TRV-2026-1466Version 1 · Certified

Written 2026-10-15 18:13:33 UTC · current record

Reason for this version

Certified into the record

Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1466
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-15T18:13:33.707113Z
status: published
lens: p_space
sector: health
headline: Contribution of artificial intelligence research to the development of inclusive applications for people with disabilities: A scoping review (2015–2025)
dek: Author summary People with disabilities (PWD) are frequently underrepresented in the design, development, implementation, and evaluation of artificial intelligence (AI) systems, which can limit the accessibility, usability, and equity of these technologies, and result failing to address their diverse needs and lived experiences. It is therefore essential to evaluate the extent to which PWD are meaningfully involved across all stages of AI development and research. In this scoping review, we examined their roles…
gain_title: (none)
problem_title: People with disabilities are often underrepresented across the design, development and evaluation of healthcare AI, which can limit accessibility, usability and equity.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: People with disabilities are often underrepresented across the design, development and evaluation of healthcare AI, which can limit accessibility, usability and equity.
problem_evidence: People with disabilities (PWD) are frequently underrepresented in the design, development, implementation, and evaluation of artificial intelligence (AI) systems
quick_read: A scoping review covering 2015 to 2025 looked at how healthcare AI research contributes to inclusive, accessible technologies for people with disabilities. It identified studies on many disabilities and AI uses, including predictive models, assistive technologies, early screening tools and decision-support systems.

The findings matter because disabled people make up 16% of the global population, yet the review describes them as disproportionately excluded from AI advances in healthcare. Recognition of inclusive development is growing, but participation and accessibility are handled unevenly, and the supplied excerpt cuts off before further detail.
limitation: 
tag: Evidence-backed problem
key_points: The review examined the roles of people with disabilities from multiple perspectives across AI development and research. | Studies covered a wide range of disabilities and AI applications, including predictive models and assistive technologies. | Early screening tools and decision-support systems were among the applications identified. | The authors report growing recognition of inclusive AI development. | Meaningful participation, accessibility, transparency and context-sensitive implementation are described as inconsistently addressed.
rundown: The authors frame the question as whether people with disabilities are meaningfully involved at every stage of AI development and research, from design through evaluation. Underrepresentation is said to limit accessibility, usability and equity, and to leave their diverse needs and lived experiences unaddressed.

The review focuses on healthcare AI research and its contribution to inclusive and accessible technologies. It reports gaps in the current literature and offers recommendations toward more equitable, accessible and person-centered AI systems for people with disabilities.
sources:
- journalism | PLOS (Public Library of Science) | https://journals.plos.org/digitalhealth/article?id=10.1371%2Fjournal.pdig.0001749 | 2026-10-15
prev: 0000000000000000000000000000000000000000000000000000000000000000
sha256
4617fa331b9067d0270f1f0d2b05bbe175a8116afe14bf2eb7f9a4a3a8d967f9
previous
0000000000000000000000000000000000000000000000000000000000000000
Verify this record
How to verify without trusting this page

Fetch the canonical text of any version from /api/record/TRV-2026-1466 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.