Cancer Deep Learning Model

Overview

This model is trained on 497 training examples and is tested for accuracy on 151 different testing examples. The accuracy is about 97%. The Python example code provides a simple example of using CSV data files with TensorFlow and training a model with three hidden layers.

I assume that you have Keras and TensorFlow installed.

Uses the Integrated Variants library to explain predictions made by a trained model

Please read this excellent paper by Mukund Sundararajan, Ankur Taly, and Qiqi Yan. When making a prediction, you can get a scaling of which input features most contributed to a classification made by the model.

A version of this code was used in a book I wrote

The github repository for my book “Introduction to Cognitive Computing” contains an older version of this example.

University of Wisconsin Cancer Data

- 0 Clump Thickness               1 - 10
- 1 Uniformity of Cell Size       1 - 10
- 2 Uniformity of Cell Shape      1 - 10
- 3 Marginal Adhesion             1 - 10
- 4 Single Epithelial Cell Size   1 - 10
- 5 Bare Nuclei                   1 - 10
- 6 Bland Chromatin               1 - 10
- 7 Normal Nucleoli               1 - 10
- 8 Mitoses                       1 - 10
- 9 Class (0 for benign, 1 for malignant)

I modified the original data slightly by removing the randomized patient ID and changing the target class values from (2,4) to (0,1) for (no cancer, cancer).

Licence: Apache License 2.0

Author: Mark Watson

Dependencies

• Python
• Keras
• tensorflow