Abstract:
Generalized nucleus segmentation can speed up visual biomarker development and validation for new digital pathology datasets. The MoNuSeg 2018 Challenge developed generalizable nuclei segmentation methods for digital pathology.
Read Also: Python Projects
32 teams from diverse institutes competed in the MICCAI 2018 satellite event. A training set of 30 images from seven organs with 21,623 nuclei was given to contestants. A test dataset with 14 images from seven organs, including two not in the training set, was released unannotated.
Read Also: Communication Projects
The test set’s average aggregated Jaccard index (AJI) was used to prioritize accurate instance segmentation over semantic segmentation. Over half of the challenge-completing teams outperformed a baseline.
Color normalization and heavy data augmentation increased accuracy. Fully convolutional networks based on ResNet or VGG and inspired by U-Net, FCN, and Mask-RCNN were also popular.
Read Also: Deep Learning Projects
Watershed segmentation on predicted semantic segmentation maps was popular post-processing. Several top techniques outperformed a human annotator and can be used for nuclear morphometrics.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
