Table 1.

Analysis of encoding and dimensionality for clinical variables

Clinical variableEncodingDimensionUnique categories / note
AgeScalar1N/A (continuous value)
SexOne-hot2[Male, Female]
ECOG PSScalar1N/A (ordinal scale 0–4)
Smoking pack-yearsScalar1N/A (continuous value)
Smoking statusOne-hot3[‘Current’, ‘Former’, ‘Never’]
Disease siteOne-hot19[‘Oropharynx’, ‘Larynx’, ‘Hypopharynx’,…]
Tumor subsiteOne-hot63[‘Tonsil’, ‘Base of tongue’, ‘Glottis’,…]
Tumor size category (T)One-hot17[‘T1’, ‘T2’, ‘T3’, ‘T4a’, ‘T4b’,…]
Nodal involvement (N)One-hot10[‘N0’, ‘N1’, ‘N2a’, ‘N2b’, ‘N2c’,…]
Metastasis status (M)One-hot2[‘M0’, ‘M1’]
Clinical stageOne-hot14[‘I’, ‘II’, ‘III’, ‘IVA’, ‘IVB’, ‘IVC’]
Pathological typeOne-hot41[‘SCC’, ‘Adenocarcinoma’,…]
HPV statusOne-hot2[‘Positive’, ‘Negative’]

or Create an Account

Close Modal
Close Modal