Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1504
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-15T22:02:39.322288Z
status: published
lens: g_space
sector: health
headline: Prospective multicentre expert evaluation of an artificial intelligence-based oncology clinical decision support system across major cancer types in India
dek: Introduction: Cancer is a major public health challenge in India, with an estimated 1.4 million new cases each year. Despite this burden, the ratio of oncologists to new cancer patients is critically low, about one specialist per 1600 new diagnoses per year. The AI-driven CDSS could help offset the oncologist shortage in LMICs. OneRx.AI is an RAG platform grounded in international and national oncology guidelines. We evaluated its Expert-rated accuracy, clinical safety, Guideline concordance and decision impact,…
gain_title: At three Indian tertiary cancer centres, oncologists using OneRx.AI reached clinical recommendations far faster, with high expert-rated accuracy, safety and guideline concordance.
problem_title: (none)
trace_subject: (none)
gain_reading: At three Indian tertiary cancer centres, oncologists using OneRx.AI reached clinical recommendations far faster, with high expert-rated accuracy, safety and guideline concordance.
gain_evidence: Expert-rated fully correct clinical accuracy was 82.7% | fully concordant guideline recommendations were 94.5%
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers tested OneRx.AI, a retrieval-augmented platform grounded in international and national oncology guidelines, at three tertiary oncology centres in India. Three oncologists independently reviewed de-identified clinical queries covering five cancer types, and 110 of 124 submitted queries were eligible.

India has very few oncologists relative to new cancer diagnoses, so tools that speed guideline-based decisions could ease specialist shortages. The authors describe the evaluation as exploratory, and results were affected by ceiling effects, so a larger follow-up study is needed before firm conclusions about real-world use.
limitation: The authors call the evaluation exploratory and ceiling-affected, and say it supports moving to a larger Stage II study rather than confirming performance.
tag: Evidence-backed gain
key_points: Of 124 submitted queries, 110 (88.7%) met the eligibility criteria. | Expert-rated clinical safety was 100%, and citation accuracy was 86.4%. | The decision-influence rate was 92.7%, and physician confidence rose by a mean of 1.69. | Inter-rater reliability for clinical accuracy was Gwet's AC1 = 0.707. | Median time to recommendation fell from 389.2 s to 47.2 s, an 87.9% reduction.
rundown: The study was a prospective, multicentre expert-validation design at three tertiary oncology centres. Three oncologist-clinicians rated queries on accuracy, safety, guideline concordance, decision impact, confidence and efficiency. Individual and common-prompt phases were analysed separately.

Co-primary endpoints used one-sample z-tests with 95% Wilson confidence intervals. Safety and guideline concordance ratings were uniform across evaluators, while clinical accuracy agreement was measured with a pre-specified statistic.

The motivating context is a shortfall of about one oncologist per 1600 new diagnoses per year, against roughly 1.4 million new cancer cases annually in India. The authors suggest AI decision support could help offset such shortages in low- and middle-income countries.
sources:
- peer_reviewed | International Journal of Medical Informatics | https://doi.org/10.1016/j.ijmedinf.2026.106739 | 2026-09-29
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

5b2ec9f78088c24d1962b4e4de1d534e81a9ad8cb016898e2f4215e093dff78f

previous

0000000000000000000000000000000000000000000000000000000000000000