and performed fairness analysis to assess the model's equity across race and gender. Our findings indicate that while LLMs can match the effectiveness of rule-based IE methods, they face significant challenges related to demographic bias and generalizability due to variability in training corpora. This evaluation revealed overfitting and insufficient generalization, especially when models were trained predominantly on limited datasets with single annotator bias. This study also revealed that LLMs offer substantial advantages in efficiency, adaptability, and scalability for biomedical IE, marking a transformative shift in clinical data extraction processes. These results provide real-world evidence of the necessity for diverse, high-quality training datasets and robust annotation processes to enhance LLM’s performance and reliability in clinical applications.

9 September 2026 in JAMIA Open
Verified
Journal Article
Open access
VTVedansh Thakkar
Vedansh Thakkar
GSGreg M Silverman
Greg M Silverman
MPMaura Putzer
Maura Putzer
AKAbhinab Kc
Abhinab Kc
SISana Ikramuddin
Sana Ikramuddin
CTChristopher Tignanelli
Christopher Tignanelli
Objectives: To produce the first structured longitudinal dataset of symptom severity trajectories in vagus nerve stimulation (VNS) patients using a 2-stage large language model pipeline for automated symptom severity extraction from clinical notes, as part of the NIH-funded research evaluating vagal excitations and anatomical linkages (U54AT012307). Materials and methods: We analyzed 1427 annotated clinical notes from 95 patients across 5 corpora. LLaMA models (3.2-1B, 3.2-3B, and 3.3-70B) were fine-tuned to classify 13 symptoms as present, absent, or negated (stage 1), then assessed severity (indeterminate, mild, moderate, or severe; stage 2). Longitudinal trajectories were tracked quarterly in 23 VNS patients against a 10-patient surgical control cohort (NSQIP). Results: Severity F1-scores ranged from 0.76 to 0.96, and symptom classification reached 0.99 for high-prevalence symptoms, with the validation corpus (VNS vs NSQIP) as the dominant factor (29.4% of variance; no significant source-corpus effect), supporting cross-domain generalizability. Macro F1-scores were lower than weighted F1-scores for smaller models and NSQIP; suicidal-ideation sensitivity was lower on NSQIP than VNS (0.80 vs 1.00, largest model), limiting generalization for psychiatric risk detection. Equity analyses were exploratory given small subgroups (gender precision 0.95). The longitudinal dataset revealed a median time to clinically meaningful seizure improvement of 0.5 quarters, with 50% responders, 25% non-responders, and 25% worsened. Post-surgical decline was absent in the unmatched NSQIP control cohort, consistent with, but not confirming, VNS specificity. Discussion: To our knowledge, this is the first system producing structured symptom trajectories from free-text VNS documentation, supporting individualized over population-level monitoring. Conclusion: This dataset supports future treatment-response prediction, personalized device programming, and a reproducible clinical natural langauge processing benchmark for neuromodulation research.