A team of researchers at the University of Erlangen-Nuremberg in Germany, led by Dr. Moritz Ronneberger, has developed a new approach called "SUBSCAN" that uses artificial intelligence (AI) to detect basal cell carcinoma, the most common type of skin cancer, under the skin and before any visible changes or signs appear on the surface. This breakthrough technology has the potential to help diagnose the disease at an early stage.
The researchers used a technique called LC-OCT, which combines optical coherence tomography and confocal microscopy, to obtain high-resolution images of the skin in real-time. The AI system analyzes the images and identifies areas that are suspicious for basal cell carcinoma, allowing doctors to select areas that require more precise examination. While the final diagnosis remains the responsibility of the doctor, this technology could enable early detection and potentially reduce the need for surgical procedures.
Basal cell carcinoma is the most common type of skin cancer worldwide, and while it usually grows locally, it can destroy surrounding tissue and cause damage to sensitive areas such as the nose and eyes if left untreated. Early detection allows for less invasive treatment options, and in some cases, a topical cream may be sufficient. The SUBSCAN approach may also expand the use of local treatments and reduce the need for surgery.
The LC-OCT technology provides detailed images of the skin, allowing doctors to visualize changes beneath the surface. The AI system analyzes these images and provides a probability index, color-coded to indicate the likelihood of basal cell carcinoma. This enables doctors to prioritize areas that require further examination. However, the technology is still in the evaluation phase and is not yet ready for routine use due to limitations in sensitivity and the time-consuming nature of the examination.
The researchers conducted a systematic survey of the facial skin of individuals at high risk of skin cancer and were able to detect cases of basal cell carcinoma in areas without visible skin changes. The study demonstrates the potential of the SUBSCAN approach to detect skin cancer at an early stage, and further research is needed to confirm its effectiveness.
The development of this technology is a significant step forward in the early detection and treatment of skin cancer. While there are still challenges to overcome, the SUBSCAN approach has the potential to improve patient outcomes and reduce the burden of skin cancer on healthcare systems. Further studies are needed to evaluate the sensitivity and specificity of the technology.
The use of AI in medical diagnosis is becoming increasingly prevalent, and this study demonstrates the potential of AI-powered technology to improve the detection and treatment of skin cancer. As research continues to evolve, we can expect to see more innovative applications of AI in healthcare, leading to better patient outcomes and more efficient healthcare systems.
Key points
- Researchers develop AI-powered method to detect skin cancer under the skin
- The SUBSCAN approach uses LC-OCT technology and AI to identify basal cell carcinoma
- Early detection of skin cancer enables less invasive treatment options and potentially reduces the need for surgical procedures
