Improving data participation for the development of artificial intelligence in dermatology.

Artificial intelligence (AI) has the potential to significantly impact many aspects of dermatology. The visual nature of dermatology lends itself to innovations in this space. The robustness of AI algorithms depends on the quality, quantity, and variety of data it is trained and tested on. Image collections can suffer from inconsistencies in image quality, underrepresentation of various anatomic sites and skin tones, and lack of benign counterparts leading to underperformance of algorithms in contexts other than one in which it is developed. Access to care, trust, rights, control, and transparency all play roles in the willingness of patients and healthcare providers and systems to collect, provide, and share data. Opportunities to improve data participation for the development of artificial intelligence include the establishment of data hubs and public algorithms, federated learning strategies, development of renumeration ecosystems for patients and systems, and development of criteria and mechanisms for transparency.

Teledermatology during COVID-19: a comparison of video and telephone visits with patient-uploaded images at an urban academic medical center.

The COVID-19 pandemic stimulated adoption of teledermatology via video and telephone modalities by outpatient dermatology clinics. However, it was unknown how patient-related factors may have impacted, whether video or phone visits were used, and if visit modality impacted management. Consequently, we conducted a retrospective cross-sectional study of teledermatology visits occurring between March 30, 2020 and May 30, 2020 at an urban tertiary care center. A total of 788 teledermatology visits including 525 video visits and 263 telephone visits, mostly supplemented by patient-uploaded images, were analyzed. Patient age (P<0.001) and visit type (new versus return patient status), (P<0.001) were significant predictors of likelihood of video visit. No significant difference between video and telephone visits was found with regard to frequency of treatment modification (P=0.52), frequency of biopsy referral (P=0.73), biopsy noncompliance rate (P=0.44), or proportion of biopsies showing a new malignant lesion (P=0.92). With age as a significant predictor of visit modality, maintaining both video and phone modalities could prove useful to maximize patient participation. It appears either can be used without concern that choice of modality would impair the ability to change treatment, recognize a lesion requiring biopsy, recognize a new malignant lesion, or negatively affect compliance with biopsy.

Painful parotid hypertrophy with bulimia: a report of medical management.

In eating disorders, such as bulimia nervosa, body image disturbance often extends beyond the realm of weight and shape into the dermatologic spectrum. While commonly associated conditions due to binging and self-induced vomiting include cutaneous entities (e.g., Russell's sign, acne, alopecia and hypertrichosis) and oral pathologies (e.g., enamel erosion, caries and mild parotid hypertrophy), a rare but troubling manifestation is disfiguring parotid enlargement (sialoadenomegaly). This article presents a case of painful sialoadenomegaly associated with hyperamylasemia in a bulimic patient successfully managed with pilocarpine.