Technicians can use a microscope to observe tissue samples that were taken from patients who are suspected to have breast cancer. By looking at the size and shape of the nuclei present within these tissue samples, one can then predict whether a given sample appears to be cancerous. In this document I demonstrate an automated methodology to predict if a sample is benign or malignant given measurements of nuclear shape that were made from digital images of fine needle aspirates of breast tissue masses from clinical samples.
3
stars
11
commits
Jupyter Notebook
primary language
Jun 17, 2024
updated
README
Machine Learning - Breast Cancer Classification
Technicians can use a microscope to observe tissue samples that were taken from patients who are suspected to have breast cancer. By looking at the size and shape of the nuclei present within these tissue samples, one can then predict whether a given sample appears to be cancerous. In this document I demonstrate an automated methodology to predict if a sample is benign or malignant given measurements of nuclear shape that were made from digital images of fine needle aspirates of breast tissue masses from clinical samples.
Dataset Used: Breast Cancer Wisconsin (Diagnostic)
Methods
- Data Prep using Pandas DataFrames
- Data Visualization using Seaborn (matplotlib)
- Using Support Vector Machine (SVM) from scikit-learn
- Improving the model, by employing MinMaxScaler
- Using GridSearchCV to fine-tune hyper parameters
Screenshot
Contributors
11 commits
Similar repos
Languages
Jupyter Notebook
100.0%
Technicians can use a microscope to observe tissue samples that were taken from patients who are suspected to have breast cancer. By looking at the size and shape of the nuclei present within these tissue samples, one can then predict whether a given sample appears to be cancerous. In this document I demonstrate an automated methodology to predict if a sample is benign or malignant given measurements of nuclear shape that were made from digital images of fine needle aspirates of breast tissue masses from clinical samples.
3
stars
11
commits
Jupyter Notebook
primary language
Jun 17, 2024
updated
README
Machine Learning - Breast Cancer Classification
Technicians can use a microscope to observe tissue samples that were taken from patients who are suspected to have breast cancer. By looking at the size and shape of the nuclei present within these tissue samples, one can then predict whether a given sample appears to be cancerous. In this document I demonstrate an automated methodology to predict if a sample is benign or malignant given measurements of nuclear shape that were made from digital images of fine needle aspirates of breast tissue masses from clinical samples.
Dataset Used: Breast Cancer Wisconsin (Diagnostic)
Methods
- Data Prep using Pandas DataFrames
- Data Visualization using Seaborn (matplotlib)
- Using Support Vector Machine (SVM) from scikit-learn
- Improving the model, by employing MinMaxScaler
- Using GridSearchCV to fine-tune hyper parameters
Screenshot
Similar repos
Contributors
11 commits
Languages
Jupyter Notebook
100.0%