软银赛富开始了他的进步谈工作，我们在22ND of March and started by discussing the automatic quantification of epidermis curvature in H&E stained microscope skin images of mice. He introduces his work by first showing us how the algorithm which is used is divided into three stages; stage 1 involves the segmentation based on the image colour intensity of the mice skin samples to separate the original image into three individual layers. By doing this, the epidermis, dermis and dispose layer are more prominent for testing. Stage 2 can be seen to incorporate further segmentation on the first layer as once completed, this will be used to measure the difference between the epidermis edges on either end. The final stage extracts these measurements so that it can be used to identify a reoccurring pattern of genes which is linked to changes in the epidermis curvature. The dataset used for this involves 5714 H&E images of wild and mutant type mice skin layers, this is used to improve the effectiveness of the curvature results aND algorithm.
Our work in progress speaker, Saif Hussein, gave us a better understanding of his research and progress in the quantification and segmentation process of H&E stained microscope skin images of mice. This seminar also gave students aND staff the opportunity to benefit from the discussion of ‘the automatic quantification of epidermis curvature in H&E stained microscope skin images of mice”，可能纳入新知识纳入自己的研究。对于谁可能会考虑在计算或为那些谁可能有兴趣一般一个疗程的个体，这些讨论提供了一个深入了解当前学生的研究，并在那里将带领他们。