Browsing by Subject "Signal-To-Noise Ratio"

Now showing 1 - 3 of 3

Open Access
A Deep Learning Approach for Rapid and Generalizable Denoising of Photon-Counting Micro-CT Images.
(Tomography (Ann Arbor, Mich.), 2023-07) Nadkarni, Rohan; Clark, Darin P; Allphin, Alex J; Badea, Cristian T
Photon-counting CT (PCCT) is powerful for spectral imaging and material decomposition but produces noisy weighted filtered backprojection (wFBP) reconstructions. Although iterative reconstruction effectively denoises these images, it requires extensive computation time. To overcome this limitation, we propose a deep learning (DL) model, UnetU, which quickly estimates iterative reconstruction from wFBP. Utilizing a 2D U-net convolutional neural network (CNN) with a custom loss function and transformation of wFBP, UnetU promotes accurate material decomposition across various photon-counting detector (PCD) energy threshold settings. UnetU outperformed multi-energy non-local means (ME NLM) and a conventional denoising CNN called UnetwFBP in terms of root mean square error (RMSE) in test set reconstructions and their respective matrix inversion material decompositions. Qualitative results in reconstruction and material decomposition domains revealed that UnetU is the best approximation of iterative reconstruction. In reconstructions with varying undersampling factors from a high dose ex vivo scan, UnetU consistently gave higher structural similarity (SSIM) and peak signal-to-noise ratio (PSNR) to the fully sampled iterative reconstruction than ME NLM and UnetwFBP. This research demonstrates UnetU's potential as a fast (i.e., 15 times faster than iterative reconstruction) and generalizable approach for PCCT denoising, holding promise for advancing preclinical PCCT research.
Open Access
Determining the Likelihood of Variant Pathogenicity Using Amino Acid-level Signal-to-Noise Analysis of Genetic Variation.
(Journal of visualized experiments : JoVE, 2019-01-16) Jones, Edward G; Landstrom, Andrew P
Advancements in the cost and speed of next generation genetic sequencing have generated an explosion of clinical whole exome and whole genome testing. While this has led to increased identification of likely pathogenic mutations associated with genetic syndromes, it has also dramatically increased the number of incidentally found genetic variants of unknown significance (VUS). Determining the clinical significance of these variants is a major challenge for both scientists and clinicians. An approach to assist in determining the likelihood of pathogenicity is signal-to-noise analysis at the protein sequence level. This protocol describes a method for amino acid-level signal-to-noise analysis that leverages variant frequency at each amino acid position of the protein with known protein topology to identify areas of the primary sequence with elevated likelihood of pathologic variation (relative to population "background" variation). This method can identify amino acid residue location "hotspots" of high pathologic signal, which can be used to refine the diagnostic weight of VUSs such as those identified by next generation genetic testing.
Open Access
Optimization of reduced-dose MDCT of thoracic aorta using iterative reconstruction.
(Journal of computer assisted tomography, 2014-01) Töre, Hüseyin Gürkan; Entezari, Pegah; Chalian, Hamid; Gonzalez-Guindalini, Fernanda Dias; Botelho, Marcos Paulo Ferreira; Yaghmai, Vahid
OBJECTIVE: To evaluate the contribution of iterative reconstruction on image quality of reduced-dose multidetector computed tomography of the thoracic aorta. METHODS: A torso phantom was scanned using two tube potentials (80 and 120 kVp) and five different tube currents (110, 75, 40, 20, and 10 mAs). All images were reconstructed with both filtered back projection (FBP) and iterative reconstruction. Aortic attenuation, image noise within the thoracic aorta, signal-to-noise ratio, and sharpness of the aortic wall were quantified in the phantom for the two reconstruction algorithms. Data were analyzed using paired t test. A value of P < 0.05 was considered significant. RESULTS: The aortic attenuation was similar for FBP and iterative reconstruction (P > 0.05). Image noise level was lower (P < 0.0001), and image sharpness was higher (P = 0.046) with iterative reconstruction. Signal-to-noise ratios were higher with iterative reconstruction compared with those with FBP (P < 0.0001). Signal-to-noise ratio at 80 kVp with iterative reconstruction (9.8 ± 4.4) was similar to the signal-to-noise ratio at 120 kVp with FBP (8.4 ± 3.3) (P = 0.196). CONCLUSIONS: Less image noise and higher image sharpness may be achieved with iterative reconstruction in reduced-dose multidetector computed tomography of the thoracic aorta.