Implementing-Deep-Learning-in-the-field-of-Radiology-and-Health-Care-using-Python

During my internship at “RAD365” (currently it is https://hailth.ai/), I was design various classification and segmentation CNN models using Python Keras, Opencv, Sklearn etc. libraries. Here I implement CNN models for CT, MRI, X-RAY images.

Libraries used in this project:

1. Numpy -- for matrix and array related operations
2. Matplotlib -- for plotting
3. Scikit image - for image read write,specificly I used it for .nii image read
4. pydicom -- for dicom image array read
5. Opencv -- for image processings opperation like blurring,thresholding etc etc.
6. Keras -- for Deep learning architecture building

Work

1. The first project is a multiclass classification problem, where I develop a Transfer learning based CNN model for four classes (1 for meningioma, 2 for glioma, 3 for pituitary tumour 4 for Normal). BRATS 2015 dataset

Deep learning architecture I used during this project:

2. The second project is a binary segmentation problem, where I develop an Encoder-Decoder based CNN model for segmenting spinal cord from MRI image.

Deep learning architecture I used during this project:

3. The third project is a segmentation problem, where I develop an Encoder-Decoder based CNN model for four classes (1 for meningioma, 2 for glioma, 3 for pituitary tumour 4 for Normal) segmentation. BRATS 2015 dataset

Deep learning architecture I used during this project: