📄 medicine

Patient Characteristics, Motivations, and Satisfaction in an AI-Assisted Rapid-Access Pathway for the Evaluation of Skin Lesions: A Cross-Sectional Survey Study

This cross-sectional survey of 254 patients in Paris demonstrates that an AI-assisted rapid-access pathway for skin lesion evaluation, which integrates primary care dermoscopy and immediate surgical management, is highly satisfactory to patients and fosters positive attitudes toward AI, particularly among those with higher educational levels.

Original authors: Laurène Masson, Raffaele Aguglia, Moshé Assouline, Theodoros Chrelias, Georgia Kanellopoulou, Haizam Oubari, Curtis L. Cetrulo, Alexandre G. Lellouch

Published 2026-06-29

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Laurène Masson, Raffaele Aguglia, Moshé Assouline, Theodoros Chrelias, Georgia Kanellopoulou, Haizam Oubari, Curtis L. Cetrulo, Alexandre G. Lellouch

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you have a mole on your arm that looks a little strange, or you just want to make sure your skin is healthy. In the past, finding a skin specialist (a dermatologist) to check it out was like trying to get a table at the most popular restaurant in town: you might wait three months, and even then, you might not get in.

This study looked at a different way of handling this problem, called the "Maison Abeille" pathway. Think of this pathway as a high-speed train station for skin checks, designed to get you from "I'm worried" to "It's fixed" without the long layovers.

Here is how the study explains this system and what the passengers (patients) thought about it.

The "High-Speed Train" Setup

Instead of waiting for a specialist, this center uses a three-step team approach:

The Scout: A general doctor (who is specially trained to use a high-powered magnifying glass called a dermoscope) takes a look at your skin.
The Digital Co-Pilot: The doctor uses an AI computer program (called Moleanalyzer Pro) to analyze the images. Think of this as a super-smart GPS that helps the doctor navigate the map of your skin to spot trouble spots.
The Fixer: If the Scout and the Co-Pilot think something looks risky, a plastic surgeon is right there in the same building. They can remove the spot immediately, right then and there. No need to go home and wait weeks for a second appointment.

Who Took the Train?

The researchers asked 254 people who used this service to fill out a survey. The group looked a lot like a crowd of early adopters at a tech conference:

Mostly Women: About 7 out of 10 people were women.
Highly Educated: Most had college degrees or higher.
At Risk: About 70% of them had reasons to worry about skin cancer (like having lots of moles, fair skin, or a family history).

Why Did They Choose This Route?

The study found that people didn't come here because they were obsessed with technology. They came because they were tired of the wait.

The Main Reason: 86% of people said they chose this place because it was much faster than waiting for a regular dermatologist.
The Goal: Nearly half came just for a routine check-up, while about a third came because they found a specific spot that worried them.

How Happy Were They?

The passengers were very happy with the ride.

Satisfaction: Over 90% said they were happy with how easy it was to get an appointment and with the consultation itself.
The "Fix-It-Now" Factor: The ability to have a suspicious spot removed on the same day was a huge bonus for many.

What Did They Think of the AI?

This is the most interesting part. People were surprisingly open to the computer helping the doctor.

Enthusiasm: About 68% said they were "enthusiastic" or "very enthusiastic" about the AI helping to check their skin.
The Education Link: The study found that people with higher levels of education were more likely to be excited about the AI. It's as if people who are used to complex technology felt more comfortable letting a computer be a "co-pilot" rather than a replacement for the human doctor.
The Reality Check: Even though people liked the AI, it wasn't the main reason they showed up. They showed up because they wanted to see a doctor today, not next month. The AI was just a helpful tool in the background.

The Bottom Line

The study concludes that this "fast-track" model works well for the people who tried it. It combines a trained doctor, a smart computer, and a surgeon who can act immediately.

However, the authors are careful to note that this study was done in a private clinic in a wealthy part of Paris with very educated patients. It's like testing a new sports car on a perfect race track; it runs great there, but we don't know yet if it will handle the bumpy roads of the general public or if it can be built cheaply enough for everyone.

In short: When skin cancer worries are high and specialist wait times are long, a system that mixes human doctors, AI helpers, and immediate surgery seems to make patients very happy and willing to trust the process. But we need more research to see if this works for everyone, everywhere.

