We are a weekly newspaper serving the communities of Exeter, Lindsay, and Woodlake California.

Issue link:

Contents of this Issue


Page 7 of 27

Working to find a skin cancer solution That's just right Researchers find the "sweet spot" in classifying images of skin lesions to more accurately identify and diagnose melanoma T oo much, too little, just right. It might seem like a line from "Goldilocks and the Three Bears," but actually describes an important finding from researchers in Florida Atlantic Uni- versity's College of Engineering and Computer Science. They have developed a technique using machine learning — a sub-field of arti- ficial intelligence (AI) — that will enhance computer-aided diagnosis (CADx) of melanoma. Thanks to the algorithm they created — which can be used in mobile apps that are being developed to diagnose suspicious moles — they were able to determine the "sweet spot" in classifying images of skin lesions. This new finding, published in the Journal of Digital Imaging, will ultimately help clinicians more reliably identify and diagnose mela- noma skin lesions, distinguishing them from other types of skin le- sions. The research was conducted in the NSF Industry/University Cooperative Research Center for Advanced Knowledge Enablement (CAKE) at FAU and was funded by the Center's industry members. Melanoma is a particularly deadly form of skin cancer when left undiagnosed. In the United States alone, there were an estimated 76,380 new cases of melanoma and an estimated 6,750 deaths due to melanoma in 2016. Malignant melanoma and benign skin lesions often appear very similar to the untrained eye. Over the years, der- matologists have developed different classification methods to diag- nose melanoma, but to limited success (65 to 80 percent accuracy). As a result, computer scientists and doctors have teamed up to de- velop CADx tools capable of aiding physicians to classify skin le- sions, which could potentially save numerous lives each year. Images of skin lesions often contain more than just skin le- sions — background noise, hair, scars, and other artifacts in the image can potentially confuse the CADx system. To prevent the classifier from incorrectly associating these irrelevant artifacts with melano- ma, the images are segmented into two parts, separating the lesion from the surrounding skin, begging the question, "How much seg- mentation is too much?" Researchers first compared the effects of no segmentation, full segmentation, and partial segmentation on classification and dem- onstrated that partial segmentation led to the best results. They then proceeded to determine how much segmentation would be "just right." To do that, they used three degrees of partial segmentation, investigating how a variable-sized non-lesion border around the seg- mented skin lesion affects classification results. They performed com- parisons in a systematic and reproducible manner to demonstrate empirically that a certain amount of segmentation border around the lesion could improve classification performance. Their findings suggest that extending the border beyond the lesion to include a limited amount of background pixels improves their clas- sifier's ability to distinguish melanoma from a benign skin lesion. Their method showed an improvement across all relevant mea- sures of performance for a skin lesion classifier. — Newswise 6 | CENTRAL VALLEY MEDICAL | WINTER 2017-18 CanCer Care

Articles in this issue

Archives of this issue

view archives of Magazines - cvm_mag_vol6_no3