Table 3 Comparison of predictors for future events.

From: Predicting long-term time to cardiovascular incidents using myocardial perfusion imaging and deep convolutional neural networks

 

AUC

P

MACE

 Clinical traditional risk factors

0.651 (0.559–0.743)

0.060

 Disease vessels numbers

0.557 (0.459–0.655)

0.304

 Combined traditional risk + Disease vessel number

0.653 (0.561–0.745)※

0.094

 MPI-AI risk score

0.727 (0.634–0.821)*Ψ※

 < 0.0001

Total CV events

 Clinical traditional risk factors

0.620 (0.544–0.695)

0.098

 Disease vessels numbers

0.577 (0.504–0.650)

0.058

 Combined traditional risk + Disease vessel number

0.636 (0.562–0.711)*※

0.075

 MPI-AI risk score

0.747 (0.679–0.814)*Ψ※

 < 0.0001

  1. Clinical risk factors included age, gender, history of hypertension and diabetes.
  2. *Indicates P < 0.05 compared with clinical risk factors.
  3. ΨIndicates p < 0.05 compared with combined traditional risk factors + disease vessel numbers.
  4. ※Indicates p < 0.05 compared with disease vessel numbers.
Back to article page