We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CV

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Electrical Engineering and Systems Science > Image and Video Processing

Title: Positron Emission Tomography (PET) image enhancement using a gradient vector orientation based nonlinear diffusion filter (GVOF) for accurate quantitation of radioactivity concentration

Abstract: To accurately quantify in vivo radiotracer uptake using Positron Emission Tomography (PET) is a challenging task due to low signal-to-noise ratio (SNR) and poor spatial resolution of PET camera along with the finite image sampling constraint. Furthermore, inter lesion variations of the SNR and contrast along with the variations in size of the lesion make the quantitation even more difficult. One of the ways to improve the quantitation is via post reconstruction filtering with Gaussian Filter (GF). Edge preserving Bilateral Filter (BF) and Nonlinear Diffusion Filter (NDF) are the alternatives to GF that can improve the SNR without degrading the image resolution. However, the performance of these edge preserving methods are only optimum for high count and low noise cases. A novel parameter free gradient vector orientation based nonlinear diffusion filter (GVOF) is proposed in this paper that is insensitive to statistical fluctuations (e. g., SNR, contrast, size etc.). GVOF method applied on the PET images collected with the NEMA phantom with varying levels of contrast and noise reveals that the GVOF method provides the highest SNR, CNR (contrast-to-noise ratio) and resolution compared to the original and other filtered images. The percentage bias in estimating the maximum activity representing SUVmax (Maximum Standardized Uptake Value) for the spheres with diameter > 2cm where the partial volume effects (PVE) is negligible is the lowest for the GVOF method. The GVOF method also improves the maximum intensity reproducibility. Robustness of the GVOF against variation in sizes, contrast levels and SNR makes it a suitable post filtering method for both accurate diagnosis and response assessment. Furthermore, its capability to provide accurate quantitative measurements irrespective of the SNR, it can also be effective in reduction of radioactivity dose.
Subjects: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Cite as: arXiv:2006.00273 [eess.IV]
  (or arXiv:2006.00273v1 [eess.IV] for this version)

Submission history

From: Mahbubunnabi Tamal [view email]
[v1] Sat, 30 May 2020 13:57:02 GMT (2217kb)

Link back to: arXiv, form interface, contact.