Which Explanation Makes Sense? A Critical Evaluation of Local Explanations for Assessing Cervical Cancer Risk Factors

Project Proposal submitted for CS598: Deep Learning For Healthcare, SP24

Instructors:

Jimeng Sun Siddhartha Laghuvarapu

Prepared by:

Himangshu Das hdas4@illinois.edu Jeremy Samuel sjeremy3@illinois.edu Mahesh Matta maheshm3@illinois.edu

Original paper - A-Critical-Evaluation-of-Local-Explanations-for-Assessing-Cervical-Cancer-Risk-Factors -

Final vide presentation (with 4 mins max limit) is here

Citations: Mustafa WA, Ismail S, Mokhtar FS, Alquran H, Al-Issa Y. Cervical Cancer Detection Techniques: A Chronological Review. Diagnostics (Basel). 2023 May 17;13(10):1763. doi: 10.3390/diagnostics13101763. PMID: 37238248; PMCID: PMC10217496.

Project final submission pdf link

Steps to run the code

1. first clone the git repo -

git clone https://github.com/dasshims/UIUC-CS598-FinalProject.git

2. Open the folder to an IDE, VSCode preferred
3. Open the ipynb file and set the python interpreter to 3.x.x, for our project we used 3.9.6.

If you are running this on Colab

1. make sure to uncomment the following cell to mount the drive folder.

from google.colab import drive
drive.mount('/content/drive')

2. Create the following directories to store the resulting files 3. data 4. documentation 5. img 6. models

Requirements

CPU: A multi-core processor with sufficient computational power for training machine learning models. A CPU with at least 4 cores and a clock speed of 2.5 GHz or higher is recommended.

RAM: Adequate RAM to handle the dataset size and model training. A minimum of 8 GB RAM is recommended for handling the preprocessing and training tasks efficiently.

For this project we used a macbook with 32Gigs of RAM and M2 Processor.