Technical Summary: Patient Characteristics, Motivations, and Satisfaction in an AI-Assisted Rapid-Access Pathway for Skin Lesion Evaluation

Problem Statement
The study addresses the growing global incidence of skin cancers, including melanoma and keratinocyte carcinomas, which has outpaced the availability of dermatology specialists. In France, this imbalance is exacerbated by a declining density of dermatologists (dropping from >5 to ~3.4 per 100,000 inhabitants) and a marked rise in new cases, resulting in average consultation waiting times of approximately three months. These delays compromise early diagnosis and treatment. While alternative models integrating primary care, teledermatology, and artificial intelligence (AI) are emerging, there is a lack of data regarding the specific patient profiles, motivations, satisfaction levels, and acceptance of AI within these alternative pathways.

Methodology
The authors conducted a cross-sectional survey study at a private outpatient center in Paris, France, operating between June 2025 and January 2026. The center utilizes an integrated care model where a general practitioner (GP) trained in dermoscopy performs initial clinical and dermoscopic evaluations. Suspicious lesions are analyzed using an AI-assisted diagnostic support tool (Moleanalyzer Pro®). If indicated, a plastic surgeon performs immediate biopsy or excision during the same visit, bypassing the need for separate surgical referrals.

Participants: From a source population of patients consulting for skin lesion evaluation, a random sample was invited to complete an anonymous, self-administered online questionnaire.
Data Collection: The survey captured sociodemographic data, skin cancer risk factors, dermatologic health literacy, consultation motives, healthcare access history, satisfaction levels, and attitudes toward AI.
Analysis: Descriptive statistics summarized patient characteristics. Associations between categorical variables (e.g., education level vs. AI enthusiasm) were assessed using chi-square tests. Statistical significance was defined as $p < 0.05$ .

Key Results
Out of 4,235 invitations, 254 completed questionnaires (6.0% response rate) were analyzed.

Patient Demographics: The cohort was predominantly female (69.3%) and highly educated, with 71.0% holding a bachelor's degree or higher. The most represented age group was 40–59 years (46.1%).
Risk Profile: 69.7% of participants reported at least one skin cancer risk factor. The most common were having >50 nevi (40.2%) and fair phototype (33.5%).
Consultation Motives: Routine skin cancer screening was the primary motive (48.2%), followed by evaluation of a self-identified suspicious lesion (31.5%).
Drivers for Center Selection: 86.6% of patients chose the center specifically for shorter waiting times compared to conventional dermatology. The possibility of immediate lesion excision and the use of advanced imaging were also significant motivators.
Satisfaction: Satisfaction was high, with 95% satisfied with access to care and 91% satisfied with the consultation itself. Satisfaction levels correlated significantly with the duration of prior waiting times for dermatology appointments ( $p = 0.039$ ).
Attitudes Toward AI: 68.1% of patients expressed enthusiasm or strong enthusiasm for AI-assisted assessment, while only 7.1% expressed opposition. Enthusiasm was significantly associated with higher educational levels ( $p = 0.02$ ) but not with age or sex.

Key Contributions
The study provides empirical data on the "patient side" of AI-integrated dermatology workflows, a dimension previously under-described. It characterizes a specific patient profile (predominantly female, highly educated) that engages with rapid-access, AI-assisted pathways. The research highlights that while AI is a component of the service, the primary driver for patient adoption is the reduction of waiting times and the efficiency of same-day surgical management. Furthermore, it identifies a correlation between educational attainment and the acceptance of AI, suggesting that digital health literacy may influence the perception of AI as a decision-support tool rather than a replacement for clinical judgment.

Significance and Claims
The authors conclude that integrated care pathways combining primary care dermoscopy, AI-assisted evaluation, and rapid surgical management are highly acceptable to patients. They posit that such models represent a pragmatic strategy to improve access to skin lesion assessment in settings with limited dermatology availability.

The paper modestly claims that these findings support the viability of a tiered screening model:

Level 1: GPs trained in dermoscopy perform initial exams.
Level 2: AI systems assist in risk estimation and triage.
Level 3: Specialist expertise is reserved for complex cases via teledermatology or direct referral.

The authors emphasize that the value of AI in this context lies not in its standalone algorithmic performance but in its integration into coordinated care pathways that accelerate diagnosis and treatment. They acknowledge limitations, including the single-center setting in a socioeconomically privileged area, the low response rate (potential non-response bias), and the cross-sectional nature of the study, which precludes conclusions on diagnostic accuracy. Consequently, they call for larger multicenter and medico-economic studies to evaluate scalability and long-term impact.