Dermatology is the branch of medicine dealing with the skin and it is a specialty with both medical and surgical aspects. Dermatological imaging plays a key role in the clinical evaluation of skin lesions. The appearance, morphology, shape, and distribution of lesions are features that allow to recognize the manifestation of cutaneous or even systemic disease. The prediction of skin lesions is a challenging task even for experienced dermatologists due to the occasional low contrast between the surrounding skin and lesions, the visual resemblance between skin lesions, a fuddled lesion border, etc. An automated computer-aided detection system can help clinicians to predict malignant skin lesions at the earliest time.

Recent advances in deep learning have paved the way for the use of convolutional neural networks (CNNs) in this medical branch as well.  Their effectiveness, however, is dependent upon the quality and quantity of data available. This research line is focused on the development of fully automated algorithms for the optimization, classification, and segmentation of skin lesion images.


  • Francesco Branciforti (Ph.D. Student)
  • Kristen M. Meiburger (Tenure-track Assistant Professor (RTD/b))
  • Massimo Salvi (Assistant Professor with time contract (RTD/a))
  • Giulia Rotunno (Scholarship holder)
  • Collaborators:

  • Prof. Paola Savoia, Elisa Zavattaro, Federica Veronese - University Hospital Maggiore della Carità, Novara