CT scan range estimation using multiple body parts detection: let PACS learn the CT image content
- 218 Downloads
The aim of this study was to develop an efficient CT scan range estimation method that is based on the analysis of image data itself instead of metadata analysis. This makes it possible to quantitatively compare the scan range of two studies.
In our study, 3D stacks are first projected to 2D coronal images via a ray casting-like process. Trained 2D body part classifiers are then used to recognize different body parts in the projected image. The detected candidate regions go into a structure grouping process to eliminate false-positive detections. Finally, the scale and position of the patient relative to the projected figure are estimated based on the detected body parts via a structural voting. The start and end lines of the CT scan are projected to a standard human figure. The position readout is normalized so that the bottom of the feet represents 0.0, and the top of the head is 1.0.
Classifiers for 18 body parts were trained using 184 CT scans. The final application was tested on 136 randomly selected heterogeneous CT scans. Ground truth was generated by asking two human observers to mark the start and end positions of each scan on the standard human figure. When compared with the human observers, the mean absolute error of the proposed method is 1.2 % (max: 3.5 %) and 1.6 % (max: 5.4 %) for the start and end positions, respectively.
We proposed a scan range estimation method using multiple body parts detection and relative structure position analysis. In our preliminary tests, the proposed method delivered promising results.
KeywordsScan range estimation Body parts detection Structural voting Machine learning Pictorial structures Image classification
Conflict of interest
Chunliang Wang and Claes Lundström are both employed by Sectra AB (a company working with PACS development), and the reported study is funded by the company.
- 2.Seifert S, Kelm M, Moeller M, Mukherjee S, Cavallaro A, Huber M, Comaniciu D (2010) Semantic annotation of medical images. In: SPIE medical imaging, pp 762808–762808Google Scholar
- 5.Wang C, Smedby Ö (2014) Automatic multi-organ segmentation in non-enhanced CT datasets using hierarchical shape priors. In: Proceedings of the 22th international conference on pattern recognition (ICPR), StockholmGoogle Scholar
- 6.Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: IEEE Computer Society conference on computer vision and pattern recognition, 2005. CVPR 2005, vol 1, pp 886–893Google Scholar
- 8.Abramson Y, Steux B, Ghorayeb H (2007) Yet even faster (YEF) real-time object detection. Int J Intell Syst Technol Appl 2(2):102–112Google Scholar
- 9.Felzenszwalb PF, Girshick RB, McAllester D, Ramanan D (2010) Object detection with discriminatively trained part-based models. IEEE Trans Pattern Anal Mach Intell 32(9):1627–1645Google Scholar
- 11.Dollár P, Tu Z, Perona P, Belongie S (2009) Integral channel features. In: BMVC, vol 2, p 5Google Scholar
- 12.Criminisi A, Shotton J, Robertson D, Konukoglu E (2011) Regression forests for efficient anatomy detection and localization in CT studies. In: Medical computer vision. Recognition techniques and applications in medical imaging. Springer, pp 106–117Google Scholar
- 14.Zheng Y, Barbu A, Georgescu B, Scheuering M, Comaniciu D (2007) Fast automatic heart chamber segmentation from 3D CT data using marginal space learning and steerable features. In: IEEE 11th international conference on computer vision, 2007. ICCV 2007, pp 1–8Google Scholar
- 15.Zhou XS, Peng Z, Zhan Y, Dewan M, Jian B, Krishnan A, Tao Y, Harder M, Grosskopf S, Feuerlein U (2010) Redundancy, redundancy, redundancy: the three keys to highly robust anatomical parsing in medical images. In: Proceedings of the international conference on multimedia information retrieval, New York, NY, USA, pp 175–184Google Scholar