Artificial intelligence in mobile health for skin cancer diagnostics at home (AIM HIGH): a pilot feasibility study

Abstract

Background: Artificial intelligence (AI)-based mobile phone apps (mHealth) have the potential to streamline care for suspicious skin lesions in primary care. This study aims to investigate the conditions and feasibility of a study that incorporates an AI-based app in primary care and evaluates its potential impact.

Methods: We conducted a pilot feasibility study from November 22nd, 2021 to June 9th, 2022 with a mixed-methods design on implementation of an AI-based mHealth app for skin cancer detection in three primary care practices in the Netherlands (Rotterdam, Leiden and Katwijk). The primary outcome was the inclusion and successful participation rate of patients and general practitioners (GPs). Secondary outcomes were the reasons, facilitators and barriers for successful participation and the potential impact in both pathways for future sample size calculations. Patients were offered use of an AI-based mHealth app before consulting their GP. GPs assessed the patients blinded and then unblinded to the app. Qualitative data included observations and audio-diaries from patients and GPs and focus-groups and interviews with GPs and GP assistants.

Findings: Fifty patients were included with a median age of 52 years (IQR 33.5-60.3), 64% were female, and 90% had a light skin type. The average patient inclusion rate was 4-6 per GP practice per month and 84% (n = 42) successfully participated. Similarly, in 90% (n = 45 patients) the GPs also successfully completed the study. GPs never changed their working diagnosis, but did change their treatment plan (n = 5) based on the app's assessments. Notably, 54% of patients with a benign skin lesion and low risk rating, indicated that they would be reassured and cancel their GP visit with these results (p < 0.001).

Interpretation: Our findings suggest that studying implementation of an AI-based mHealth app for detection of skin cancer in the hands of patients or as a diagnostic tool used by GPs in primary care appears feasible. Preliminary results indicate potential to further investigate both intended use settings.

Funding: SkinVision B.V.

Keywords: Artificial intelligence; Convolutional neural network; General practitioners; Mobile health; Primary care; Skin cancer.

Fig. 1

Flowchart of the study inclusion rate of patients with suspicious skin lesions consulting one of the three GP practices between November 2021 and June 2022. The patient trajectory from inclusion until usage of the app and completion of the questionnaire was defined as phase one. Phase two was defined as participation by the GP, which was the consult with the patient and filling in of the GP specific questions. Criteria for successful or unsuccessful participation are defined in Supplemental Table S1. Abbreviations: GP; general practitioner.

Fig. 2

Flowchart describing the risk classification by the general practitioner (GP) and the app in comparison to the actual final diagnosis of the lesion according to the gold standard. Abbreviations: GP; General practitioner, SK; Seborrheic keratosis, BU; Benign unspecified, DF; Dermatofibroma, LK; Lichen planus-like keratosis, AN; Atypical nevus, AK; Actinic keratosis, BCC; Basal cell carcinoma, MM; Malignant melanoma.

Fig. 3

Factors related to (un)successful usage of the app. Abbreviation: GP; General practitioner.