Table 1 Radiomics risk score model for prognosis and the associated radiomics features.

From: Radiomics-guided deep neural networks stratify lung adenocarcinoma prognosis from CT scans

Category

Feature

Cox-LASSO coefficient

Shape

Maximum 2D dameter

0.1581

GLCM

Autocorrelation

0.0409

GLCM

IMC2

āˆ’0.0050

GLSZM

Large area emphasis

āˆ’0.0067

Margin

CDF slope skewness

0.1930

  1. GLCM gray-level co-occurrence matrix, GLSZM gray-level size-zone matrix, IMC informational measure of correlation, CDF cumulative distribution function. The optimal penalty is the penalty term of the Cox-LASSO model.
Back to article page