Skip to main content

A package for breast cancer diagnosis using MLP classifier.

Project description

Breast Cancer Diagnosis with MLP 🩺💻

GitHub PyPI License

🧠 This project harnesses the power of a Multi-Layer Perceptron (MLP) neural network, implemented with scikit-learn, to perform breast cancer diagnosis based on tumor characteristics extracted from biopsy samples. The MLP model is a type of artificial neural network designed to learn complex patterns in data, making it well-suited for tasks like medical diagnosis.

🔬 The MLP model is trained on a comprehensive dataset containing various features derived from digital images of breast tissue samples. These features include mean radius, texture, perimeter, area, smoothness, compactness, concavity, concave points, symmetry, and fractal dimension. Each feature provides valuable information about the physical properties and spatial arrangements of cells within the tissue, enabling the model to learn to distinguish between benign and malignant tumors.

💡 By analyzing these features, the MLP model can effectively classify breast tumors as either benign or malignant, providing valuable diagnostic information to healthcare professionals. This approach offers a non-invasive and automated method for cancer detection, potentially improving patient outcomes through earlier detection and treatment.

Purpose 🎯

The primary objective of this project is to develop an accurate and reliable system for diagnosing breast cancer based on quantitative analysis of cell nuclei characteristics. By leveraging machine learning techniques, specifically MLP neural networks, we aim to create a predictive model capable of classifying tumors as either malignant (cancerous) or benign (non-cancerous) with high accuracy. Early and accurate diagnosis of breast cancer can significantly improve patient outcomes by enabling timely treatment and intervention.

Key Features 🔑

  • Utilizes an MLP neural network for breast cancer diagnosis. 🤖
  • Preprocesses input data using feature scaling with StandardScaler. 📊
  • Implements training and evaluation functionalities. 📈
  • Provides prediction capabilities for new biopsy samples. ⚡
  • Offers detailed model evaluation metrics, including accuracy and confusion matrix. 📊
  • Supports easy integration into Python applications for breast cancer diagnosis tasks. 🐍

Installation 🚀

You can easily install BreastCancerMLPModel using pip:

pip install BreastCancerMLPModel

How BreastCancerMLPModel Works 🤖

BreastCancerMLPModel utilizes an MLP neural network for breast cancer diagnosis. Here's how it works:

  1. Initializing the Model 🛠️:

    • The model is initialized using the BreastCancerMLPModel class from the package.
    • This class encapsulates an MLPClassifier from scikit-learn with predefined parameters.
  2. Preprocessing Input Data 📊:

    • Input data undergoes preprocessing using feature scaling with StandardScaler.
    • Scaling ensures that features are on the same scale, improving model performance.
  3. Training and Evaluation 📈:

    • The model is trained using the fit() method, which splits the dataset, scales features, and trains the MLP model.
    • Evaluation metrics, including accuracy and confusion matrix, are provided to assess model performance.
  4. Making Predictions ⚡:

    • The predict() method enables prediction capabilities for new biopsy samples.
    • Input data, such as tumor characteristics, is provided to the model for prediction.
  5. Integration with Python Applications 🐍:

    • BreastCancerMLPModel supports easy integration into Python applications for breast cancer diagnosis tasks.
    • This allows seamless incorporation of the model into existing workflows for efficient diagnosis.

This approach ensures accurate and reliable breast cancer diagnosis based on tumor characteristics, enabling better patient care and treatment decisions.

from BreastCancerMLPModel.BreastCancerMLPModel import BreastCancerMLPModel

# Example usage

# Initialize the model
model = BreastCancerMLPModel()

# Train the model
model.fit()

# Make predictions

# Data for prediction 1
data1 = "mean_radius: 17.99, mean_texture: 10.38, mean_perimeter: 122.8, mean_area: 1001, mean_smoothness: 0.1184, mean_compactness: 0.2776, mean_concavity: 0.3001, mean_concave_points: 0.1471, mean_symmetry: 0.2419, mean_fractal_dimension: 0.07871, se_radius: 1.095, se_texture: 0.9053, se_perimeter: 8.589, se_area: 153.4, se_smoothness: 0.006399, se_compactness: 0.04904, se_concavity: 0.05373, se_concave_points: 0.01587, se_symmetry: 0.03003, se_fractal_dimension: 0.006193, worst_radius: 25.38, worst_texture: 17.33, worst_perimeter: 184.6, worst_area: 2019, worst_smoothness: 0.1622, worst_compactness: 0.6656, worst_concavity: 0.7119, worst_concave_points: 0.2654, worst_symmetry: 0.4601, worst_fractal_dimension: 0.1189"
prediction1 = model.predict(data1)
print("Predicted diagnosis for data 1:", prediction1) ## ('Maligno', 1.0)

