Breast Cancer Diagnostic Metrics

Columns

• id: Unique identification number assigned to each patient.
• diagnosis: The target variable indicating the cancer status, classified as either Malignant (M) or Benign (B).
• radius_mean: The mean of distances from the centre to points on the perimeter.
• texture_mean: Standard deviation of grey-scale values (mean).
• perimeter_mean: Mean size of the core tumour.
• area_mean: Mean area of the tumour.
• smoothness_mean: Mean of local variation in radius lengths.
• compactness_mean: Mean of perimeter^2 / area - 1.0.
• concavity_mean: Mean severity of concave portions of the contour.
• concave points_mean: Mean number of concave portions of the contour.
• symmetry_mean: Mean symmetry of the nucleus.
• fractal_dimension_mean: Mean 'coastline approximation' - 1.
• radius_se: Standard error for the mean of distances from centre to points on the perimeter.
• texture_se: Standard error for texture.
• perimeter_se: Standard error for perimeter size.
• area_se: Standard error for the area.
• smoothness_se: Standard error for local variation in radius lengths.
• compactness_se: Standard error for compactness.
• concavity_se: Standard error for concavity.
• concave points_se: Standard error for concave points.
• symmetry_se: Standard error for symmetry.
• fractal_dimension_se: Standard error for fractal dimension.
• radius_worst: "Worst" or largest mean value for radius.
• texture_worst: "Worst" or largest mean value for texture.
• perimeter_worst: "Worst" or largest mean value for perimeter.
• area_worst: "Worst" or largest mean value for area.
• smoothness_worst: "Worst" or largest mean value for smoothness.
• compactness_worst: "Worst" or largest mean value for compactness.
• concavity_worst: "Worst" or largest mean value for concavity.
• concave points_worst: "Worst" or largest mean value for concave points.
• symmetry_worst: "Worst" or largest mean value for symmetry.
• fractal_dimension_worst: "Worst" or largest mean value for fractal dimension.

Distribution

• Format: CSV (Comma Separated Values)
• Size: Approximately 125.2 kB
• Structure: 32 Columns and 569 Valid Rows
• Data Types:
  • Numeric (Float) for all feature columns (mean, standard error, worst).
  • Categorical for diagnosis (2 unique values: B and M).
  • Integer/Numeric for id.

Usage

• Binary Classification: Training algorithms to predict if a tumour is Malignant or Benign based on physical features.
• Exploratory Data Analysis (EDA): Visualising correlations between tumour size, texture, and malignancy.
• Feature Selection: Identifying which cell nucleus characteristics are most indicative of cancer.
• Educational Training: Teaching beginners the fundamentals of healthcare data science and logistic regression.
• Diagnostic Tool Development: Prototyping support systems for medical professionals.

Coverage

• Domain: Healthcare and Oncology.
• Scope: The data focuses on the physical attributes of cell nuclei derived from breast mass images.
• Demographics: Patient IDs are anonymised; data represents a cohort of 569 distinct cases.
• Completeness: The file reports 0% missing data and 0% mismatched values across all columns.

License

Who Can Use It

• Data Scientists: For building and testing classification algorithms.
• Medical Researchers: To analyse morphological features of breast masses.
• Students: As a standard practice set for machine learning coursework.
• Healthcare Analysts: To model diagnostic trends.

Dataset Name Suggestions

• Breast Cancer Diagnostic Metrics
• Benign vs Malignant Tumour Classification Data
• Cell Nuclei Features for Cancer Detection
• Oncology Diagnostic Parameter Set

Attributes