iDEA: Drexel E-repository and Archives >
Drexel Theses and Dissertations >
Drexel Theses and Dissertations >
Novel automated motion compensation algorithm for producing cumulative maximum intensity images from subharmonic ultrasound imaging of breast lesions
Please use this identifier to cite or link to this item:
|Title: ||Novel automated motion compensation algorithm for producing cumulative maximum intensity images from subharmonic ultrasound imaging of breast lesions|
|Authors: ||Dave, Jaydev Kardam|
|Keywords: ||Biomedical engineering|
|Issue Date: ||29-Oct-2008|
|Abstract: ||Objective: To develop a novel automated motion compensation algorithm for producing cumulative maximum intensity (CMI) images from Sub-Harmonic Imaging (SHI) of breast lesions. To compare CMI image processing techniques in SHI mode for better visualization of lesions and the associated vascularity.
Methods and Materials: In CMI technique, a composite image depicting vascular architecture and blood flow is constructed through maximum intensity projection of image data over consecutive images. SHI data (transmitting/receiving at 4.4/2.2 MHz) was obtained from 16 breast lesions using a modified Logiq 9 scanner (GE Healthcare, Milwaukee, WI). Manual method involved aligning the individual frames based on visually matching the common regions in each frame. The automated technique proposed uses a block matching algorithm on the above acquired frames. The user selects a rectangular region of interest – kernel, in the first frame. The algorithm then employs sum of absolute difference (SAD) technique to scan all the remaining frames in order to identify motion induced shifting of this kernel. For each frame the displacement of the kernel is stored and the image is then added to the cumulative image after compensating for this displacement. The reliability of the displacement calculated in each frame is estimatedon the basis of a test statistic: the reliability parameter (RP) defined as the ratio of minimum SAD to average of all SAD values. Different threshold levels were used to eliminate image frames with very low parameter values i.e. noisy frames from subsequent processing.SHI data from 16 lesions was processed manually and by the automated technique, using 3 different threshold values for the RP to reduce motion artifacts. In every case an image of peak contrast flow was chosen as control. Six blinded and independent readers scored all the randomized images for vessel continuity, detail resolution, presence of artifacts, overall image quality and SNR on a 7 point scale (poor-excellent). Following the initial study, readers also ranked all images within each case on a scale from 1-5 (best-worst). Scores were compared using double, repeated measures ANOVA.
Results: The processing techniques were significantly different with regards to vessel continuity, detail resolution and image quality (p<0.001). The single (control) frame was scored significantly worse compared to the manual and automated CMI techniques. For all parameters assessed significant differences were observed between users (p<0.001). One of the users differed markedly from the rest, but repeating the analysis without this user yielded similar results. The rank obtained by each of the techniques (control: 3.85±1.40; manual CMI: 3.45±1.31; automatic CMI 2.70±1.46 to 2.97±1.35; averaged over all users with lower scores being better) was also significantly different (p<0.025).
Conclusion: CMI processing techniques visualize vascularity and produce better image quality then the best single SHI frame. Moreover, automated techniques save processing time, eliminate user bias and are more reproducible than the manual method. The CMI images obtained seem to be a valuable tool to assess breast lesions vascularity. The robust reconstructed image can yield hidden data which may not be evident on viewing the video file. However more patient studies are required to validate the use of this novel technique in breast lesions diagnosis. The usefulness of this algorithm can be further extended to other imaging modalities.
Relevance: CMI processing of SHI data improve the depiction of breast tumor vascularity and may in the future assist in the characterization of breast lesions.|
|Appears in Collections:||Drexel Theses and Dissertations|
Items in iDEA are protected by copyright, with all rights reserved, unless otherwise indicated.