Breast Cancer Tumor Classification with ANN

Syed Ashar

Completed work

Data Scientist

Keras

pandas

scikit-learn

Healthcare

This project builds an Artificial Neural Network (ANN) model to classify breast cancer tumors as benign or malignant using the Wisconsin Breast Cancer Dataset.

The aim is to demonstrate how deep learning can support early and accurate diagnosis in medical applications.

🔍 Dataset

Source: UCI Machine Learning Repository

The dataset contains 30 numerical features derived from digitized images of breast mass samples.

Target variable: diagnosis

M = Malignant

B = Benign

💻 Technologies Used

Python 3.x

Keras (TensorFlow backend)

Pandas, NumPy

Seaborn, Matplotlib

Scikit-learn

🧠 Model Architecture

Input Layer: 30 neurons (1 per feature)

Hidden Layers:

Dense (16 units) + ReLU + Dropout

Output Layer:

Dense (1 unit) + Sigmoid (for binary classification)

from keras.models import Sequential
from keras.layers import Dense, Dropout, Input

model = Sequential()
model.add(Input(shape=(30,)))
model.add(Dense(16, activation='relu', kernel_initializer='uniform'))
model.add(Dropout(0.1))
model.add(Dense(16, activation='relu', kernel_initializer='uniform'))
model.add(Dropout(0.1))
model.add(Dense(1, activation='sigmoid', kernel_initializer='uniform'))

🧪 Evaluation Metrics

Accuracy

Confusion Matrix

Precision / Recall (optional)

Sample Confusion Matrix Result:

[[65  2]
 [ 3 44]]

True Negatives: 65

True Positives: 44

False Positives: 2

False Negatives: 3

📊 How to Run

Clone the repository:

git clone https://github.com/your-username/breast-cancer-ann.git
cd breast-cancer-ann

Install dependencies:

pip install -r requirements.txt

Run the notebook:

Open Classification_ANN.ipynb in Jupyter or Google Colab

📂 File Structure

├── data.csv                  # Dataset file
├── Classification_ANN.ipynb # Main notebook
├── README.md                 # This file
├── download.png              # Confusion matrix image
└── requirements.txt          # (Optional) Python dependencies

✅ Key Learnings

Applied neural networks to a real-world medical dataset

Understood the importance of reducing false negatives in healthcare

Gained experience with dropout regularization and evaluation metrics

📬 Contact

For feedback or collaboration, feel free to reach out via LinkedIn or GitHub.

Like this project

Completed work

Posted Aug 6, 2025

Built an ANN model to classify breast cancer tumors using the Wisconsin Breast Cancer Dataset.

Likes

Views