Georgia Tech computational scientist advancing mental health analytics, clinical NLP, and AI ethics in healthcare
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
64
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Insufficient data to score this axis. Output measures publication volume and productivity relative to field norms — total and per-year work counts. The underlying data was not assembled in this snapshot, so no score is estimated.
Impact—
Insufficient data to score this axis. Impact measures citation-based influence — total citations, field-normalized citation rates, and downstream reference reach. These indicators were not available for computation here, so no score is estimated.
Focus100/100
Focus scores a perfect 100, reflecting tightly concentrated topics. 'Mental Health via Writing' alone carries 116 works — more than double the next theme, 'Sentiment Analysis and Opinion Mining' (47) — while 'Misinformation and Its Impacts' (42), 'Ethics and Social Impacts of AI' (36), and 'Digital Mental Health Interventions' (34) form an adjacent cluster around mental health, language, and responsible AI.
Collaboration62/100
The collaboration score is 62. The snapshot records 21 distinct recent coauthors, and the catalog points to a recurring core group including Aiwei Ivy Zhang, Nimra Ishfaq, Mohit Chandra, Santiago Alvarez Lesmes, and Adam Coscia. On CliniCIRCA the researcher holds the last, typically senior, author position, and a middle position on the 2026 oncology workflow — a hands-on, mid-sized network rather than an expansive multi-institution consortium.
Activity100/100
Activity scores a perfect 100. The countsByYear series shows sustained and accelerating production: 22 works in 2019, 32 in 2021, 30 in 2023, 29 in 2024, 33 in 2025, and 26 already recorded in 2026, with 88 works across the last three years. This multi-decade cadence, traceable to 2007, sits in the platform's top band.
Reproducibility55/100
Reproducibility scores 55, computed as a mean readiness of 55 across 1 reviewed paper. CliniCIRCA is rated medium: Evolve's verdict describes a clinician-validated, honest LLM pipeline over MIMIC-III psychiatric notes, but flags that the modular design is never ablated — leaving each component's contribution unverified and tempering an otherwise positive readiness picture.
Research summary
Munmun De Choudhury is a researcher at the Georgia Institute of Technology whose profile is name-resolved from OpenAlex bibliographic data. The snapshot records an h-index of 64 and an i10-index of 168, with 88 works in the most recent three-year window and steadily rising annual output — from 22 works in 2019 to 33 in 2025 — underpinning perfect activity and focus scores on the Evolve radar. Topic assignment is strongly concentrated: 'Mental Health via Writing' accounts for 116 works, more than double 'Sentiment Analysis and Opinion Mining' (47), with 'Misinformation and Its Impacts' (42), 'Ethics and Social Impacts of AI' (36), and 'Digital Mental Health Interventions' (34) completing a coherent agenda on language, mental health, and responsible AI. Recent cataloged work translates this agenda into clinical settings. CliniCIRCA (2026), a modular LLM framework building longitudinal mental health patient journeys from raw MIMIC-III narratives, was reviewed by Evolve with medium reproducibility readiness — clinician-validated and honest, but without ablation of its modular design — and lists the researcher as last author. A 2026 evidence-linked multi-agent workflow for oncology feature extraction is also cataloged, though not yet reviewed. Citation-volume and venue-normalized impact axes remain uncomputed in this profile.
Collaboration pattern
Recent work involves 21 distinct coauthors, with a recurring core group including Aiwei Ivy Zhang, Nimra Ishfaq, Mohit Chandra, Santiago Alvarez Lesmes, and Adam Coscia. Authorship positions in the catalog — last author on CliniCIRCA (2026) and a middle position on the 2026 oncology workflow — suggest a senior, supervisory role within a moderately sized team. The collaboration radar score of 62 reflects a substantial but not maximal network footprint.
Research keywords
Highlights
CliniCIRCA: LLM-built longitudinal patient journeys from EHR narratives
Evidence-linked multi-agent workflow for oncology feature extraction
A 2026 cataloged paper, 'From Analytics to Tumor Boards', applies a multi-agent workflow with evidence-linked outputs to oncology feature extraction, with the researcher in a middle author position. Evolve has not yet published a review verdict, keywords, or reproducibility rating for it.
Two decades concentrated on mental health and language
Topic assignment across the OpenAlex record concentrates 116 works on 'Mental Health via Writing' — more than double any other theme — with sentiment analysis and opinion mining (47), misinformation (42), AI ethics (36), and digital mental health interventions (34) completing a tightly focused agenda.
Timeline
- 2007Earliest year in the countsByYear record, with 1 work — the start of a publication history now spanning two decades.
- 2013Output begins to climb: 8 works recorded in 2013, rising to 10 in 2014.
- 2019A jump to 22 works — nearly triple 2018's 8 — marks the start of a sustained high-output phase.
- 202132 works recorded, with output holding at a high level through 2022 (25), 2023 (30), and 2024 (29).
- 202533 works — the highest annual count in the record.
- 2026CliniCIRCA is cataloged and reviewed by Evolve with medium reproducibility readiness, listing the researcher as last author; 26 works are already recorded this year.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric indicators (h-index, i10-index, countsByYear, topics, coauthor counts) are derived from OpenAlex only; identity is name-resolved from public bibliographic data and may not capture every work.
- Output and impact radar dimensions are uncomputed; no citation counts beyond the h-index and i10-index were available.
- Only two papers, both from 2026, are in the Evolve review catalog, so reviewed-paper evidence covers only a slice of the recent record.
- The reproducibility score rests on a single reviewed paper (reproPapers: 1) and should be read as indicative rather than stable.