Tools

UMAP is a machine learning algorithm that is commonly used for dimensionality reduction and data visualization.

See the document on https://umap-learn.readthedocs.io/en/latest/basic_usage.html for usage of UMAP in python

HDBSCAN is a clutering algorithm which extends the DBSCAN algorithm. In our project we used it for clustering of UMAP results.

See the document on https://hdbscan.readthedocs.io/en/latest/how_hdbscan_works.html for usage of HDBSCAN

Convolutional Neural Networks (CNN or ConvNet) are complex feed forward neural networks. CNNs are used for image classification and recognition because of its high accuracy. In our project, we used this in hope to find the hidden pattern difference on the UMAP images of the cancerous patients

See a example of image recognition using CNN: https://towardsdatascience.com/wtf-is-image-classification-8e78a8235acb

PyTorch is a Python package that provides two high-level features:

• Tensor computation (like NumPy) with strong GPU acceleration

• Deep neural networks built on a tape-based autograd system

See the document on https://github.com/pytorch/pytorch for Pytorch tutorials