2442

View Abstract / View Presentation

Deep learning-based DWI Denoising method that suppressed the "instability" problem

Hayato Nozaki^1,2, Yasuhiko Tachibana³, Yujiro Otsuka⁴, Wataru Uchida^1,2, Yuya Saito¹, Koji Kamagata¹, and Shigeki Aoki¹
¹Department of Radiology, Graduate School of Medicine, Juntendo University, Tokyo, Japan, ²Graduate School of Human Health Sciences, Tokyo Metropolitan University, Tokyo, Japan, ³Department of Molecular Imaging and Theranostics National Institute of Radiological Sciences National Institutes for Quantum and Radiological Science and Technology, Chiba, Japan, ⁴Miliman, Tokyo, Japan

The deep learning-based method which can effectively denoise DWI images almost without risked to output outliers due to the instability problem in deep learning was developed and evaluated.

Figure 1. Outline of the neural network architecture.

The output value has small risk to become an outlier because it is generated by the combination of the weighted averages for the neighboring pixels in the original image. The loss to be minimized consists of both the mean absolute error between the output and the target images and the Euclid distance between the derived diffusion tensors for efficient optimization.

Figure 4. The results of the ROI-based analysis.

dNR was closer to NEX8 than NEX1 in most regions and parameter maps, and some significant differences between NEX1 and NEX8 were resolved in between dNR and NEX8. However, dNR was far from NEX8 than NEX1 in some combinations of the region and the parameter. Especially, in the regions of deep white matter and periventricular white matter in ODI, the difference compared to NEX8 was significant in dNR and not in NEX1.