Method and system for analyzing skin lesions

A method of analyzing a digital image of a skin lesion, the method comprising:

• using a dermoscopic camera to acquire the digital image of the skin lesion, the image including pixels distributed over a lesion area;
• calculating a first set of values indicative of a distribution of the image pixels, wherein the first set of values includes a first discrimination value indicative of a weighted average of the image pixels with weighing at a border of the skin lesion;
• calculating a second set of values indicative of lesion shape, wherein the second set of values includes a second discrimination value indicative of skewness and kurtosis of the distribution of the image pixels;
• calculating a third set of values indicative of lesion symmetry, wherein the third set of values includes a third discrimination value indicative of a ratio of symmetry and gray-level power of the distribution of the image pixels;
• providing a total additive score of the values in the first, second and third sets of values;
• comparing the total additive score with a total score threshold;
• comparing the first, second and third discrimination values with respective first, second and third discrimination threshold values;
• and
• providing an output classification for the image analyzed as a function of results of comparing the total additive score with the total score threshold and comparing the first, second and third discrimination values with respective first, second and third discrimination threshold values.

A system for analyzing an digital image of a skin lesion, the image including pixels distributed over a lesion area, the system comprising:

• processing circuits configured for processing the pixels of an analyzed image;
• and
• a non-transitory storage medium including software code for execution by the processing circuits, the software code including instructions for implementing a method comprising:
• calculating a total additive score of values in first, second and third sets of values, the first set of values including a first discrimination value indicative of a distribution of the image pixels, the second set of values including a second discrimination value indicative of lesion shape, and the third set of values including a third discrimination value indicative of lesion symmetry;
• comparing the total additive score with a total score threshold;
• comparing the first, second and third discrimination values with respective first, second and third discrimination threshold values;
• and
• providing an output classification for the image analyzed as a function of results of comparing the total additive score with the total score threshold and comparing the first, second and third discrimination values with respective first, second and third discrimination threshold values.