# Data for prediction 2
data2 = "mean_radius: 13.08, mean_texture: 15.71, mean_perimeter: 85.63, mean_area: 520, mean_smoothness: 0.1075, mean_compactness: 0.127, mean_concavity: 0.04568, mean_concave_points: 0.0311, mean_symmetry: 0.1967, mean_fractal_dimension: 0.06811, se_radius: 0.1852, se_texture: 0.7477, se_perimeter: 1.383, se_area: 14.67, se_smoothness: 0.004097, se_compactness: 0.01898, se_concavity: 0.01698, se_concave_points: 0.00649, se_symmetry: 0.01678, se_fractal_dimension: 0.002425, worst_radius: 14.5, worst_texture: 20.49, worst_perimeter: 96.09, worst_area: 630.5, worst_smoothness: 0.1312, worst_compactness: 0.2776, worst_concavity: 0.189, worst_concave_points: 0.07283, worst_symmetry: 0.3184, worst_fractal_dimension: 0.08183"
prediction2 = model.predict(data2)
print("Predicted diagnosis for data 2:", prediction2) ##('Benigno', 0.9999982189891156)

Dataset 📊

The dataset used in this project is the Breast Cancer Wisconsin (Diagnostic) dataset, available in scikit-learn's built-in datasets module. It consists of features computed from digital images of fine needle aspirate (FNA) of breast masses. Each feature represents various characteristics of cell nuclei present in the images. The dataset contains both malignant and benign tumor samples, making it suitable for binary classification tasks.

Features and Descriptions

Label Meaning Weight in Diagnosis Description
Diagnosis Diagnosis (M = malignant, B = benign) Not used Result of breast cancer diagnosis
mean_radius Mean radius of cell nuclei High Average distance from the center to the points on the perimeter of cell nuclei
mean_texture Mean texture of cell nuclei Low Standard deviation of gray-scale values in the image of cell nuclei
mean_perimeter Mean perimeter of cell nuclei High Average lengths of perimeters of cell nuclei
mean_area Mean area of cell nuclei Very High Average areas of cell nuclei
mean_smoothness Mean smoothness of cell nuclei Low Local variation in lengths of cell nuclei radii
mean_compactness Mean compactness of cell nuclei High (Perimeter^2 / area) - 1.0
mean_concavity Mean concavity of cell nuclei Very High Severity of concave portions of cell nuclei contour
mean_concave_points Mean concave points of cell nuclei Very High Number of concave portions of cell nuclei contour
mean_symmetry Mean symmetry of cell nuclei Low Symmetry of cell nuclei
mean_fractal_dimension Mean fractal dimension of cell nuclei Low Coastline approximation of cell nuclei
se_radius Standard error of radius Medium Standard error of cell nuclei radius
se_texture Standard error of texture Low Standard error of cell nuclei texture
se_perimeter Standard error of perimeter Medium Standard error of cell nuclei perimeter
se_area Standard error of area Medium Standard error of cell nuclei area
se_smoothness Standard error of smoothness Low Standard error of cell nuclei smoothness
se_compactness Standard error of compactness Medium Standard error of cell nuclei compactness
se_concavity Standard error of concavity High Standard error of cell nuclei concavity
se_concave_points Standard error of concave points High Standard error of cell nuclei concave points
se_symmetry Standard error of symmetry Low Standard error of cell nuclei symmetry
se_fractal_dimension Standard error of fractal dimension Low Standard error of cell nuclei fractal dimension
worst_radius Worst value of radius High Worst value of cell nuclei radius
worst_texture Worst value of texture Low Worst value of cell nuclei texture
worst_perimeter Worst value of perimeter High Worst value of cell nuclei perimeter
worst_area Worst value of area Very High Worst value of cell nuclei area
worst_smoothness Worst value of smoothness Low Worst value of cell nuclei smoothness
worst_compactness Worst value of compactness High Worst value of cell nuclei compactness
worst_concavity Worst value of concavity Very High Worst value of cell nuclei concavity
worst_concave_points Worst value of concave points Very High Worst value of cell nuclei concave points
worst_symmetry Worst value of symmetry Low Worst value of cell nuclei symmetry
worst_fractal_dimension Worst value of fractal dimension Low Worst value of cell nuclei fractal dimension

Usage 🚀

  1. Training the Model: The model is trained using the fit method, which loads the dataset, preprocesses the input features, and trains the MLP classifier.

