Automatic heart positioning method in computed tomography scout images

Hong Li1, Kaihua Liu2, Hang Sun2

  • 1School of Sino-Dutch Biomedical & Information Engineering, Northeastern University, Shenyang 110819, China Key Laboratory of Medical Imaging Computing (Ministry of Education), Northeastern University, Shenyang 110819, China.

Insights

This study presents an automatic method for heart positioning in computed tomography (CT) scout images, crucial for reducing radiation dose via region of interest (ROI) CT scans. The technique accurately locates the heart in both anteroposterior and lateral views, aiding targeted cardiac imaging.

Area of Science:

  • Medical Imaging
  • Radiology
  • Computational Anatomy

Background:

  • Computed tomography (CT) radiation dose reduction is critical for patient safety.
  • Region of interest (ROI) CT scans offer a method to minimize radiation exposure.
  • Accurate automatic heart positioning in scout images is a prerequisite for ROI cardiac CT.

Purpose of the Study:

  • To develop and validate a fully automatic method for heart positioning in both anteroposterior (A-P) and lateral CT scout images.
  • To enable precise localization of the heart for subsequent ROI CT scans, thereby reducing patient radiation dose.

Main Methods:

  • Heart positioning in A-P scout images involves identifying feature points and boundary segments.
  • Polar coordinate transformation and slant elliptic equation curve fitting are used to determine the complete heart boundary.
  • Top and bottom heart boundaries from A-P images are utilized for positioning in lateral scout images.

Main Results:

  • The proposed automatic heart positioning method was evaluated on a dataset of 30 clinical routine cases.
  • Successful and accurate heart positioning was achieved in 26 out of 30 cases (86.7%) for both A-P and lateral scout images.
  • The method demonstrated high efficacy in localizing the heart for potential ROI CT applications.

Conclusions:

  • The developed fully automatic heart positioning technique is effective for CT scout images.
  • This method shows significant potential for facilitating ROI CT scans of the heart.
  • Successful heart localization contributes to reduced radiation dose in cardiac CT imaging.