Tech Check: AI and Emerging Technology in Dermatology Practice

Summary for practice owners

The episode looks at technologies that may change how dermatology practices evaluate lesions and organize clinical work. Dr. David Carter explains how in vivo imaging has progressed from reflectance confocal microscopy and optical coherence tomography toward combinations of imaging methods and machine learning. A cited study uses multiphoton microscopy and optical coherence angiography to distinguish melanocytic lesions, with learned features used to classify difficult examples. He also describes millimeter wave imaging as a possible point of care tool: the device scans a lesion, processes many spatial measurements, and returns a malignancy probability. The discussion places these systems in a research and development stage, instead of presenting them as routine replacements for clinician assessment.

For owners, the useful lens is procurement readiness. The conversation gives a plain account of the processing chain behind image based AI, including image capture, lesion localization, feature extraction followed by classification. Carter identifies practical weaknesses that affect workflow: systems can confuse markers or artifacts with lesions, produce too many outliers, and react to small differences in follow up image angles. These issues can create extra review work and complicate responsibility for decisions. The episode also surveys molecular testing, including DNA and RNA methods, as well as protein approaches, and describes how gene expression profiling may contribute information when a biopsy result is indeterminate. A short segment on sunscreen sampling illustrates a low cost way to make product trials easier for patients. Taken together, the video helps practice leaders ask vendors about validation evidence, review workload, system fit, and staff time needed to interpret results before adopting emerging diagnostic tools.

Owner takeaways

• 3:05 A study of difficult lesions illustrates where combined imaging and machine learning might add information after dermoscopy.
• 7:19 Millimeter wave scanning is presented as a possible in clinic classification aid, with considerable image data processed into a probability score.
• 10:34 Image artifacts and lesion selection remain practical hurdles that vendors should explain clearly.
• 15:22 Whole body AI can create a substantial review burden when many lesions are flagged or images shift between visits.
• 17:01 Molecular tests measure different biological signals, giving owners context for the range of diagnostic tools entering dermatology.

Why it made the list

Dr. David Carter discusses imaging, machine learning, molecular tests, and product sampling. The video has 62 views. It has 62 views.

Related videos

This video is published by Dermatology Education Foundation on YouTube. Dermatologists.com is not affiliated with the creator, and inclusion is not an endorsement by either party. Watch it on YouTube.