  2. Making Predictions: After training, the model can be used to make predictions on new biopsy samples using the predict method. The input data should be provided in a specific format, including features such as mean radius, texture, perimeter, etc.

  3. Evaluation: The model's performance can be evaluated using various metrics, including accuracy and confusion matrix, to assess its diagnostic capabilities.

Dependencies 🛠️

  • scikit-learn
  • numpy

License 📜

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments 🙏

This dataset is a copy of the UCI ML Breast Cancer Wisconsin (Diagnostic) datasets. UCI Machine Learning Repository.

The input features are computed from a digitized image of a fine needle aspirate (FNA) of a breast mass. They describe characteristics of the cell nuclei present in the image.

The separation plane described above was obtained using the Multiple Surface Method Tree (MSM-T) [K. P. Bennett, "Constructing a Decision Tree by Linear Programming". Proceedings of the 4th Midwest Artificial Intelligence and Cognitive Science Society, pp. 97-101, 1992], a classification method that uses linear programming to build a decision tree. Relevant features were selected through an exhaustive search in the space of 1-4 features and 1-3 separation planes.

The actual linear program used to obtain the separation plane in the three-dimensional space is described in: [K. P. Bennett and O. L. Mangasarian: "Robust Linear Programming Discrimination of Two Linearly Inseparable Sets", Optimization Methods and Software 1, 1992, 23-34].

This database is also available through the UW CS ftp server:

ftp ftp.cs.wisc.edu cd math-prog/cpo-dataset/machine-learn/WDBC/

References:

  1. W.N. Street, W.H. Wolberg and O.L. Mangasarian. Nuclear feature extraction for breast tumor diagnosis. IS&T/SPIE 1993 International Symposium on Electronic Imaging: Science and Technology, volume 1905, pages 861-870, San Jose, CA, 1993.
  2. O.L. Mangasarian, W.N. Street and W.H. Wolberg. Breast cancer diagnosis and prognosis via linear programming. Operations Research, 43(4), pages 570-577, July-August 1995.
  3. W.H. Wolberg, W.N. Street, and O.L. Mangasarian. Machine learning techniques to diagnose breast cancer from fine-needle aspirates. Cancer Letters 77 (1994) 163-171.

Contribution

Contributions to BreastCancerMLPModel are highly encouraged! If you're interested in adding new features, resolving bugs, or enhancing the project's functionality, please feel free to submit pull requests.

Get in Touch 📬

BreastCancerMLPModel is developed and maintained by Sergio Sánchez Sánchez (Dream Software). Special thanks to the open-source community and the contributors who have made this project possible. If you have any questions, feedback, or suggestions, feel free to reach out at dreamsoftware92@gmail.com.

Visitors Count

Please Share & Star the repository to keep me motivated.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

breastcancermlpmodel-0.0.32.tar.gz (9.5 kB view details)

Uploaded Source

Built Distribution

BreastCancerMLPModel-0.0.32-py3-none-any.whl (8.8 kB view details)

Uploaded Python 3

File details

Details for the file breastcancermlpmodel-0.0.32.tar.gz.

File metadata

  • Download URL: breastcancermlpmodel-0.0.32.tar.gz
  • Upload date:
  • Size: 9.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.11.6

File hashes

Hashes for breastcancermlpmodel-0.0.32.tar.gz
Algorithm Hash digest
SHA256 2bdf0865d3e0f417c92df4879b1027aed8b4e3b599a58e2466855a7f5a0fc53b
MD5 14c950f36866c50afa8271aca31504b0
BLAKE2b-256 c4c7b3cd9fdfa3e8b608d03d6eb51179a1050f2f9863b4bffeaa8a299eb77ac2

See more details on using hashes here.

File details

Details for the file BreastCancerMLPModel-0.0.32-py3-none-any.whl.

File metadata

File hashes

Hashes for BreastCancerMLPModel-0.0.32-py3-none-any.whl
Algorithm Hash digest
SHA256 2b1b2868e0b69e1e9e5b84fde9b5fd9ef0b8bdce6c6893f132f4d5e55af0f511
MD5 0ec37961b4ca08b857d588ca6a4358b2
BLAKE2b-256 5a3040e4f8a011536463fc3f5118bc8c11332539068ca06ae052e223a63d5698

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page