Technique for manipulating medical images

Details Technique for manipulating medical images Technique for manipulating medical images

2001-09-14

2004-09-28

Smith, Ruth S. (Department: 3737)

Surgery

Diagnostic testing

Detecting nuclear, electromagnetic, or ultrasonic radiation

C382S128000, C382S199000

Reexamination Certificate

active

06799066

ABSTRACT:

BACKGROUND OF THE INVENTION
MR imaging is currently the most accurate technique for assessing the articular cartilage non-invasively in vivo. A large array of different pulse sequences can be used for imaging articular cartilage. However, at present, the diagnosis of cartilage loss is based mostly on qualitative, visual analysis by the radiologist. One of the major obstacles in evaluating patients with osteoarthritis has been the lack of accurate and reproducible quantitative image processing and analysis techniques for monitoring progression of cartilage loss and response to medical or surgical therapy.
Some investigators reported the use of three-dimensional reconstruction of the articular cartilage with subsequent volumetric quantification of the entire cartilage surface. In one study, cartilage was segmented from the surrounding tissues using a signal intensity based thresholding technique applied to magnetization transfer subtraction images or fat-saturated T1-weighted images [1]. Since some of the adjacent tissues demonstrated signal intensity values overlapping those of articular cartilage, additional manual disarticulation of the cartilage was performed on selected image slices. Knees were imaged repeatedly. Cartilage volumes determined from the 3D reconstructions of MT subtraction and T1-weighted fat-saturated images were correlated to values obtained with water displacement of surgically retrieved tissue. They reported an intra-observer reproducibility error of 0.20-0.65 mL (3.6%-6.4%) for MT subtraction imaging and 0.21-0.58 mL (4.2%-6.4%) for T1-weighted fat-saturated imaging [1]. Inter observer error was less than 0.62 mL and 7.8%. In a subsequent study involving the metacarpophalangeal joints, they found a reproducibility error of 5.2% (95% confidence interval 2.9% to 7.6%) for metacarpal cartilage and 9.9% (5.4% to 15.1%) for proximal phalangeal cartilage [2]. They concluded that three-dimensional data analysis of MR images allows reproducible volumetric quantification of articular cartilage in the knee and metacarpophalangeal joints.
Piplani et al. refined this approach by combining signal intensity based thresholding of the articular cartilage with a connected-components or seed growing algorithm thereby obviating the need for manual disarticulation of the cartilage in those areas where adjacent tissues demonstrated overlapping signal intensities [3].
Stammberger et al. used B-spline snakes for semi-automated segmentation of articular cartilage [4]. A continuous and smooth parametric curve that depends on a number of control points is fit around the object by means of minimizing different energy terms. These energy terms control the smoothness of the curve and its attraction to certain image features, e.g. high graylevel gradients, causing it to act much like a rubber band.
Lynch et al. demonstrated a variation of the snake algorithm, in which the spline is adjusted to minimize costs calculated from Gauss and Canny filter responses [5]. The user initializes the system by selecting different control points in the medial and lateral tibio-femoral and the patello-femoral compartments. These control points are subsequently automatically adjusted as far as possible.
However, at present, there are no techniques available that perform reliably when used for segmentation of cartilage affected by advanced osteoarthritis. In these cases, MR images typically show a high degree of texture inhomogeneity of the cartilage, irregular and interrupted contours, and low contrast between the cartilage and surrounding tissue. These situations require a different technique for segmentation of the cartilage.
We developed a system for the calculation of the 3-dimensional cartilage thickness that is based on a 3D Euclidian distance transformation (EDT). For a given set of feature points in a binary volume, the EDT computes the distance to the closest feature point for each non-feature point of the volume. By using the points on the cartilage-bone interface (inner cartilage surface, ICS) as feature points, the EDT measures the distance to the closest voxel on the ICS for all other points, including the ones on the outer cartilage surface (OCS), resulting in a truly three-dimensional distance value determined normal to the ICS.
SUMMARY OF THE INVENTION
The general purpose of the invention and the embodiments described in this invention is to provide new techniques for extracting tissues from medical images. These techniques can be applied to diagnosing arthritis and for monitoring disease progression or response to therapeutic intervention.
In one embodiment, the invention provides for means to estimate the volume of cartilage layers in articular joints such as the knee joint using magnetic resonance imaging (MRI) scans. In another embodiment, the invention provides for means to estimate the thickness distribution of articular cartilage layers using an MRI scan. In another embodiment, the invention provides for means to measure volume and thickness distribution of specific volumes of interest (VOIs) in an MRI scan. In another embodiment, the invention provides for means to compare baseline and follow-up MRI scans of a patient. In another embodiment, the invention provides for means to identify the articular cartilage in an image, such as an MRI. In another embodiment, the invention provides for means to extract the articular cartilage from medical images for analysis purposes.


