Statistical shape models for 3D medical image segmentation: a review.

[서평]「Component Software」 - Beyond Object-Oriented Programming -

Medical image analysis

A 2D/3D correspondence building method for reconstruction of a patient-specific 3D bone surface model using point distribution models and calibrated X-ray images.

Instantiation and registration of statistical shape models of the femur and pelvis using 3D ultrasound imaging.

https://www.cs.unc.edu/Research/MIDAG/pubs/papers/Rajamani_MedImageAnal_2007.pdf

Statistical deformable bone models for robust 3D surface extrapolation from sparse data

Background

Minimally invasive surgical interventions performed using computer-assisted surgery (CAS) systems require reliable registration methods for pre-operatively acquired patient anatomy representations that are compatible with the minimally invasive paradigm. The use of brightness-mode ultrasound seems to be promising, if associated devices work in a computationally efficient and fully automatic manner.



Methods

This paper presents a rapid and fully automatic segmentation approach for ultrasound B-mode images capable of detecting echoes from bony structures. The algorithm focuses on the precise and rapid detection of bone contours usable for minimally invasive registration. The article introduces the image-processing scheme and a set-up enabling a direct comparison between manually digitized reference points and the segmented bone contours. The segmentation accuracy was assessed using cadaveric material.



Results

The experimental evaluation revealed results in the same order of magnitude as a pointer-based surface digitization procedure.



Conclusion

The suggested segmentation approach provides a reliable means of detecting bony surface patches in ultrasound images. Copyright © 2007 John Wiley & Sons, Ltd.

Automated bone contour detection in ultrasound B-mode images for minimally invasive registration in computer-assisted surgery-an in vitro evaluation.

This article presents a feasibility and evaluation study for using 2D ultrasound in conjunction with our statistical deformable bone model within the scope of computer-assisted surgery. The final aim is to provide the surgeon with enhanced 3D visualization for surgical navigation in orthopedic surgery without the need for preoperative CT or MRI scans. We unified our earlier work to combine several automatic methods for statistical bone shape prediction and ultrasound segmentation and calibration to provide the intended rapid and accurate visualization. We compared the use of a tracked digitizing pointer and ultrasound for acquiring landmarks and bone surface points for the estimation of two cast proximal femurs.

/pdf/a-comparison-study-assessing-the-feasibility-of-ultrasound-1cxpkro921.pdf

A comparison study assessing the feasibility of ultrasound-initialized deformable bone models.

This paper presents methods based on information filters for solving matching problems with emphasis on real time, or effectively real-time applications. Both applications discussed in this paper deal with ultrasound-based rigid registration in computer-assisted orthopedic surgery. In the first application, the usual workflow of rigid registration is reformulated such that registration algorithms would iterate while the surgeon is acquiring ultrasound images of the anatomy to be operated. Using this effectively real-time approach to registration, the surgeon would then receive feedback in order to better gauge the quality of the final registration outcome. The second application considered in this paper circumvents the need to attach physical markers to bones for anatomical referencing. Experiments using anatomical objects immersed in water are performed in order to evaluate and compare the different methods presented herein, using both 2-D as well as real-time 3-D ultrasound.

Information Filtering for Ultrasound-Based Real-Time Registration

We present a method for evaluating the quality of acetabular fracture reductions in computer-assisted orthopaedic surgery (CAOS), which is analogous to evaluating the accuracy reached by a partial surface matching algorithm. We formulate the problem as a two-step procedure: (1) Characterizing the fracture surfaces and (2) providing a useful quantitative value to assess a fracture reduction procedure or a partial surface matching algorithm. To characterize 3D models of the fracture surfaces, we use clustering in conjunction with two similarity measures that we introduce in this paper. The similarity measures take into account features such as orientation and size of the triangles, as well as connectivity information. Our method for obtaining a quantitative value enables for the approximation of the interstitial volume left between bone fragments resulting from a fracture, and thus allows for the evaluation of the accuracy of the fragments' repositioning.

Haydar Talib

Papers

Statistical deformable bone models for robust 3D surface extrapolation from sparse data

Automated bone contour detection in ultrasound B-mode images for minimally invasive registration in computer-assisted surgery-an in vitro evaluation.

A comparison study assessing the feasibility of ultrasound-initialized deformable bone models.

Information Filtering for Ultrasound-Based Real-Time Registration

Evaluating partial surface matching for fracture reduction assessment