A 3D Scanner is Helping Diagnose Skin Cancers
A non-invasive and affordable diagnosis tool
The most common type of cancer is skin cancer. The World Health Organization (WHO) reports that there are between two to three million non-melanoma biopsies performed worldwide, and 132,000 diagnosis of melanoma skin cancer each year.
A biopsy is an invasive surgical procedure that is done to determine if a suspicious lump or patch of discolored skin is cancerous. It is expensive, and can cause discomfort for the patient. There is a tool that can reduce the number of surgical biopsies, and lower the cost of a skin cancer diagnosis.
Spanish researchers have developed a 3D scanning tool that is showing promise at being able to diagnose skin cancers without the need for invasive surgery. Based on machine learning and fringe projection, the protype can distinguish between moles and melanomas by measuring the shape and volume of the lesion.
Fringe projection is used in the 3D scanner to get information on the skin in a matter of seconds. It creates a map of the various skin heights, similar to 3D laser printing. It allows physicians to see in minutes what used to require surgery. With high-resolution digital cameras and powerful imaging software, this technology shows promise as a noninvasive and affordable diagnostic tool.
The project is led by researchers at the Universitat Politècnica de Catalunya, headed by Santiago Royo and Meritxell Vilaseca Ricart. They designed a handheld protype that has a PEk3 picoprojector and two monochrome CCD cameras. A fringe pattern is created on the skin as the protype is moved across the area horizontally. When finished, the images from the three cameras are calibrated and go through an imaging process to create a 3D map of the skin. A machine that has a learning-based classification system is then used to identify the type of lesions.
The research team used the protype in a study that examined 654 skin lesions from patients at the Università degli Studi di Modena e Reggio Emilia and Hospital Clinic i Provincial de Barcelona. One of the project leaders spoke to Physics World stating that the patients used in the study were enrolled at one of the hospitals for treatment, and were mainly Caucasian.
The study may have been unknowingly limited to a certain population, but the early results were promising. The 3D scanning system had an 80-percent sensitivity rating and 76.6-percent specificity.
Santiago Royo stated, “In our pilot, we analyzed all suspect lesions, which were subsequently confirmed by histology. It was unfortunate that there were so few squamous cell and basal cell carcinomas in this population. It would be interesting to add more of these lesions to improve the specificity regarding these two cases, and to better separate them from melanoma. We are currently applying for funds to conduct this type of research.”
By adjusting the contrast and intensity of the projected fringes, the team hopes to expand their research to evaluating skin lesions on persons of African and Asian descent.