A new screening method promises the early detection of basal cell carcinoma, the most common type of skin cancer globally. This technique combines high-precision microscopic imaging with artificial intelligence, allowing doctors to identify tumors in real-time, even when they are invisible to the naked eye.
The Technology: LC-OCT and AI
The technique, known as Linear-field Optical Coherence Tomography (LC-OCT), is a hybrid approach. It combines Optical Coherence Tomography (OCT), which visualizes the depth and extent of a tumor, with confocal microscopy, which allows for imaging at the cellular level. Together, they produce vertical, horizontal, and 3D images of the skin.
The LC-OCT functions at a micrometer resolution, enabling physicians to examine structures only a few thousandths of a millimeter wide. By comparison, a human hair is approximately 50 micrometers thick. AI analyzes these images as they are captured, providing color-coded indicators of the probability of cancer presence, though the final diagnosis remains the physician's responsibility.
The SUBSCAN Study Results
In a feasibility study involving 150 high-risk patients, systematic facial skin imaging identified subclinical tumors in 14 participants (9.3%). The positive predictive value of the AI-assisted method reached 83.3%, with most flagged lesions confirmed by biopsy.
Dr. Moritz Rennicke notes that early detection through the "SUBSCAN" approach could make treatments significantly less invasive. In some instances, early diagnosis allows for the use of topical creams, potentially eliminating the need for surgery.
Despite its potential, the method is not yet ready for widespread clinical use. Researchers state that the process is currently too time-consuming and more data is needed to determine the system's overall sensitivity—specifically, how many tumors it might potentially miss.
