Computational framework for structuring and analyzing clinical trial criteria for AI-guided fine-grained matching
Open accessAbstract
While artificial intelligence (AI) has demonstrated potential in automating clinical trial matching, most existing solutions rely on high-level structured data or oversimplified criteria. This study introduces a framework to structure and analyze eligibility criteria across three real-world trial protocols, aiming to inform more granular AI-driven trial matching strategies. Trial criteria from three protocols were decomposed into individual variables and evaluated based on data type, scope, and dependency. Complexity was assessed using a novel formula incorporating the number of independent and dependent variables, alongside the Flesch-Kincaid reading grade level. Quantitative analysis explored variation across trials. Protocols contained between 22-160 eligibility variables, with 4-22% showing interdependence. Reading grade levels ranged from sixth grade to first-year college. Complexity scores varied significantly, with some trials exhibiting particularly high cognitive and logical burdens. Recursive and hierarchical structures were prevalent in high-complexity protocols. This study reveals the substantial variability and structural complexity of clinical trial criteria, highlighting challenges for AI matching systems. A standardized approach to measuring trial complexity can enhance algorithm transparency, scalability, and interpretability. These findings underscore the need for structured, computable frameworks to improve equity and efficiency in clinical trial recruitment.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
TRIALSCOPE demonstrates oncology-scale text normalization and ontology alignment.
“Pipelines such as TRIALSCOPE demonstrate oncology-scale text normalization and ontology alignment.13”
From Transforming oncology clinical trial matching through neuro-symbolic, multi-agent AI and an oncology-specific knowledge graph: a prospective evaluation in 3804 patients (2026) · cites this paper as [13]
The Methods section describes structuring eligibility criteria from three trial protocols, including one oncology trial. Searches for TRIALSCOPE and ontology alignment found no passage describing that pipeline or demonstrating oncology-scale text normalization and ontology alignment. The paper’s discussion of oncology criteria does not establish the attributed claim. (LLM, conf=0.76)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Innovation in the pharmaceutical industry: New estimates of R&D costs
DiMasi, J. A., Grabowski, H. G., & Hansen, R. W. (2016) · Journal of Health Economics · doi:10.1016/j.jhealeco.2016.01.012 · Cited by 3,359 · see citing works
Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic review
Ford, J. G., Howerton, M. W., Lai, G. Y., Gary, T. L., Bolen, S., Gibbons, M. C., Tilburt, J., Baffi, C., Tanpitukpongse, T. P., Wilson, R. F., et al. (2008) · Cancer · doi:10.1002/cncr.23157 · Cited by 1,083 · see citing works
Corbie-Smith, G. M., Thomas, S. B., & George, D. M. M. S. (2002) · Archives of Internal Medicine · doi:10.1001/archinte.162.21.2458 · Cited by 1,067 · see citing works
The Role of Clinical Trial Participation in Cancer Research: Barriers, Evidence, and Strategies
Unger, J. M., Cook, E., Tai, E., & Bleyer, A. (2016) · American Society of Clinical Oncology Educational Book · doi:10.1200/edbk_156686 · Cited by 665 · see citing works
The TRIPOD-LLM reporting guideline for studies using large language models
Gallifant, J., Afshar, M., Ameen, S., Aphinyanaphongs, Y., Chen, S., Cacciamani, G., Demner-Fushman, D., Dligach, D., Daneshjou, R., Fernandes, C., et al. (2025) · Nature Medicine · doi:10.1038/s41591-024-03425-5 · Cited by 615 · see citing works
Unger, J. M., Vaidya, R., Hershman, D. L., Minasian, L. M., & Fleury, M. E. (2018) · JNCI Journal of the National Cancer Institute · doi:10.1093/jnci/djy221 · Cited by 578 · see citing works
Carlisle, B., Kimmelman, J., Ramsay, T., & MacKinnon, N. (2015) · Clinical Trials · doi:10.1177/1740774514558307 · Cited by 367 · see citing works
Matching patients to clinical trials with large language models
Jin, Q., Wang, Z., Floudas, C. S., Chen, F., Gong, C., Bracken-Clarke, D., Xue, E., Yang, Y., Sun, J., & Lu, Z. (2024) · Nature Communications · doi:10.1038/s41467-024-53081-z · Cited by 287 · see citing works
Clinical Trials in Rare Disease
Augustine, E. F., Adams, H., & Mink, J. W. (2013) · Journal of Child Neurology · doi:10.1177/0883073813495959 · Cited by 223 · see citing works
