Robot-assisted mirror ultrasound scanning for deep venous thrombosis detection using RGB-D sensor

Bo Meng, Ying ying Zhao, Lei Chen, Fereshteh Aalamifar, Xue jun Liu, Emad Boctor

Research output: Contribution to journalArticlepeer-review

7 Scopus citations


Deep Venous Thrombosis (DVT) is a major cause of morbidity and mortality. Manually scanning with Ultrasound (US) probe brings heavy work load for the sonographers. This paper proposes a novel “Mirror robotic US scanning system. The system is composed of two-arm robot, linear US probes for master and slave side, Kinect sensor as a vision servo. On the master side, sonographers hold one robot arm and operate probe to inspect one leg. On the slave side, the robot follows the master probe on the navigation of Kinect and scans the other leg. 3D images of legs are segmented and register to get mirror matrix. Both Clustered Viewpoint Feature Histogram (CVFH) descriptors of segmented probe and CAD (Computer-Aided Design) training data were calculated to probe recognition. The leg phantom platform was built up. The mirror matrixes were obtained. The correlation coefficients between the two legs are calculated. Times of ICP (Iterative Closest Point) have been calculated for the platform. Results from the initial experiment indicate the idea is feasible and promising greatly by improving the inspection efficiently. Clinically, the method can be implemented for pre-operative procedures to predict the risk of DVT, and this may improve the US scanning efficacy.

Original languageEnglish (US)
Pages (from-to)14247-14261
Number of pages15
JournalMultimedia Tools and Applications
Issue number22
StatePublished - Nov 1 2016


  • CAD model
  • Camera calibration
  • ICP registraion
  • RGB-D sensor
  • Robot
  • Vision servo

ASJC Scopus subject areas

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications


Dive into the research topics of 'Robot-assisted mirror ultrasound scanning for deep venous thrombosis detection using RGB-D sensor'. Together they form a unique fingerprint.

Cite this