Patients may be receptive to skin cancer screening via artificial intelligence

Three-quarters of patients would recommend artificial intelligence as a component of clinical decision-making for skin cancer, according to a survey.

“The use of artificial intelligence (AI) is expanding throughout the field of medicine,” Caroline A. Nelson, MD, of the department of dermatology at Yale School of Medicine, and colleagues wrote, adding that researchers are investigating the use of AI in classifying skin lesions. “Although AI is poised to change how patients engage in health care, patient perspectives remain poorly understood.”

The researchers conducted semi-structured interviews with 48 patients in general dermatology clinics at Brigham and Women’s Hospital and melanoma clinics at Dana-Farber Cancer Institute to determine how patients think about the risks, benefits, strengths and weakness of AI as it pertains to skin cancer screening. They also aimed to determine how patients feel about the differences between human and AI clinical decision-making.

Participants were interviewed between May 6 and July 8, 2019. The cohort was 54% women with a mean age of 53.3 years. One-third of the cohort had a history of melanoma, one-third had a history of non-melanoma skin cancer, and one-third had no history of skin cancer.

For half of the cohort, the content of the interview pertained to a direct-to-patient AI tool, while the other half were queried about a clinician decision-support AI tool.

Results showed that 60% of participants perceived increased diagnostic speed to be a benefit of AI, while 60% also felt AI could increase health care access. Conversely, 40% suggested that increased patient anxiety may constitute the biggest risk of AI.

While 69% of patients perceived AI to offer a more accurate diagnosis, 85% suggested that AI would provide a less accurate diagnosis.

Symbiosis between humans and AI was a theme that emerged in 94% of interviews.

In discussing conflict between human and AI clinical decision-making, 67% of patients suggested that a biopsy was the best way to resolve the issue.

Seventy-five percent of the cohort reported that they would recommend AI to family or friends.

“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,” the researchers said.

In an accompanying editorial, Carrie L. Kovarik, MD, of the department of dermatology at the Perelman School of Medicine at the University of Pennsylvania, suggested that the paper gives valuable information about patient preferences with regard to AI, albeit for a small subset of patients.

“Vulnerable patients, including racial and ethnic minorities, the underinsured or uninsured, economically disadvantaged, and those with chronic health conditions, may be at risk for improper consent for or use of AI,” Kovarik wrote. “Patients who better understand privacy protection regulations and privacy preferences in digital devices may be less concerned about the loss of privacy than those who lack the knowledge and tools to manage these options.”

For Kovarik, privacy is a key consideration for the use of AI. “People who are healthy may have a smaller amount of personal health information and less need to benefit from shared information, so they may not see the value in sharing their personal health information for the development of AI,” she wrote. “Patients with certain medical issues or difficulty obtaining medical coverage have expressed concern that sharing their medical data could be used by insurance companies to deny them or their children insurance or to increase premiums.”

Kovarik noted that the American Academy of Dermatology offers a position statement on the use of augmented intelligence that includes language about transparency and the sharing of medical information. However, she believes there is more work to be done.

“There should be clarity in the symbiotic and synergistic roles of augmented intelligence and human judgment so that it is clear to the patient and provider when and how this technology is utilized to augment human judgment and interpretation,” she wrote. “For this goal to be reached, the current patient knowledge and perspective in each community needs to be understood.” – by Rob Volansky

Disclosures: Nelson reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures. Kovarik reports she is a member of the artificial intelligence task force for the American Academy of Dermatology.