Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1968
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-20T01:11:00.572231Z
status: published
lens: g_space
sector: health
headline: Comparison of an Artificial Intelligence Nutrition Platform With Dietitians in Providing Nutrition Support to Patients With Cancer
dek: Purpose Patients with cancer experience nutrition challenges, but access to registered dietitians (RDs) is limited. The purpose of this study was to determine whether a specialized Artificial Intelligence (AI) Nutrition Platform trained on curated high-quality evidence is equivalent in providing nutrition information to patients with cancer compared with oncology-credentialed RDs. Methods Twenty nutrition queries were developed by an expert panel, encompassing common cancer types, symptoms, and food allergies/pr…
gain_title: A specialized AI platform trained on curated evidence was rated equivalent to oncology-credentialed dietitians on cancer nutrition information, with better clarity, relevance, usefulness and readability in less time.
problem_title: (none)
trace_subject: (none)
gain_reading: A specialized AI platform trained on curated evidence was rated equivalent to oncology-credentialed dietitians on cancer nutrition information, with better clarity, relevance, usefulness and readability in less time.
gain_evidence: The AI Nutrition Platform met equivalence criteria with human dietitians | delivering responses with greater clarity, relevance, usefulness, and readability in less time
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers compared a specialized AI nutrition platform with four nationally recruited, oncology-credentialed dietitians. An expert panel wrote twenty queries spanning common cancer types, symptoms and food allergies or preferences, and both the platform and the dietitians answered them.

Three experienced reviewers scored the responses blind, and the platform met the preset equivalence criteria. The authors suggest it could widen access to nutrition support where dietitians are scarce, though the study tested written answers to prepared queries rather than care delivered to real patients.
limitation: The comparison used only twenty prepared queries and four dietitians, so it does not show how the platform performs in actual patient care.
tag: Evidence-backed gain
key_points: Twenty expert-panel queries covered common cancer types, symptoms, and food allergies or preferences. | Three reviewers with more than 10 years of cancer nutrition experience scored responses blind using a modified QAMAI instrument. | Mean total score difference was 0.9 points (90% CI, -0.27 to 2.15), and equivalence was tested with TOST. | Relevance scored 4.1 versus 3.9 and usefulness 3.8 versus 3.5, both favoring the AI platform. | AI responses had a Flesch-Kincaid grade level of 7.2 versus 9.8 for dietitian responses.
rundown: The motivation was limited access to registered dietitians for patients with cancer, who often face nutrition challenges. The platform was trained on curated high-quality evidence, and the study asked whether its information was equivalent to that from oncology-credentialed dietitians.

The primary outcome was the mQAMAI total score, which sums 1 to 5 ratings for accuracy, clarity, relevance, completeness and usefulness. Equivalence was assessed a priori with two one-sided tests, and secondary outcomes included readability and response time.

The reported results favored the AI platform on relevance, usefulness and readability, and the abstract also reports a shorter response time. The conclusion is that the tool could broaden access in settings with limited dietitian availability.
sources:
- peer_reviewed | JCO Clinical Cancer Informatics | https://doi.org/10.1200/cci-25-00337 | 2026-10-15
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

0d953fa42d136cd3834176e3c73cabd1b8f71c27488a7795d9c462d4ea87d379

previous

0000000000000000000000000000000000000000000000000000000000000000