International Journal of Scientific & Technology Research

IJSTR@Facebook IJSTR@Twitter IJSTR@Linkedin
Home About Us Scope Editorial Board Blog/Latest News Contact Us

IJSTR >> Volume 5 - Issue 6, June 2016 Edition

International Journal of Scientific & Technology Research  
International Journal of Scientific & Technology Research

Website: http://www.ijstr.org

ISSN 2277-8616

Optimization Of Scan Range For 3d Point Localization In Statscan Digital Medical Radiology

[Full Text]



Jacinta S. Kimuyu



Optimization, Lodox Statscan, Radiology, Modality, Localization, DLT.



The emergence of computerized medical imaging in early 1970s, which merged with digital technology in the 1980s, was celebrated as a major breakthrough in three-dimensional (3D) medicine. However, a recent South African innovation, the high speed scanning Lodox Statscan Critical Digital Radiology modality, posed challenges in X-ray photogrammetry due to the system’s intricate imaging geometry. The study explored the suitability of the Direct Linear Transformation as a method for the determination of 3D coordinates of targeted points from multiple images acquired with the Statscan X-ray system and optimization of the scan range. This investigation was carried out as a first step towards the development of a method to determine the accurate positions of points on or inside the human body. The major causes of errors in three-dimensional point localization using Statscan images were firstly, the X-ray beam divergence and secondly, the position of the point targets above the X-ray platform. The experiments carried out with two reference frames showed that point positions could be established with RMS values in the mm range in the middle axis of the X-ray patient platform. This range of acceptable mm accuracies extends about 15 to 20 cm sideways towards the edge of the X-ray table and to about 20 cm above the table surface. Beyond this range, accuracy deteriorated significantly reaching RMS values of 30mm to 40 mm. The experiments further showed that the inclusion of control points close to the table edges and more than 20 cm above the table resulted in lower accuracies for the L - parameters of the DLT solution than those derived from points close to the center axis only. As the accuracy of the L - parameters propagates into accuracy of the final coordinates of newly determined points, it is essential to restrict the space of the control points to the above described limits. If one adopts the usual approach of surrounding the object by known control points, then the limited space with an acceptable accuracy potential for the L - terms would not be large enough to enclose an adult human body surrounded by suitably positioned control points. This shortcoming can be overcome by making use of two further observations made in the course of this investigation. These observations were firstly, that the best image orientation angles are 00 and 400 to 600, and secondly, that no significant improvement could be achieved when using more than two images. This observation contradicts the theory of adjustment and observations, and can be investigated in further research. The possible observation method deduced from this is as follows: First, a frame with well distributed control points with accurate 3D coordinates and of approximately the size of a human body is placed on the X-ray table and imaged with the X-ray beam in the 0 degree position. This makes it possible to determine L - parameters for this ray orientation; 2. The frame is removed; the patient is positioned in the control space; and an X-ray image of the patient is taken 3. The X-ray source is rotated to a new position between 400 and 600 and a second image of the patient is acquired and fourth, the patient is removed and replaced by the frame. A final image of the frame is now acquired. Steps 1 and 4 serve to determine the L-parameters for the two X-ray source positions, while steps 2 and 3 provide the image coordinates of the required object points on or inside the patient’s body. This approach can only then result in accurate point positions, if the patient remains motionless for the duration of steps 2 and 3. An alternative to this observation design would be simultaneous imaging from two X-ray sources, one with 00 orientations and the other with an orientation between 400 and 600.



[1] Fraser, C.S., (2001); Australis User Manual, (University of Melbourne).

[2] Karara, H.M. (1989); Non-Topographic Photogrammetry (2nd Edition).

[3] Kasser, M. and Egels, Y., (2002); Digital Photogrammetry.

[4] Lodox website; http://www.lodox.com/.

[5] Marzan, G.T. and Karara, H.M., (1976, January); “Rational Design for Close Range Photogrammetry”; A report on a study sponsored by the National Science Foundation – as a part of research Grant GK-11655; Civil Engineering Studies-Photogrammetry Series No. 43; University of Illinois at Urbana-Chamaign (Urbana, Illinois 61801), pp 156-185.

[6] Mikhail, E.M., Bethel, J.S. and McGlone, J.C., (2001); Introduction to Modern Photogrammetry.

[7] Rüther, H., (2005, APG313S); Numerical Methods in Geomatics lecture notes, (unpublished).