Research
Research behind TrialMind
Peer-reviewed methods behind TrialMind, grouped by the six types of team that use them.
Research areas
Each research area corresponds to a TrialMind solution page.
For biometrics and quantitative science teams
Biostatistics & Trial Design
Methods for designing more feasible trials, simulating outcomes, and creating analysis-ready evidence for statistical teams.
- Trial design and simulation
- Statistical programming
- Digital twins and synthetic controls
Representative papers
All 4AutoTrial · EMNLP 2023
Prompting Language Models for Clinical Trial Design
Wang, Xiao, Sun
Trial2Vec · Findings of EMNLP 2022
Zero-Shot Clinical Trial Document Similarity Search using Self-Supervision
Wang, Sun
TWIN · KDD 2023
Personalized Clinical Trial Digital Twin Generation
Das, Wang, Sun
Generative trial prediction · Patterns 2025
Clinical trial outcome prediction with multi-modal generative AI
Theodorou, Wang, Sun
For real-world evidence (RWE), epidemiology, and data science teams
Real-World Evidence
Reliable methods for cohort analysis, longitudinal patient modeling, comparative effectiveness, and safety studies on real-world data.
- Cohort feasibility
- Comparative effectiveness
- Treatment patterns and safety
Representative papers
All 9MediTab · IJCAI 2024
Scaling Medical Tabular Data Predictors via Data Consolidation, Enrichment, and Refinement
Wang et al.
TransTab · NeurIPS 2022
Learning Transferable Tabular Transformers Across Tables
Wang, Sun
Longitudinal EHR synthesis · Nature Communications 2023
Synthesize high-dimensional longitudinal electronic health records via hierarchical autoregressive language model
Theodorou, Xiao, Sun
Generative balancing · npj Digital Medicine 2025
Improving medical machine learning models with generative balancing for equity and excellence
Theodorou et al.
For strategy, search, and portfolio teams
Competitive & Portfolio Intelligence
Clinical-trial foundation models and predictive methods for landscape analysis, probability of success, and evidence-backed portfolio decisions.
- Trial landscape and key opinion leader (KOL) discovery
- Probability of technical success
- Portfolio and indication assessment
Representative papers
All 5TrialPanorama · npj Digital Medicine 2025
Benchmarking and Developing Large Language Models Using One Million Clinical Trials
Wang et al.
Panacea · medRxiv 2024
A foundation model for clinical trial search, summarization, design, and recruitment
Lin et al.
SPOT · ACM-BCB 2023
Sequential Predictive Modeling of Clinical Trial Outcome with Meta-Learning
Wang, Sun
HINT · Patterns 2022
Hierarchical Interaction Network for Clinical Trial Outcome Predictions
Wang, Sun
For clinical development and trial operations teams
Clinical Development & Operations
Research that improves site selection, patient-trial matching, recruitment, and the operational decisions that determine whether a trial finishes.
- Evidence-based site selection
- Patient matching and recruitment
- Enrollment and dropout prediction
Representative papers
All 4FRAMM · Patterns 2024
Fair ranking with missing modalities for clinical trial site selection
Theodorou et al.
TrialGPT · Nature Communications 2024
Matching patients to clinical trials with large language models
Jin, Wang et al.
COMPOSE · KDD 2020
Cross-modal pseudo-siamese network for patient-trial matching
Gao, Xiao, Glass, Sun
Doctor2Vec · AAAI 2020
Dynamic Doctor Representation Learning for Clinical Trial Recruitment
Gao, Xiao, Glass, Sun
For health economics and outcomes research (HEOR), evidence synthesis, and market access teams
Health Economics & Market Access
Human-AI evidence synthesis methods for systematic reviews, reproducible screening, and evidence bases that can stand up to health technology assessment (HTA) review.
- Systematic literature review
- Screening and data extraction reported to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses)
- Reusable evidence bases for HTA submissions
Representative papers
All 3DeepEvidence · Nature Machine Intelligence 2026
Empowering Biomedical Evidence Exploration and Synthesis with Deep Knowledge Graph Research
Wang et al.
TrialMind-SLR · npj Digital Medicine 2025
Accelerating clinical evidence synthesis with large language models
Wang et al.
LEADS · Nature Communications 2025
A foundation model for human-AI collaboration in medical literature mining
Wang et al.
For investigators, departments, and research informatics
Research Institutions
Foundational AI, including drug-discovery and multimodal methods, plus data-science agents that help investigators run reproducible biomedical studies.
- Investigator-led data science
- Multimodal biomedical research
- Reproducible methods and training
Representative papers
All 6DSWizard · Nature Biomedical Engineering 2026
Making large language models reliable data science programming copilots for biomedical research
Wang et al.
BIODSA-1K · Preprint 2025
Benchmarking Data Science Agents for Biomedical Research
Wang, Danek, Sun
BioBridge · ICLR 2024
Bridging Biomedical Foundation Models via Knowledge Graph
Wang et al.
MedCLIP · EMNLP 2022
Contrastive Learning from Unpaired Medical Images and Text
Wang, Sun
How we approach research
Research starts from a real decision
Each project starts from a decision a biometrics, real-world evidence, operations, competitive intelligence, health economics, or academic team has to make.
Performance is measured and published
Peer review, purpose-built benchmarks, and established baselines make performance inspectable.
Every output stays traceable
Sources, assumptions, and intermediate decisions stay attached to the output, so work can be reviewed and reproduced.
All publications
Every paper sits under its primary area, with keyword filters for methods that cross workflows.