From: Application of machine learning techniques for predicting survival in ovarian cancer
Feature | Domain of values | Type of feature |
---|---|---|
County | 165 different values | Categorical |
Histologic type ICD-O-3 | 150 different values | Categorical |
Laterality | 1. Bilateral, single primary 2. Paired site, but no information concerning laterality 3. Right - origin of primary 4. Left - origin of primary 5. Only one side - side unspecified | Categorical |
Radiation sequence with surgery | 1. No radiation and/or cancer-directed surgery 2. Radiation after surgery 3. Radiation prior to surgery 4. Sequence unknown, but both were given 5. Radiation before and after surgery 6. Intraoperative radiation | Categorical |
Reason no cancer-directed surgery | 1. Surgery performed 2. Not recommended 3. Recommended but not performed, unknown reason 4. Not recommended, contraindicated due to other cond; autopsy only 5. Not performed, patient died prior to recommended surgery | Categorical |
Sequence number | 1. One primary only 2. 1st of 2 or more primaries | Categorical |
Race recode | 1. White 2. Black 3. Asian or Pacific Islander 4. American Indian/Alaska Native | Categorical |
Marital status at diagnosis | 1. Married (including common law) 2. Widowed 3. Single (never married) 4. Divorced 5. Separated 6. Unmarried or Domestic Partner | Categorical |
PRCDA region | 1. Pacific Coast 2. East 3. Northern Plains 4. Southwest 5. Alaska | Categorical |
Summary stage | 1. Distant 2. Regional 3. Localized | Categorical |
Insurance recode | 1. Insured 2. Insured/No specifics 3. Any Medicaid 4. Insurance status unknown 5. Uninsured | Categorical |
CS site-specific factor 1 | 6 different numeric values (Mean:509, Standard deviation: 535.71, Range: 10–999) | Numerical |
Year of diagnosis | 17 different years (Range: 2000–2016) | Numerical |
Age at diagnosis | 100 different ages (Range: 0-113) | Numerical |
Chemotherapy recode | 1. yes 2. no | Categorical |
Rural-Urban continuum code | 1. Counties in metropolitan areas GE 1Â million pop 2. Counties in metropolitan areas of 250,000 to 1Â million pop 3. Counties in metropolitan areas of LT 250 thousand pop 4. Urban pop of 2,500 to 19,999, adjacent to a metro area 5. Urban pop of 2,500 to 19,999, not adjacent to a metro area 6. Urban pop of GE 20,000 adjacent to a metropolitan area 7. Urban pop of GE 20,000 not adjacent to a metropolitan area 8. Comp rural LT 2,500 urban pop, not adjacent to metro area 9. Comp rural LT 2,500 urban pop, adjacent to a metro area | Categorical |
Grade | 1. Well differentiated; Grade I 2. Moderately differentiated; Grade II 3. Poorly differentiated; Grade III 4. Undifferentiated; anaplastic; Grade IV | Categorical |