Evaluation of an artificial intelligence clinical trial matching system in Australian lung cancer patients
Problem
Matching lung cancer patients to eligible clinical trials requires evaluating thousands of patient attributes against complex trial eligibility criteria, which is slow to do manually.
Why it matters
Trial eligibility matching affects patient access to experimental cancer treatments.
What the AI did
An AI-based clinical trial matching system evaluated 7252 patient attributes against 11,467 trial eligibility criteria across ten phase I-III lung cancer trials, compared to a clinician-consensus gold standard.
Results
System evaluated 7252 patient attributes (median 74 per patient) against 11,467 criteria (median 597 per trial); median query time 15.5 s (range 7.2-37.8); accuracy vs gold standard not stated in source.
Impact
Peer-reviewed evaluation of AI trial matching feasibility in 102 lung cancer patients at an Australian cancer hospital.
Record
| Venue | JAMIA Open |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Marliese Alexander, Benjamin J. Solomon, David Ball, M. Sheerin, Irene Dankwa‐Mullan, Anita M. Preininger, Gretchen Purcell Jackson, Dishan Herath |
| Citations | 85 (OpenAlex, at import time) |
| References | 23 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1093/jamiaopen/ooaa002 |
| Open access | gold |
| Keywords | clinical trial matching, lung cancer, AI evaluation, eligibility criteria |
Importance rating
significant — thousands to millions affected
Trial matching affects lung cancer patients' access to experimental treatments; single-hospital evaluation of 102 patients; broader scale not stated.