Table 2 Resource efficiency evaluation of the proposed method using the generated carbon emissions, the consumed energy and the equivalent distance a car could travel during the final training (250 epochs).

From: Resource efficient aortic distensibility calculation by end to end spatiotemporal learning of aortic lumen from multicentre multivendor multidisease CMR images

Model

CO\(_2\)eq (g)

Energy (kWh)

Equivalent distance travelled by car (km)

Training time (h:min:s)

Average inference time (ms)

Proposed

2093.571

5.984

17.388

06:46:11

2.768

SOTA24

5785.498

23.184

48.052

35:09:20

7.544

Unpruned31

11031.780

39.574

91.626

63:06:35

12.25

  1. Also shown are the training and average inference times. The experiments were conducted on a workstation with an NVIDIA RTX A6000 (48GB) GPU. Bold face indicates best performance.