REFERENCES:
patent: 4655227 (1987-04-01), Gracovetsky
patent: 4699156 (1987-10-01), Gracovetsky
patent: 4813436 (1989-03-01), Au
patent: 4823807 (1989-04-01), Russell et al.
patent: 5099859 (1992-03-01), Bell
patent: 5154178 (1992-10-01), Shah
patent: 5246013 (1993-09-01), Frank et al.
patent: 5320102 (1994-06-01), Paul et al.
patent: 5413116 (1995-05-01), Radke et al.
patent: 5433215 (1995-07-01), Athanasiou et al.
patent: 5445152 (1995-08-01), Bell et al.
patent: 5503162 (1996-04-01), Athanasiou et al.
patent: 5541515 (1996-07-01), Tsujita
patent: 5564437 (1996-10-01), Bainville et al.
patent: 5682886 (1997-11-01), Delp et al.
patent: 5749362 (1998-05-01), Funda et al.
patent: 5772595 (1998-06-01), Votruba et al.
patent: 5779651 (1998-07-01), Buschmann et al.
patent: 5810006 (1998-09-01), Votruba et al.
patent: 5824085 (1998-10-01), Sahay et al.
patent: 5853746 (1998-12-01), Hunziker
patent: 5880976 (1999-03-01), DiGioia III et al.
patent: 5899859 (1999-05-01), Votruba et al.
patent: 5913821 (1999-06-01), Farese et al.
patent: 5928945 (1999-07-01), Seliktar et al.
patent: 5995738 (1999-11-01), DiGioia, III et al.
patent: 6002859 (1999-12-01), DiGioia, III et al.
patent: 6078680 (2000-06-01), Yoshida et al.
patent: 6161080 (2000-12-01), Aouni-Ateshian et al.
patent: 6175655 (2001-01-01), George, III et al.
patent: 6205411 (2001-03-01), DiGioia, III et al.
patent: 6249692 (2001-06-01), Cowin
patent: 6289753 (2001-09-01), Basser et al.
patent: 6310477 (2001-10-01), Schneider
patent: 6310619 (2001-10-01), Rice
patent: 6316153 (2001-11-01), Goodman et al.
patent: 6334066 (2001-12-01), Rupprecht et al.
patent: 6450978 (2002-09-01), Brosseau et al.
patent: 6533737 (2003-03-01), Brosseau et al.
patent: 6560476 (2003-05-01), Pelletier et al.
patent: 2002/0016543 (2002-02-01), Tyler
patent: 2002/0087274 (2002-07-01), Alexander et al.
patent: 2002/0147392 (2002-10-01), Steines et al.
patent: 2003/0015208 (2003-01-01), Lang et al.
patent: WO 02/22014 (2002-03-01), None
Van der Linden et al. “MR imaging of hyaline cartilage of 0.5 T: a quantitative and qualitative in vitro evaluation of three types of sequences.”Skeletal Radiol1998, 27: 297-305.
Herberhold C, et al. “An MR-Based Technique for Quantifying the Deformation of Articular Cartilage During Mechanical Loading in an Intact Cadaver Joint.”Magnetic Resonance in Medicine1998, 39(5): 843-850.
Adam G, et al. “MR Imaging of the Knee: Three-Dimensional Volume Imaging Combined with Fast Processing.”J Comput Asst Tomogr1989 Nov.-Dec.; 13(6): 984-988.
Adams ME, et al. “Quantitative Imaging of Osteoarthritis.”Semin

Affiliated with

Also associated with

Rating

Technique for manipulating medical images does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Technique for manipulating medical images, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Technique for manipulating medical images will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3255939