Medical image processing method and device, electronic device and storage medium
A medical image and processing method technology, applied in the information field, can solve problems such as inaccurate target segmentation and poor medical image processing
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 1
[0239] Increase the diversity of input images through a large number of sample data enhancements to avoid network overfitting.
[0240] Training neural network, structure extraction features, including residual network and deep layer aggregation (deep layeraggregation, DLA). Feature upsampling using a U-Net-like network. The network outputs two channels, including: a channel that outputs a masked image of the nucleus, and another channel that outputs a masked image of the overlapping portion of the nucleus boundary.
[0241] The outputs of the two channels were post-processed using a random walk algorithm and a watershed algorithm to separate adjacent nuclei.
[0242] Through a random forest classifier, some false detections are eliminated and the accuracy rate is improved.
[0243] The implementation of increasing image diversity can be as follows:
[0244] Use random cutting, random rotation, random flip, color perturbation, brightness contrast perturbation, limited contr...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


