Are Patients Receptive to AI Use for Skin Cancer Screening?

The use of artificial intelligence (AI) in health care is becoming more widespread, but how patients feel about it is a question up for debate. A new study found that when it comes to skin cancer screening, patients may be receptive to its use.

This was a qualitative study that used a grounded theory approach to semistructured interview analysis. A total of 48 patients were included in the study, which was conducted in general dermatology clinics at the Brigham and Women’s Hospital and melanoma clinics at the Dana-Farber Cancer Institute. Two researchers conducted independent interviews regarding skin cancer screening with interrater reliability measurement. Code frequency was analyzed using reconciled codes. The primary outcome measures were AI intelligence concept, perceived benefits and risks of AI, strengths and weaknesses of AI, AI implementation, response to conflict between human and AI clinical decision-making, and recommendation for or against AI.

Is AI the Future of Skin Cancer Screening?

Just over half of patients (n=26, 54%) were female, and the mean age (SD) was 53.3 (21.7) years. A third of patients (n=16, 33%) had a history of melanoma, another third had a history of nonmelanoma skin cancer only, and the remaining third had no skin cancer history. Half of patients (n=24) were interviewed about a direct-to-patient AI tool, and the other half were asked about a clinician decision-support AI tool. The two coding teams had interrater reliability ratings of κ = 0.94 and κ = 0.89. Patients tended to think of AI of how it pertained to cognition; the main benefits they reported were increased diagnostic speed (n=29, 60%) and health care access (n=29, 60%), while the most common risk was increased patient anxiety (n=19, 40%). The greatest perceived strength of AI was more accurate diagnosis (n=33, 69%), while the greatest perceived weakness was less accurate diagnosis (n=41, 85%). The overarching theme observed in the interviews was the significance of symbiosis between humans and AI (n=45, 94%). The primary response to conflict between human-AI clinical decision-marking was seeking biopsy (n=32, 67%). Three-quarters of patients (n=36) said they would recommend AI to family and friends.

The study was published in JAMA Dermatology.

“In this qualitative study, patients appeared to be receptive to the use of AI for skin cancer screening if implemented in a manner that preserves the integrity of the human physician-patient relationship,” concluded the study authors.