Table 3 Sensitivity analysis to measuring the collider on the wrong scale.

From: Eliminating biasing signals in lung cancer images for prognosis predictions with deep learning

Model

Actual \(x\)

Measured \(x^\prime\)

\({\mathrm{MSE}}_{{\mathrm{y}}}\)

\({\mathrm{ATE}}\)

Regression*

Area

Area

1.99

1.00

CausalNet

Area

Area

2.23

1.02

CausalNet

Diameter

Volume

2.24

0.99

CausalNet

Volume

Diameter

2.21

1.02

  1. Mean squared error for survival (MSEy) along with estimated average treatment effect (ATE). The regression* results indicate the optimal results attainable for this simulated scenario