Focal cartilage defect progression detection : measurement of precision and variation in natural characteristics of cartilage thickness maps derived from 3D MRI data

1
Focal cartilage defect progression detection
measurement of precision and variation in natural
characteristics of cartilage thickness maps
derived from 3D MRI data

• Author Monica Barbu-McInnis, Jose G. Tamez-Pena,
  
• Sara Totterman
• Source IEEE International Symposium on
  Biomedical Imaging
• April 2004 Page(s) 840 - 843 Vol. 1
• Speaker Ren-Li Shen
• Advisor Ku-Yaw Chang

2
Outline

• Introduction
• Materials and methods
• Results
• Conclusion

3
Introduction

• MRI
• Noninvasive imaging modality
• Quantitative evaluation
• Measure OA(OsteoArthritic) progression
• Measure overall cartilage volume
• Not easily reflect focal or local changes
• Quantitatively measuring changes is to compute
  cartilage thickness map

4
Introduction

• Segment
• Compute the tibia medial cartilage thickness map
• Construct an inter-subject atlas
• Describe the shape and natural thickness
  variation
• Analyse

5
Outline

• Introduction
• Materials and methods
• Results
• Conclusion

6
Materials and methods

• Ten subjects
• Five male and five female
• 44.8 years on average
• 5 with no history of OA, 5 with mild OA
• Scanned 5 times
• Using Phillips Gyroscan Intra 1.5T MR imaging
  system
• Images acquired in sagittal plane
• Region growing algorithm
• Segmentation

7
Materials and methods-Computing the thickness
map Inter-subject registration

• Deal of variability
• Cartilage thickness
• Surface area
• Wide range of cartilage shapes

8
Materials and methods-Computing the thickness
map Inter-subject registration

• All maps rotated to a unique coordinate system
• Principal moments of inertia
• Principal axis method
• Ten rotated maps were scaled to a common space

9
Materials and methods-Correlating the thickness
mapsIntra-subject registration

• Calculate origin and orientation of the thickness
  map
• In intra-subject registration
• Assumes the differences in coordinates of
  thickness points likely due to global changes
• Good origin approximation of the common
  coordinate system
• Derived from the centroid to the two thickness
  maps

10
Materials and methods-Correlating the thickness
mapsIntra-subject registration

• Translation vector
• Origin
• Rotation matrix
• Orientation
• Thickness maps have been correlated
• Point by point mean and standard deviation can be
  calculated

11
Outline

• Introduction
• Materials and methods
• Results
• Conclusion

12
Results
13
Results
14
Results
15
Results
16
Outline

• Introduction
• Materials and methods
• Results
• Conclusions

17
Conclusion

• The developed methodology can determine shape and
  thickness variations
• Within the tibia medial cartilage among a set
  population
• Future work
• Planning on applying these methods on a much
  larger population set