The earlier an accurate skin cancer diagnosis is made, the better the chances of cure. For the early detection of skin cancer, non-invasive imaging or physical methods of routine diagnostics have become established in specialized institutions in recent years, especially for malignant melanoma, basal cell carcinoma and squamous cell carcinoma (and actinic keratosis as a precursor), in addition to reflected light microscopy.
Modern imaging, biophysical methods, and artificial intelligence (AI) have great potential to improve early detection of skin cancer and precancerous changes, reduce unnecessary excisions of benign skin lesions, and enable the best possible decision-making for treatment.
Innovative imaging enables ‘optical biopsies’
In addition to confocal laser microscopy (in vivo and ex vivo) and optical coherence tomography (OCT), line-field confocal OCT (LC-OCT), dermatofluoroscopy and optoacoustic imaging are among the most innovative imaging techniques in non-invasive dermatological diagnostics. “Imaging in dermatology has developed rapidly over the last 20 years. This has benefited skin tumor diagnostics in particular, but now also the diagnosis of many inflammatory dermatoses,” explained Prof. Julia Welzel, MD, Director of the Clinic for Dermatology and Allergology at Augsburg University Hospital (D), Medical Campus South. “These ‘optical biopsies’ allow us to make an immediate classification and a quick therapy decision. And this is non-invasive – i.e. without surgery, pain or scars,” added Prof. Welzel. Another advantage is that diagnostics and therapy can take place immediately during one visit, a concept that is beginning to be known as ‘one stop store’.
“Non-invasive diagnostics therefore enable us to start non-surgical therapy immediately if necessary,” Prof. Welzel continued. One example is the treatment of basal cell carcinomas and actinic keratoses by means of local treatment measures or photodynamic therapy (PDT).
Modernize dermoscopy and combine it with AI
Dermatoscopy is a diagnostic tool that has been established for many years. Digital dermoscopy provides digital images and has been combined with AI for decision support. AI-based software analysis systems help detect initial melanomas. Sequential video dermatoscopy and whole-body photography also use digital evaluation tools with AI, supporting dermatologists’ work with automated lesion detection and tracking. “Sequential videodermatoscopy in combination with whole-body photography has the advantage of showing all changes over time, even if typical malignancy criteria are absent but pigmentary moles have morphologic or color dynamics,” Prof. Welzel explained.
With advanced imaging techniques and the use of AI for automated analysis and clinical decision support, the number of unnecessary biopsies and excisions for benign lesions can be reduced. “With these techniques, we can significantly improve early diagnosis as well as therapy monitoring of skin tumors and numerous other inflammatory dermatoses,” summed up Prof. Michael Hertl, MD, Director of the Department of Dermatology and Allergology at the University Hospital Marburg (D).
«Source: DDG-Tagung 2023: Nicht-invasive Diagnostik und Künstliche Intelligenz bei Hauttumoren», Deutsche Dermatologische Gesellschaft, 17.04.2023.
DERMATOLOGIE PRAXIS 2023; 33(3): 23
InFo ONKOLOGIE & HÄMATOLOGIE 2023; 11(4): 39