II-FCN for skin lesion analysis towards melanoma detection
This work addresses the problem of low-contrast skin lesion segmentation for medical imaging, representing an incremental improvement with domain-specific impact.
The paper tackled melanoma detection from dermoscopy images by developing a symmetrical identity inception fully convolutional network with reversible inception blocks and an approximate Jaccard loss function, achieving a Jaccard index of 0.82 on the ISBI 2017 lesion segmentation challenge.
Dermoscopy image detection stays a tough task due to the weak distinguishable property of the object.Although the deep convolution neural network signifigantly boosted the performance on prevelance computer vision tasks in recent years,there remains a room to explore more robust and precise models to the problem of low contrast image segmentation.Towards the challenge of Lesion Segmentation in ISBI 2017,we built a symmetrical identity inception fully convolution network which is based on only 10 reversible inception blocks,every block composed of four convolution branches with combination of different layer depth and kernel size to extract sundry semantic features.Then we proposed an approximate loss function for jaccard index metrics to train our model.To overcome the drawbacks of traditional convolution,we adopted the dilation convolution and conditional random field method to rectify our segmentation.We also introduced multiple ways to prevent the problem of overfitting.The experimental results shows that our model achived jaccard index of 0.82 and kept learning from epoch to epoch.