Evaluation of Pectus Deformities using Optical Imaging Techniques

1

• Evaluation of Pectus Deformities
  
  using Optical Imaging Techniques
• T. Richardson1, J. Ronsky2, P. Poncet2, R.
  Evison2, D.Sigalet3, D. Kravarusic3
• Department of Civil Engineering, University of
  Calgary 2) Department of Mechanical and
• Manufacturing Engineering, University of Calgary
  3) Alberta Childrens Hospital

Introduction

Current Methods

• Computed Tomography (CT) scan of patients chest
  to obtain a cross sectional image of the
  deformity (Figure 2).
• Doctors calculate the Haller Index from the CT
  scan at the point of maximum intrusion of the
  deformity for Excavatum patients.
• No current standard for evaluating the deformity
  in Carinatum patients.
• CT scans emit high levels of harmful radiation
  therefore cannot be taken regularly.

• Pectus deformities are caused by an excessive
  growth of the ribs and costal cartilage that
  buckles inward (Excavatum) or outward (Carinatum)
  (Figure 1).
• The deformity affects chest wall flexibility and
  hinders chest expansion during inhalation.
• The deformity also affects the performance of
  the heart and lungs during exercise.
• Adolescent patients often have psychological
  problems due to cosmetic issues.

Figure 1 Pectus Excavatum (left)
and Carinatum (right)
Figure 2 CT scan of a PE patient
Purpose The objective of the Pectus group is to
use the proposed Optical Imaging procedure
clinically at the Alberta Childrens Hospital.
In order to be accepted as a method of
evaluation, the Pectus group must obtain results
from the Optical Imaging system that correlate
well with the results from the current CT scan
procedure.
Methods

• Procedure adapted from the Scoliosis protocol.
• The Inspeck Imaging System is used to capture
  and process images to create a 3-dimensional
  reconstruction of the patients torso (Figure 3).
  
• Digitizers capture 1.3 million points of
  geometry and texture.
• Data is collected every month at the Alberta
  Childrens Hospital.
• A qualified nurse places markers on the patients
  that relate internal bony structures to external
  surface landmarks.

(a)
(b)
(c)
(d) (e)
(f) Figure 3 Optical Image Processing
Steps (a) Patient stands in the positioning frame
while 4 Inspeck digitizers capture images of each
side of the torso. (b) Cameras use the Moiré
Fringe Contouring technique and the Inspeck
software assigns 3D coordinates to each point on
the torso surface. (c) FAPS (Fringe Acquisition
and Processing Software) is used to select the
are of interest (torso outline). (d) Inspeck EM
application is used to merge images into a 3D
model. (e) Cross section of interest (greatest
protrusion or indentation) is located. (f) Cross
section is isolated to be further analyzed.

• Results
• 10 patients have had CTs and Optical Images
  taken with HI and HImod calculated from each
• 5 Carinatum patients 5 Excavatum patients
• Results from CT and Optical Imaging correlated
  strongly (Figure 5)

Calculations From the optical image cross
section, the Haller Index (HI) and the modified
Haller Index (HImod) were calculated using the
steps shown in figure 4.

• (a)
  (b) (c)
  (d)
• Figure 4 Haller Index Calculation Steps
• Cross section of maximum protrusion or
  indentation is isolated. (b) Cross section is
  rotated so that the line tangent to the back
  humps is vertical. (c) Outermost points on the
  torso are found and a transverse line is drawn
  connecting them (1). Also, tangent lines are
  drawn to the point of maximum deformity (2) and
  spinous process (3). (d) The following
  dimensions are measured dLa transverse
  diameter of the chest dAP anterior-posterior
  diameter at the point of greatest intrusion to
  the spinous process dAPmod distance from the
  point of greatest intrusion to the transverse
  diameter line.
• The Haller Index and modified Haller Index can be
  calculated using the dimensions
• HI dLa HImod dLa
• dAP dAPmod

(a)
(b) Figure 5 Correlation between (a) Haller
Indices and (b) Modified Haller Indices from
Optical Imaging and CT techniques

• Future Work
• Decrease processing time to under 30 minutes
• Create computer programs to perform calculation
  steps automatically
• Determine indices that can evaluate Scoliosis
  progression in Pectus patients
• Create a simplified format to present information
  to patients and doctors (aesthetic score)

• Conclusion
• Optical Imaging method is non-invasive and can be
  used regularly to monitor patient progress
  through treatment or surgery. This will allow
  doctors to understand the history of Pectus
  disorders.
• To be used clinically, the Optical Imaging
  process must take under 30 minutes. Computer
  programs must be produced to automatically
  perform calculation steps.
• Indices must me found that can evaluate the
  progression of Scoliosis, that often occurs in
  Pectus patients.
• So far, correlations between CT results and
  Optical Imaging results have been strong. More
  data must be processed to further correlate the
  two methods.

Acknowledgements Funding support from Markin
Flanagan USRP NSERC. D. Visser, S. Anderson,
H.Wu, K. Truong, V. Komisar from the U of C
Scoliosis Research Group.