Machine learning based skin lesion segmentation method with novel borders and hair removal techniques

Abstract

The effective segmentation of lesion(s) from dermoscopic skin images assists the Computer-Aided Diagnosis (CAD) systems in improving the diagnosing rate of skin cancer. The results of the existing skin lesion segmentation techniques are not up to the mark for dermoscopic images with artifacts like varying size corner borders with color similar to lesion(s) and/or hairs having low contrast with surrounding background. To improve the results of the existing skin lesion segmentation techniques for such kinds of dermoscopic images, an effective skin lesion segmentation method is proposed in this research work. The proposed method searches for the presence of corner borders in the given dermoscopc image and removes them if found otherwise it starts searching for the presence of hairs on it and eliminate them if present. Next, it enhances the resultant image using state-of-the-art image enhancement method and segments lesion from it using machine learning technique namely, GrabCut method. The proposed method was tested on PH2 and ISIC 2018 datasets containing 200 images each and its accuracy was measured with two evaluation metrics, i.e., Jaccard index, and Dice index. The evaluation results show that our proposed skin lesion segmentation method obtained Jaccard Index of 0.77, 0.80 and Dice index of 0.87, 0.82 values on PH2, and ISIC2018 datasets, respectively, which are better than state-of-the-art skin lesion segmentation techniques.

Fig 1. Dermoscopic image containing corner borders and hairs.

Fig 2. Flow diagram of our proposed skin lesion segmentation method.

Fig 3

a). Input image with corner borders, b) extreme outer contour of corner borders, c) the extreme inner contour of corner borders, d) the result after corner borders removal.

Fig 4. Extreme end point calculation for inner rectangle.

Fig 5. Workflow diagram of hairs removal process.

Fig 6. In-painting process of a point P.

Fig 7

Hair removal process a) result of the previous module b) the contour of hair in Gray Scale Image c) Mask Computed from contour image d) RGB image after Hair removal.

Fig 8

a) Original Image, b) Histogram equalization c) CLAHE d)modified histogram and log exponential transformation [16].

Fig 9

a) Input obtained from previous module b) Results of segmentation module.

Fig 10. Sample results of our proposed method.

Fig 11. Results of our proposed method over dermoscopic images taken from ISIC 2018 dataset containing small size infected region.