Non-local means filter denoising for DEXA images

Abstract

Dual high and low energy images of Dual Energy X-ray Absorptiometry (DEXA) suffer from noises due to the use of weak amount of X-rays. Denoising these DEXA images could be a key process to enhance and improve a Bone Mineral Density (BMD) map which is derived from a pair of high and low energy images. This could further improve the accuracy of diagnosis of bone fractures, osteoporosis, and etc. In this paper, we present a denoising technique for dual high and low energy images of DEXA via non-local means filter (NLMF). The noise of dual DEXA images is modeled based on both source and detector noises of a DEXA system. Then, the parameters of the proposed NLMF are optimized for denoising utilizing the experimental data from uniform phantoms. The optimized NLMF is tested and verified with the DEXA images of the uniform phantoms and real human spine. The quantitative evaluation shows the improvement of Signal-to-Noise Ratio (SNR) for the high and low phantom images on the order of 30.36% and 27.02% and for the high and low real spine images on the order of 22.28% and 33.43%, respectively. Our work suggests that denoising via NLMF could be a key preprocessing process for clinical DEXA imaging.

Hand Gesture Recognition Using Single Patchable Six-Axis Inertial Measurement Unit via Recurrent Neural Networks

“…Using RNN-GRU, about 1% of the V and X shapes we gate recurrent unit (GRU) re misclassified with the public data and 3% to 8% with the collected data. To increase the performance metrics (i.e., accuracy, precision, and recall) with our patchable IMU, the RNN-based models can be fine-tuned with our collected data [ 21 , 54 ].…”

Object and Instance Detection Within Image Scenes

“…YOLO is a unified system that is able to detect the potential region of interests (ROIs) from an entire whole image and directly predict their class probabilities [5]. It posses a single neural network that uses features from an entire image to predict bounding boxes across all classes of the image.…”

Breast mass detection in digital mammography based on anchor-free architecture

“…Because of the clear imaging and sensitivity to early breast masses, mammography images have become the preferred imaging diagnostic method [6]. However, current examination of mammography images is mainly dependent on the subjective experience of physicians, which would lead to the missed detection and misdetection due to visual fatigue and loss of attention [7][8][9]. To effectively avoid the cumbersome manual labeling and the variability of detection results, the development of a robust breast mass automatic detection system has important clinical significance [10].…”