MétaCan
Menu
← Back to cohort
Record W7164835344 · doi:10.2196/81829

Frontline Health Workers’ Perspectives of the WHO skin NTD app in Kenya – a Qualitative Study on AI-embedded mHealth Implementation (Preprint)

2025· article· en· W7164835344 on OpenAlexvenueno aff
Emily E. V. Quilter, Ruth Nyangacha, Esther Kinyeru, Aïna Fuster Casanovas, Amberly Brigden, Kenton O'Hara, Diana Atieno, José A Ruiz-Postigo, Carme Carrión

Bibliographic record

VenueJMIR mhealth and uhealth · 2025
Typearticle
Languageen
FieldHealth Professions
TopicMobile Health and mHealth Applications
Canadian institutionsnot available
Fundersnot available
KeywordsmHealthQualitative researchDigital healthTelemedicineHealth careFocus group

Abstract

fetched live from OpenAlex

Background: Skin neglected tropical diseases (NTDs) pose significant diagnostic and management challenges in resource-limited settings due to constrained dermatological expertise, frontline health worker (FHW) training, and limited access to diagnostic resources. Mobile health apps with artificial intelligence (AI)-enabled diagnostic imaging capabilities have the potential to enhance clinical decision-making and professional development at the primary care level. The World Health Organization (WHO) skin NTD mobile app uses convolutional neural networks to analyze images of skin lesions and generate differential diagnoses, intended to be used alongside clinical history and examination, to support FHWs in identifying 12 skin NTDs and 24 common skin conditions. Beyond clinical decision support, the app also aims to upskill FHWs in the recognition and management of these diseases. However, the success of such tools depends on understanding users' needs and the realities of implementation in diverse clinical contexts. Objective: This study aimed to explore FHWs' perspectives on the real-world use and impact of the AI-embedded WHO Skin NTDs app on diagnostic workflows, dermatological understanding, clinical decision-making, and FHW-patient interactions across diverse health care delivery settings in Kenya. Methods: This qualitative study involved 36 FHWs from 5 skin NTD-endemic counties in Kenya. Following a training workshop, FHWs integrated the app into routine clinical workflows from June to October 2024. Data were collected through 15 semistructured interviews (each 30-45 minutes) and 4 focus group discussions (1-1.5 hours) exploring FHW experiences across diverse health care delivery contexts. All sessions were audio-recorded, transcribed verbatim, and thematically analyzed using NVivo (QSR International), using a bottom-up inductive coding approach. Results: FHWs reported that the app facilitated a shift from habitual referral to more proactive case management at the local-level facility, reinforcing clinical ownership and positioning them as local dermatology reference points. It was perceived to enhance diagnostic confidence, strengthen patient trust, and encourage community engagement. Some FHWs described how the app helped mitigate situations for patient stigma due to decreased reliance on public colleague consultations. However, technical limitations (eg, internet dependency and algorithmic errors) constrained consistent use. While most FHWs used the app in line with its intended role as an assistive tool, a minority reported situations of diagnostic deferral to the AI output, highlighting potential considerations of clinical autonomy. Conclusions: The WHO Skin NTDs app shows strong potential to strengthen frontline dermatological capacity that aligns with WHO strategies to decentralize NTD care and promote "skin health for all." Our findings underscore the importance of embedding such tools within ethical and pedagogical frameworks that protect clinical autonomy and foster sustainable capacity building. Further research will examine real-world use in situ to guide context-specific governance, ensuring that this AI-embedded tool enhances-rather than displaces-clinical reasoning and epistemic authority.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Direct model labels (unvalidated)

Per-model category and study-design labels from the labeling rounds. They are machine output, unvalidated, and the disagreement between models ships as data. No study design here is MEDLINE-validated yet.

Model armCategoriesStudy designConfidence
gemmano category
Domain: not available · Genre: Empirical
About the Canadian research system: no · About a Canadian topic: no
Qualitativelow
gptno category
Domain: not available · Genre: Empirical
About the Canadian research system: no · About a Canadian topic: no
Qualitativehigh
models agreeAgreement compares identical category sets and study designs across arms.

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.011
metaresearch head score (Gemma)0.011
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Qualitative · Consensus signal: Qualitative
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.032
Threshold uncertainty score0.064

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0110.011
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0010.001
Science and technology studies0.0160.006
Scholarly communication0.0050.005
Open science0.0010.005
Research integrity0.0030.005
Insufficient payload (model declined to judge)0.0060.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.068
GPT teacher head0.546
Teacher spread0.478 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Labeled directly by 2 models reading the full record.

The models applied no category: nothing in the taxonomy fit this work.
Study designQualitative
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractno

Explore more

Same venueJMIR mhealth and uhealth→Same topicMobile Health and mHealth Applications→French-language works237,207→