Search

Three different computed tomography (CT) reconstruction algorithms: Filtered Back Projection (FBP), Unified Tomographic Reconstruction (UTR) and customized Simultaneous Algebraic Reconstruction Technique (cSART), have been systematically compared and evaluated using experimental data from CT scans of ten fresh mastectomy samples collected at the Imaging and Medical beamline of the Australian Synchrotron. All the scans were collected at the mean glandular dose of 2 mGy, using monochromatic X-rays with 32 keV energy, flat-panel detectors with 0.1 mm pixels and 6 meter distance between the rotation stage and the detector. Paganin's phase retrieval method was used in conjunction with all three CT reconstruction algorithms. The reconstructed images were compared in terms of the objective image quality characteristics, including spatial resolution, contrast, signal-to-noise, and contrast-to-noise ratios. The images were also evaluated by seven experienced medical imaging specialists, rating perceptible contrast, sharpness of tissue interfaces, image noise, calcification visibility and overall image quality. Of the three compared algorithms, cSART was clearly superior to UTR and FBP in terms of most measured objective image quality characteristics. At the same time, the results of the subjective quality evaluation consistently favoured the images reconstructed by FBP, followed by UTR, with cSART receiving lower scores on average. We argue that this apparent disagreement between the objective and subjective assessments of image quality can be explained by the importance assigned to image contrast in the subjective assessment, while the signal-to-noise ratio seemed to receive relatively low weighting. This study was conducted in preparation for phase-contrast breast CT imaging of live patients at Australian Synchrotron (Melbourne, Australia)., Comment: 21 pages, 5 figures, 2 tables

In dusty plasma environments, the spontaneous growth of nanoparticles from reactive gases has been extensively studied for over three decades, primarily focusing on hydrocarbons and silicate particles. Here, we introduce the growth of titanium dioxide, a wide band gap semiconductor, as dusty plasma nanoparticles. The resultant particles exhibited a spherical morphology and reached a maximum homogeneous radius of 230 $\pm$ 17 nm after an elapsed time of 70 seconds. The particle grew linearly and the growth displayed a cyclic behavior; that is, upon reaching their maximum radius, the largest particles fell out of the plasma, and a new growth cycle immediately followed. The particles were collected after being grown for different amounts of time and imaged using scanning electron microscopy. Further characterization was carried out using energy dispersive X-ray spectroscopy, X-ray diffraction and Raman spectroscopy to elucidate the chemical composition and crystalline properties of the maximally sized particles. Initially, the as-grown particles after 70 seconds exhibited an amorphous structure. However, annealing treatments at temperatures of 400 $^\circ$C and 800 $^\circ$C induced crystallization, yielding anatase and rutile phases, respectively. Notably, annealing at 600 $^\circ$C resulted in a mixed phase of anatase and rutile. These findings open new avenues for a rapid and controlled growth technique of titanium dioxide as dusty plasma., Comment: 8 pages, 5 figures

Subject: In this article, convolutional networks of one, two, and three dimensions are compared with respect to their ability to distinguish between the drawing tests produced by Parkinson's disease patients and healthy control subjects. Motivation: The application of deep learning techniques for the analysis of drawing tests to support the diagnosis of Parkinson's disease has become a growing trend in the area of Artificial Intelligence. Method: The dynamic features of the handwriting signal are embedded in the static test data to generate one-dimensional time series, two-dimensional RGB images and three-dimensional voxelized point clouds, and then one-, two-, and three-dimensional CNN can be used to automatically extract features for effective diagnosis. Novelty: While there are many results that describe the application of two-dimensional convolutional models to the problem, to the best knowledge of the authors, there are no results based on the application of three-dimensional models and very few using one-dimensional models. Main result: The accuracy of the one-, two- and three-dimensional CNN models was 62.50%, 77.78% and 83.34% in the DraWritePD dataset (acquired by the authors) and 73.33%, 80.00% and 86.67% in the PaHaW dataset (well known from the literature), respectively. For these two data sets, the proposed three-dimensional convolutional classification method exhibits the best diagnostic performance.

Automatic synthesis of analog and Radio Frequency (RF) circuits is a trending approach that requires an efficient circuit modeling method. This is due to the expensive cost of running a large number of simulations at each synthesis cycle. Artificial intelligence methods are promising approaches for circuit modeling due to their speed and relative accuracy. However, existing approaches require a large amount of training data, which is still collected using simulation runs. In addition, such approaches collect a whole separate dataset for each circuit topology even if a single element is added or removed. These matters are only exacerbated by the need for post-layout modeling simulations, which take even longer. To alleviate these drawbacks, in this paper, we present FuNToM, a functional modeling method for RF circuits. FuNToM leverages the two-port analysis method for modeling multiple topologies using a single main dataset and multiple small datasets. It also leverages neural networks which have shown promising results in predicting the behavior of circuits. Our results show that for multiple RF circuits, in comparison to the state-of-the-art works, while maintaining the same accuracy, the required training data is reduced by 2.8x - 10.9x. In addition, FuNToM needs 176.8x - 188.6x less time for collecting the training set in post-layout modeling., Comment: 8 pages, 13 figures, 8 tables, accepted on International Conference on Computer-Aided Design (ICCAD)

