This is a vignette to perform supervised classification and weakly supervised regression for Label-free assessment of red blood cell storage lesions, as well as a comparative analysis using conventional image processing (CellProfiler)
A full installation guide can be found here. Briefly, core requirements include:
- Python 3.6
- Tensorflow-gpu 1.9.0
- Keras 2.1.5
- Numpy 1.18.1
- Scipy 1.4.1
- Keras-resnet 0.0.7
- Python-bioformats 1.5.2
- Jupyter notebook
- CellProfiler 3.1.8+
Please use the provided Jupyter notebooks following the step-wise order.