scholarly journals Automatic Validation of Textual Attribute Values in E-commerce Catalog by Learning with Limited Labeled Data

Author(s):  
Yaqing Wang ◽  
Yifan Ethan Xu ◽  
Xian Li ◽  
Xin Luna Dong ◽  
Jing Gao
Keyword(s):  
2021 ◽  
Vol 27 (1) ◽  
pp. 57-62
Author(s):  
Rakhma Noviliawati ◽  
Choirul Anam ◽  
Heri Sutanto ◽  
Geoff Dougherty ◽  
Muhammad Ridha Mak’ruf

Abstract The purpose of this study was to develop an automatic method for validating the computed tomography gantry tilt. A head polymethyl methacrylate phantom with a diameter of 16 cm was used. Gantry tilt angles were measured both manually and automatically. Manual measurements were performed by measuring the length of the anteroposterior and lateral diameters from acquired images using electronic calipers. Automatic measurements consisted of a number of steps: phantom segmentation, determination of the center of the phantom, measurement of the anteroposterior and lateral diameters, and computation of the gantry tilt angle. The method was implemented on the gantry angles from 0° to 15°. The proposed method of measuring gantry angles produced accurate gantry tilt angles. The differences with the angles displayed on the gantry were less than 1°. The results of the automatic method were the same as those of the manual method (R2 > 0.98).


Sign in / Sign up

Export Citation Format

Share Document