Briel, M., Elger, B. S., McLennan, S., Schandelmaier, S., von Elm, E. B., & Satalkar, P. P. (2021) · Trials · doi:10.1186/s13063-021-05818-0 · Cited by 113 · see citing works
Schroen, A. T., Petroni, G. R., Wang, H., Gray, R. J., Wang, X. F., Cronin, W. M., Sargent, D. J., Benedetti, J., Wickerham, D. L., Djulbegović, B., et al. (2010) · Clinical Trials · doi:10.1177/1740774510374973 · Cited by 100 · see citing works
Effort Required in Eligibility Screening for Clinical Trials
Penberthy, L. T., Dahman, B. A., Petkov, V. I., & DeShazo, J. P. (2012) · Journal of Oncology Practice · doi:10.1200/jop.2012.000646 · Cited by 96 · see citing works
Williams, C., Senft, N., Shelburne, N., & Norton, W. E. (2021) · JAMA Network Open · doi:10.1001/jamanetworkopen.2021.27792 · Cited by 95 · see citing works
Alexander, M., Solomon, B., Ball, D., Sheerin, M., Dankwa‐Mullan, I., Preininger, A. M., Jackson, G. P., & Herath, D. (2020) · JAMIA Open · doi:10.1093/jamiaopen/ooaa002 · Cited by 83 · see citing works
Zero-Shot Clinical Trial Patient Matching with LLMs
Wornow, M., Lozano, A., Dash, D., Jindal, J., Mahaffey, K. W., & Shah, N. H. (2025) · NEJM AI · doi:10.1056/aics2400360 · Cited by 59 · see citing works
Kumar, G., Chaudhary, P., Quinn, A., & Su, D. (2022) · Contemporary Clinical Trials Communications · doi:10.1016/j.conctc.2022.100939 · Cited by 51 · see citing works
Ünlü, O., Varugheese, M., Shin, J., Subramaniam, S., Stein, D. W. J., Laurent, J. J. S., Mailly, C. J., McPartlin, M., Wang, F., Oates, M., et al. (2025) · JAMA · doi:10.1001/jama.2024.28047 · Cited by 41 · see citing works
Combining human and machine intelligence for clinical trial eligibility querying
Fang, Y., Idnay, B., Sun, Y., Liu, H., Chen, Z., Marder, K. S., Xu, H., Schnall, R., & Weng, C. (2022) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocac051 · Cited by 29 · see citing works
Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow
Jain, N. M., Culley, A., Knoop, T., Micheel, C., Osterman, T. J., & Levy, M. A. (2019) · JCO Clinical Cancer Informatics · doi:10.1200/cci.19.00033 · Cited by 27 · see citing works
Self-citationLearnings From Precision Clinical Trial Matching for Oncology Patients Who Received NGS Testing
Jain, N. M., Culley, A., Micheel, C., Osterman, T. J., & Levy, M. A. (2021) · JCO Clinical Cancer Informatics · doi:10.1200/cci.20.00142 · Cited by 15 · see citing works
Self-citationSun, Y., Butler, A. M., Diallo, I., Kim, J. H., Ta, C., Rogers, J. R., Liu, H., & Weng, C. (2021) · Applied Clinical Informatics · doi:10.1055/s-0041-1733846 · Cited by 6 · see citing works
Jain, N. M., Holt, M. E., Micheel, C., & Levy, M. A. (2021) · JCO Clinical Cancer Informatics · doi:10.1200/cci.21.00082 · Cited by 2 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Loaiza-Bonilla, A., Yost, C., Kurnaz, S., Tuysuz, E., Thaker, N. G., Giritlioglu, D., & Noel Meza, J. P. (2026) · ESMO Real World Data and Digital Oncology · doi:10.1016/j.esmorw.2026.100706 · Cited by 8 · see citing works
“Prior systems have demonstrated that LLMs, retrieval, and ontology alignment can help identify candidate trials, 3,4,11-13 but many published approaches remain retrospective, disease-limited, or optimized for text similarity and retrieval rather than auditable patient — trial execution over…”
Huang, G., Li, L.-J., Li, P.-S., Fan, Z., & Lu, B.-Q. (2026) · European Archives of Oto-Rhino-Laryngology · doi:10.1007/s00405-026-10133-y · Cited by 3 · see citing works
Blom, J. M. C., & Pani, L. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02987-7 · Cited by 2 · see citing works
- Authors
- Habib, D. R. S., Mahajan, I., Evancha, B., Micheel, C., & Fabbri, D.
- Affiliations
- Vanderbilt University (US)
- Corresponding author
- Fabbri, Daniel
- Venue
- Journal of Medical Systems
- Published
- 2025-11-22
- Volume
- 49
- Issue
- 1
- Pages
- 168-168
- ISSN
- 0148-5598 (ISSN-L)
- 1573-689X
- PMC
- PMC12640310
- OpenAlex
- W4416552902
- Cited by
- 3
- References
- 22
- Citations by year
- 2026: 3
- FWCI
- 0.72
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Locations
- Journal of Medical Systems (publishedVersion, cc-by)
- Research Square (acceptedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Ethics in Clinical Research
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- inclusion and exclusion criteria, interdependent variables, Flesch-Kincaid readability, clinical trial recruitment, algorithmic transparency
- MeSH terms
- Algorithms, Artificial Intelligence, Clinical Trials as Topic, Humans, Patient Selection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
