Objective: The goal of this study was to compare live and digital image examination of hospitalized dermatology consultation patients.
Design: A medical student digitally imaged patients with a Nikon Coolpix 950 digital camera (Nikon Corp., Tokyo, Japan). A resident reviewed these images and the student recorded the differential diagnosis. Digital examination was followed by live dermatologic examination by the dermatology resident and then by a supervising board certified dermatologist. The diagnoses in 29 of these cases were compared with definitive diagnostic tests (e.g., skin biopsy).
Setting: Montefiore Hospital, Bronx, NY Patients. In March, 2000, 48 of 58 consecutive hospitalized patients for whom dermatology consultations were requested and from whom consent for photographs could be obtained.
Main outcome measures: The accuracy and consistency of the examiners top (up to three) digital diagnoses, (up to three) live diagnoses, and histopathologic diagnoses were compared.
Results: Eighty-one percent (38/47) of resident digital consult diagnoses agreed with live resident diagnoses. Live resident diagnosis agreement with live supervising board certified dermatologists diagnosis was 90% (18/20). Resident digital image diagnosis correlation with definitive diagnostic test results was 76% (22/29). Live supervising board certified dermatologist diagnosis correlation with definitive diagnostic test results was 73% (14/19).
Conclusions: Digital image review results in reliable and accurate diagnosis compared to live consultations. For inpatient consultations, three-dimensional lesions are more amenable to digital diagnosis than flat, generalized rashes.