3D U-Net for Semantic Segmentation
Offered By: DigitalSreeni via YouTube
Course Description
Overview
Learn to implement a 3D U-Net for semantic segmentation in this comprehensive tutorial video. Explore the application of 3D U-Net to various volumetric imaging modalities such as FIB-SEM, CT, and MRI, with a focus on the BRATS dataset. Dive into the unit architecture, Python libraries, and annotation techniques using APEER. Follow along with the provided code and dataset to train and test machine learning algorithms for multiclass semantic segmentation. Gain insights into data preprocessing, model training, result evaluation, and handling multichannel images. Master the implementation of 3D U-Net for basic and advanced segmentation tasks, including OEM TIFF and multichannel image processing.
Syllabus
Introduction
Data types
Why 3D UNet
Unit Architecture
Python libraries
Annotation
Notebook
Running the code
Verify tensorflow and Keras
Import data
Local file location
Number of classes
Image dimensions
Multiclass classification
Learning rate
Preprocessing
Results
Testing
Saving
Multichannel
Basic segmentation
Final segmentation
OEM TIFF
Multichannel image
Summary
Taught by
DigitalSreeni
Related Courses
Computer Vision: The FundamentalsUniversity of California, Berkeley via Coursera Image and Video Processing: From Mars to Hollywood with a Stop at the Hospital
Duke University via Coursera Fundamentals of Digital Image and Video Processing
Northwestern University via Coursera 医学图像处理技术 Medical Image Analysis
Shanghai Jiao Tong University via Coursera Image Processing and Analysis for Life Scientists
École Polytechnique Fédérale de Lausanne via edX