Skip to content

Detect and segment brain tumors precisely and fast.(accuracy=94%)

License

Notifications You must be signed in to change notification settings

sairbarbaros/deepcancer

Repository files navigation

DeepCancer, AI for Brain Tumor Detection and Segmentation

Detect and segment brain tumors precisely and fast. It's accuracy is 94.74% in detection and 96% in segmentation so far.

+This model uses MobileNetV3 architecture to build a model which can operate both precisely and fast.

+For segmentation, DeepCancer uses its own configured U-Net architecture written by sairbarbaros.

+Since there is a few datasets for biological segmentation, I used an ultrasonic breast dataset for my model. I saved my model and weights, and you can find those in my repo.

+You can use transfer learning for applying my model and its weights to brain tumor datasets.

+You can find this brain tumor detection dataset from https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection?datasetId=165566

+You can find precision graphs and further recommendations in my repo.

About

Detect and segment brain tumors precisely and fast.(accuracy=94%)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published