Table 1 Basic characteristics of enrolled subjects, number of CT images from different manufacturers, and number of different diseases.

From: Rapid vessel segmentation and reconstruction of head and neck angiograms using 3D convolutional neural network

Parameters

Patient (images) metric

Patients characteristics

 Number of patients (training/independent testing/clinical evaluation and application data sets)

16,433/1826/152/3430

 Male to female (training/independent testing/clinical evaluation and application data sets)

1.05:1 (6539:6242)/1.07:1 (2831:2647)/1.17:1 (82:70)/1.03:1(1738:1692)

 Age (y) (training/independent testing/clinical evaluation and application data sets)

64 ± 11/62 ± 14/66 ± 13/64 ± 13

Different manufactures

Number of patients in training/testing data sets

 GE

 7071/1011

 Siemens

 5271/523

 Philip

 2556/197

 Toshiba

 1535/95

Different diseases

Number of patients in training/testing/clinical evaluation/application data sets

 Atherosclerosis

 11,503/1223/128/2982

 Cerebrovascular disease

 5752/548/52/1744

 Arterial aneurysm

 679/70/7/176

 Moyamoya disease

 159/17/3/45

 Interventional therapy

 448/46/2/138

 Vascular variation

 247/26/2/59