+1 877 705 1914• 24 / 7
HIPAA Compliant
ISO 27001 Certified

Actively Recruiting

Phase Not Applicable
All Ages
All Genders
ID07682831

Evaluating Machine Learning Analysis of Two-Photon Fluorescence Microscopy for Diagnosing Skin Biopsy Specimens

Led by University of Rochester · Updated on 2026-08-19

92

Participants Needed

1

Research Sites

4 weeks

Total Duration

AI-Summary

What This Trial Is About

Researchers are evaluating a machine learning model to analyze two-photon fluorescence microscopy images of skin biopsy specimens taken during dermatology clinic visits. The study aims to see how well this imaging and analysis method agrees with traditional histology, which is the standard way to diagnose skin conditions like basal cell carcinoma and squamous cell carcinoma. Because two-photon images can be obtained from small biopsy samples within minutes, this approach could potentially allow faster diagnosis at the point of care. During the study, biopsy specimens such as punch, excisional, or shave biopsies will be imaged shortly after collection using two-photon fluorescence microscopy. A machine learning model will then assess these images and provide confidence scores for possible diagnoses including basal cell carcinoma, squamous cell carcinoma, or non-cancer. If the model has high confidence, it will provide a diagnosis otherwise, the specimen will be flagged for manual review by a pathologist. After imaging, specimens will undergo conventional histological processing to guide patient treatment. Participants will have their biopsy specimens imaged and analyzed on the same day of biopsy, while their final diagnosis will be based on histology slides reviewed by a pathologist. The study will compare the machine learning-based diagnosis with the gold-standard histology diagnosis. Researchers will measure the sensitivity and specificity of the machine learning analysis, the proportion of discordant diagnoses due to model errors, and the proportion of specimens with a definitive machine learning diagnosis. Treatment decisions remain based on conventional histology, not on the imaging method studied.

CONDITIONS

Brief Title

Machine Learning Analysis of Two-photon Fluorescence Microscopy of Dermatologic Biopsies

Research Team

M

Michael Giacomelli, Ph.D

Not the Right Trial for You?

Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.

Already have an account? Log in here