Semantic Segmentation for Medical Ultrasound Imaging
This is a video artifact generated during my Capstone project for my Master of Science in Data Science studies at the University of Wisconsin. The project is called Semantic Segmentation for Medical Ultrasound Imaging, and it has produced deep learning models trained to detect benign and malignant tumors in breast ultrasound images. The models identify the possible area of the lesion, and then mark it with colors: green if the lesion appears to be benign (not cancer), or red if it appears to be malignant (cancer). The two models used for segmentation were U-Net and SegFormer, written in PyTorch. Their predictions are labeled in each frame. The project repository is here: https://github.com/FlorinAndrei/datascience_capstone_project
This is a video artifact generated during my Capstone project for my Master of Science in Data Science studies at the University of Wisconsin. The project is called Semantic Segmentation for Medical Ultrasound Imaging, and it has produced deep learning models trained to detect benign and malignant tumors in breast ultrasound images. The models identify the possible area of the lesion, and then mark it with colors: green if the lesion appears to be benign (not cancer), or red if it appears to be malignant (cancer). The two models used for segmentation were U-Net and SegFormer, written in PyTorch. Their predictions are labeled in each frame. The project repository is here: https://github.com/FlorinAndrei/datascience_capstone_project