In recent years, deep learning methods have achieved great success in various fields due to their strong performance in practical applications. In this paper, we present a light-weight neural network for Parkinson's disease diagnostics, in which a series of hand-drawn data are collected to distinguish Parkinson's disease patients from healthy control subjects. The proposed model consists of a convolution neural network (CNN) cascading to long-short-term memory (LSTM) to adapt the characteristics of collected time-series signals. To make full use of their advantages, a multilayered LSTM model is firstly used to enrich features which are then concatenated with raw data and fed into a shallow one-dimensional (1D) CNN model for efficient classification. Experimental results show that the proposed model achieves a high-quality diagnostic result over multiple evaluation metrics with much fewer parameters and operations, outperforming conventional methods such as support vector machine (SVM), random forest (RF), lightgbm (LGB) and CNN-based methods.

A method for extracting the dark-field signal in propagation-based phase-contrast imaging is proposed. In the case of objects consisting predominantly of a single material, or several different materials with similar ratios of the real decrement to the imaginary part of the complex refractive index, the proposed method requires a single image for extraction of the dark-field signal in two-dimensional projection imaging. In the case of three-dimensional tomographic imaging, the method needs only one image to be collected at each projection angle. A preliminary example demonstrates that this method can improve the visualization of microcalcifications in propagation-based X-ray breast cancer imaging. It is suggested that the proposed approach may be useful in other forms of biomedical imaging, where it can help one to obtain additional small-angle scattering information without increasing the radiation dose to the sample.

Event cameras are bio-inspired sensors that differ from conventional frame cameras: Instead of capturing images at a fixed rate, they asynchronously measure per-pixel brightness changes, and output a stream of events that encode the time, location and sign of the brightness changes. Event cameras offer attractive properties compared to traditional cameras: high temporal resolution (in the order of microseconds), very high dynamic range (140 dB vs. 60 dB), low power consumption, and high pixel bandwidth (on the order of kHz) resulting in reduced motion blur. Hence, event cameras have a large potential for robotics and computer vision in challenging scenarios for traditional cameras, such as low-latency, high speed, and high dynamic range. However, novel methods are required to process the unconventional output of these sensors in order to unlock their potential. This paper provides a comprehensive overview of the emerging field of event-based vision, with a focus on the applications and the algorithms developed to unlock the outstanding properties of event cameras. We present event cameras from their working principle, the actual sensors that are available and the tasks that they have been used for, from low-level vision (feature detection and tracking, optic flow, etc.) to high-level vision (reconstruction, segmentation, recognition). We also discuss the techniques developed to process events, including learning-based techniques, as well as specialized processors for these novel sensors, such as spiking neural networks. Additionally, we highlight the challenges that remain to be tackled and the opportunities that lie ahead in the search for a more efficient, bio-inspired way for machines to perceive and interact with the world.

Detection of salient objects in image and video is of great importance in many computer vision applications. In spite of the fact that the state of the art in saliency detection for still images has been changed substantially over the last few years, there have been few improvements in video saliency detection. This paper investigates the use of recently introduced non-local neural networks in video salient object detection. Non-local neural networks are applied to capture global dependencies and hence determine the salient objects. The effect of non-local operations is studied separately on static and dynamic saliency detection in order to exploit both appearance and motion features. A novel deep non-local neural network architecture is introduced for video salient object detection and tested on two well-known datasets DAVIS and FBMS. The experimental results show that the proposed algorithm outperforms state-of-the-art video saliency detection methods., Comment: Submitted to Journal of Visual Communication and Image Representation

Deep networks are now able to achieve human-level performance on a broad spectrum of recognition tasks. Independently, neuromorphic computing has now demonstrated unprecedented energy-efficiency through a new chip architecture based on spiking neurons, low precision synapses, and a scalable communication network. Here, we demonstrate that neuromorphic computing, despite its novel architectural primitives, can implement deep convolution networks that i) approach state-of-the-art classification accuracy across 8 standard datasets, encompassing vision and speech, ii) perform inference while preserving the hardware's underlying energy-efficiency and high throughput, running on the aforementioned datasets at between 1200 and 2600 frames per second and using between 25 and 275 mW (effectively > 6000 frames / sec / W) and iii) can be specified and trained using backpropagation with the same ease-of-use as contemporary deep learning. For the first time, the algorithmic power of deep learning can be merged with the efficiency of neuromorphic processors, bringing the promise of embedded, intelligent, brain-inspired computing one step closer., Comment: 7 pages, 6 figures