Dermatologists Receptive to AI, AuI for Diagnosis of Skin Lesions
Implementation challenges include disruption of the human physician-patient relationship, threats to accuracy
By Dermsquared Editorial Team | June 02, 2021
Dermatologists are receptive to artificial intelligence (AI) and augmented intelligence (AuI), but do have concerns regarding their use, according to a research letter published online May 26 in JAMA Dermatology.
Caroline A. Nelson, M.D., from the Yale School of Medicine in New Haven, Connecticut, and colleagues examined dermatologists' perspectives on AI and AuI in a cross-sectional survey, completed by 121 dermatologists.
The researchers found that 46 percent of dermatologists thought that AI would positively influence their practice compared with 64 percent for AuI. To help diagnose skin lesions in a clinic and to help monitor skin lesions, 57 and 74 percent of respondents, respectively, reported that they would use an AI tool with similar accuracy to a human dermatologist. If an AI tool indicated a possible malignant diagnosis, 76 percent of respondents stated that they would be more likely to perform a biopsy of a lesion that was not clinically suggestive of cancer. Dermatologists prioritized more efficient triage, improved health care access, and quicker diagnosis as benefits of AI tools for skin cancer screening, while lack of patient follow-up, clinician loss of control to AI, and human deskilling were prioritized as risks. Implementation challenges identified by most dermatologist respondents included disruption of the human physician-patient relationship and threats to accuracy.
"Dermatologists are receptive to assistance from AI tools in diagnosing and monitoring skin lesions but value the human physician-patient relationship and accuracy," the authors write.