There are other examples in the cast_examples directory.
Here are gui screenshots:
We include the submodule in the ref directory.
This Segmentation is use the Clarius ultrasound device HD3 L7 to collect the ultrasound images. Here is the application and whole project detials for the segmentation of ultrasound images using Unet model. In order to get the mask of the vein in the ultrasound images.
The result of the prediction on the test dataset:
Infering on Tesla V100-PCIE-32GB
[Inference performance] Total number of samples: 290
[Inference performance] Total inference time: 0.6602s
[Inference performance] Average inference time per image: 0.0023s
[Inference performance] Inference frame rate (FPS): 439.25
[TEST] loss: 0.0155 | dice: 0.9866
A real time deomo of the segmentation model can be found in the GUI directory.
The video show below:
