Scan Order Research Articles

The appropriateness of sinonasal CT scan orders according to the American College of Radiology Guidelines

The appropriateness of sinonasal CT scan orders according to the American College of Radiology Guidelines

Efficient and Effective Context-Based Convolutional Entropy Modeling for Image Compression.

Precise estimation of the probabilistic structure of natural images plays an essential role in image compression. Despite the recent remarkable success of end-to-end optimized image compression, the latent codes are usually assumed to be fully statistically factorized in order to simplify entropy modeling. However, this assumption generally does not hold true and may hinder compression performance. Here we present contextbased convolutional networks (CCNs) for efficient and effective entropy modeling. In particular, a 3D zigzag scanning order and a 3D code dividing technique are introduced to define proper coding contexts for parallel entropy decoding, both of which boil down to place translation-invariant binary masks on convolution filters of CCNs. We demonstrate the promise of CCNs for entropy modeling in both lossless and lossy image compression. For the former, we directly apply a CCN to the binarized representation of an image to compute the Bernoulli distribution of each code for entropy estimation. For the latter, the categorical distribution of each code is represented by a discretized mixture of Gaussian distributions, whose parameters are estimated by three CCNs. We then jointly optimize the CCNbased entropy model along with analysis and synthesis transforms for rate-distortion performance. Experiments on the Kodak and Tecnick datasets show that our methods powered by the proposed CCNs generally achieve comparable compression performance to the state-of-the-art while being much faster.

A Gradient-Aware Line Sampling Algorithm for LiDAR Scanners

Due to a slow acquisition, the spatial resolution of a depth-acquisition device, such as a Light Detection and Ranging (LiDAR) sensor, is limited, which strictly requires scanning a field of view (FOV) in a particular order. To accurately reconstruct a depth image from limited spatial samples, two-stage sampling has been widely used. However, from the perspective of a LiDAR, two-stage sampling requires scanning the FOV two times to build a depth map, and eventually becomes counterintuitive in practice. To address this problem, this study presents a LiDAR framework and a gradient-aware line-based sampling algorithm. Different from previous works, the proposed scanning algorithm allows a LiDAR to scan the FOV from top to bottom while simultaneously seeking sampling points along object boundary. By utilizing the information of the previous line during scanning, the proposed method maintains the conventional scanning order in a LiDAR, while efficiently predicting the sampling locations of the current line. The experimental results demonstrate that the proposed sampling outperforms grid sampling by at most 9.79 dB on the synthetic datasets. Consequently, the proposed sampling achieves reconstructed quality similar to that of optimal sampling in the previous design, while substantially reducing the computation time and memory requirements. The experimental results with the laser range data and the real data captured by the LiDAR system demonstrate that the proposed method can reduce the averaged mean-absolute-error (MAE) by 34.91%, 47.23%, 54.88%, and 57.99% for the sampling ratios of 20%, 30%, 40%, and 50%, respectively, compared to the conventional LiDAR sampling.

Resource-Efficient Image Buffer Architecture for Neighborhood Processors

Neighborhood image processing operations on Field Programmable Gate Array (FPGA) are considered as memory intensive operations. A large memory bandwidth is required to transfer the required pixel data from external memory to the processing unit. On-chip image buffers are employed to reduce this data transfer rate. Conventional image buffers, implemented either by using FPGA logic resources or embedded memories are resource inefficient. They exhaust the limited FPGA resources quickly. Consequently, hardware implementation of neighborhood operations becomes expensive, and integrating them in resource constrained devices becomes unfeasible. This paper presents a resource efficient FPGA based on-chip buffer architecture. The proposed architecture utilizes full capacity of a single Xilinx Block-RAM (BRAM36 primitive) for storing multiple rows of input image. To get multiple pixels/clock in a user defined scan order, an efficient duty-cycle based memory accessing technique is coupled with a customized addressing circuitry. This accessing technique exploits switching capabilities of BRAM to read 4 pixels in a single clock cycle without degrading system frequency. The addressing circuitry provides multiple pixels/clock in any user defined scan order to implement a wide range of neighborhood operations. With the saving of 83% BRAM resources, the buffer architecture operates at 278 MHz on Xilinx Artix-7 FPGA with an efficiency of 1.3 clock/pixel. It is thus capable to fulfill real time image processing requirements for HD image resolution ( $1080\times 1920$ ) @103 fps.

SUV variability in EARL-accredited conventional and digital PET

BackgroundA high SUV-reproducibility is crucial when different PET scanners are in use. We evaluated the SUV variability in whole-body FDG-PET scans of patients with suspected or proven cancer using an EARL-accredited conventional and digital PET scanner.In a head-to-head comparison we studied images of 50 patients acquired on a conventional scanner (cPET, Ingenuity TF PET/CT, Philips) and compared them with images acquired on a digital scanner (dPET, Vereos PET/CT, Philips). The PET scanning order was randomised and EARL-compatible reconstructions were applied.We measured SUVmean, SUVpeak, SUVmax and lesion diameter in up to 5 FDG-positive lesions per patient. The relative difference ΔSUV between cPET and dPET was calculated for each SUV-parameter. Furthermore, we calculated repeatability coefficients, reflecting the 95% confidence interval of ΔSUV.ResultsWe included 128 lesions with an average size of 19 ± 14 mm. Average ΔSUVs were 6-8% with dPET values being higher for all three SUV-parameters (p < 0.001). ΔSUVmax was significantly higher than ΔSUVmean (8% vs. 6%, p = 0.002) and than ΔSUVpeak (8% vs. 7%, p = 0.03). Repeatability coefficients across individual lesions were 27% (ΔSUVmean and ΔSUVpeak) and 33% (ΔSUVmax) (p < 0.001).ConclusionsWith EARL-accredited conventional and digital PET, we found a limited SUV variability with average differences up to 8%. Furthermore, only a limited number of lesions showed a SUV difference of more than 30%. These findings indicate that EARL standardisation works.Trial registrationThis prospective study was registered on the 31th of October 2017 at ClinicalTrials.cov. URL: https://clinicaltrials.gov/ct2/show/NCT03457506?id=03457506&rank=1.

Appropriate use of CT for patients presenting with suspected renal colic: a quality improvement study

IntroductionCT use for renal colic has increased costs, radiation exposure and frequently does not alter management. Consequently, choosing wisely (CW) recommends avoiding CT imaging of otherwise healthy patients younger than...

Effect of Electronic Clinical Decision Support on Imaging for the Evaluation of Acute Low Back Pain in the Ambulatory Care Setting

To assess the effectiveness of a clinical decision support tool consisting of an electronic medical record best practice alert (BPA) on the frequency of lumbar imaging in patients with acute low back pain in the ambulatory care setting, and to explore why providers order imaging outside of clinical guidelines. On March 23, 2016, we implemented a BPA pop-up alert that informed the ordering physician of the Choosing Wisely recommendation to not order imaging within the first 6 weeks of low back pain in the absence of red flags. We calculated imaging rates 1 year before and after implementation of the BPA. To override the BPA, providers could ignore the alert or explain their rationale for ordering imaging using either preset options or a free-text submission. We tracked preset options and manually reviewed 125 free-text submissions. Significant decreases in both total imaging rate (9.6% decrease; P= 0.02) and magnetic resonance imaging rate (14.9% decrease; P < 0.01) were observed after implementation of the BPA. No change was found in the rates of X-ray or computed tomography scan orders. Almost two-thirds (64%) of the providers used preset options in overriding the BPA, and 36% of the providers entered a free-text submission. Among those providers using a free-text submission, 56% entered a non-guideline-supported rationale. The present study demonstrates the effectiveness of a simple, low-cost clinical decision support tool in reducing imaging rates for patients with acute low back pain. We also identify reasons why providers order imaging outside of clinical guidelines.

A Low Complexity System for Multiple Data Embedding Into H.264 Coded Video Bit-Stream

This article investigates the relative performance of multiple data embedding into H.264 compressed video under two schemes of closed-loop and open-loop methods. In closed-loop, a part of an encoder is modified to embed data within the encoding loop during re-compression, while in the open-loop scenario, a part of decoder is modified to embed data out of decoding loop. It is shown, while for the first instance of embedding, both methods behave almost similarly for all picture types, for later instances of embedding, open-loop outperforms the closed-loop method. Moreover, the required time for embedding and extracting processes of open-loop method is only 1-1.5% of that of the closed-loop counterpart. In both methods, the quality of watermarked video and the data hiding capacity are controlled by the position of last non-zero (LNZ) coefficient in the H.264 zigzag scanning order. However, for B-pictures, the side-effect of embedding distortion is very limited (in the order of 0.002 in terms of SSIM), but for I- and P-pictures it can be significant. Picture degradations in I- and P-frames are alleviated by confining data embedding only into the last block of macroblocks of I-frames and the last P-frame in the GOP, respectively. Finally, it is shown that while with CAVLC type entropy coder, the number of increased bits due to data embedding can vary from 5%-65% of the metadata volume (depending on quality degradation) this value with CABAC coder is only less than 20% of that under CAVLC.

Interactive OCT-Based Tooth Scan and Reconstruction.

Digital dental reconstruction can be a more efficient and effective mechanism for artificial crown construction and period inspection. However, optical methods cannot reconstruct those portions under gums, and X-ray-based methods have high radiation to limit their applied frequency. Optical coherence tomography (OCT) can harmlessly penetrate gums using low-coherence infrared rays, and thus, this work designs an OCT-based framework for dental reconstruction using optical rectification, fast Fourier transform, volumetric boundary detection, and Poisson surface reconstruction to overcome noisy imaging. Additionally, in order to operate in a patient’s mouth, the caliber of the injector is small along with its short penetration depth and effective operation range, and thus, reconstruction requires multiple scans from various directions along with proper alignment. However, flat regions, such as the mesial side of front teeth, may not have enough features for alignment. As a result, we design a scanning order for different types of teeth starting from an area of abundant features for easier alignment while using gyros to track scanned postures for better initial orientations. It is important to provide immediate feedback for each scan, and thus, we accelerate the entire signal processing, boundary detection, and point-cloud alignment using Graphics Processing Units (GPUs) while streamlining the data transfer and GPU computations. Finally, our framework can successfully reconstruct three isolated teeth and a side of one living tooth with comparable precisions against the state-of-art method. Moreover, a user study also verifies the effectiveness of our interactive feedback for efficient and fast clinic scanning.

Fast string matching for DNA sequences

In this paper we propose the Maximal Average Shift (MAS) algorithm that finds a pattern scan order that maximizes the average shift length. We also present two extensions of MAS: one improves the scan speed of MAS by using the scan result of the previous window, and the other improves the running time of MAS by using q-grams. These algorithms show better average performances in scan speed than previous string matching algorithms for DNA sequences.

DCT-based color image compression algorithm using adaptive block scanning

A lossy compression algorithm for still color images is presented. Based on DCT and using adaptive block scanning, the proposed method utilizes a simple technique to encode efficiently the DCT coefficients. The required image quality is guaranteed by using the bisection method to threshold the DCT coefficients of the YCbCr image gotten from the input RGB image. Four scan orders (zigzag, horizontal, vertical and hilbert) are used as adaptive block scanning to read the DCT coefficients from the retained DCT block coefficients. Following a scan order, an index vector is formed by the length of the zero-run sequence that preceded a nonzero DCT coefficient. The lowest value of the four index vectors maximums determines the best scan. Finally, the nonzero DCT coefficients and the index vector for each block are encoded to form the compressed image. The obtained results faced to those reported in recent methods show clearly that the proposed technique achieves high performances.

Image Error Concealment Based on Deep Neural Network

In this paper, we propose a novel spatial image error concealment (EC) method based on deep neural network. Considering that the natural images have local correlation and non-local self-similarity, we use the local information to predict the missing pixels and the non-local information to correct the predictions. The deep neural network we utilize can be divided into two parts: the prediction part and the auto-encoder (AE) part. The first part utilizes the local correlation among pixels to predict the missing ones. The second part extracts image features, which are used to collect similar samples from the whole image. In addition, a novel adaptive scan order based on the joint credibility of the support area and reconstruction is also proposed to alleviate the error propagation problem. The experimental results show that the proposed method can reconstruct corrupted images effectively and outperform the compared state-of-the-art methods in terms of objective and perceptual metrics.

Remote Sensing Image Compression Based on Adaptive Directional Wavelet Transform With Content-Dependent Binary Tree Codec

Remote sensing images provide a wealth of information for a variety of applications, but it is at the expense of huge data. In this paper, we present a novel compression method based on optimum adaptive directional lifting (OADL) with content-dependent binary tree codec. First, the OADL model is designed, which calculates the optimal prediction direction of each image block and performs the weighted directional adaptive interpolation during the process of lifting. The former aims to reduce the edge and texture energy of the non-horizontal and non-vertical directions in the high-frequency subbands, and the latter focuses on preserving the directional characteristics of remote sensing images as much as possible. Second, a binary tree codec with content-based adaptive scanning is introduced, which can provide different scanning orders and scanning manners among and within subbands, respectively. In addition, it can encode more significant coefficients at the same bit rate. Experimental results show that, compared with other scan-based compression methods, the proposed compression method can always provide better coding performance in terms of some evaluation indexes.

A zigzag scanning ordering of four-dimensional Walsh basis for single-pixel imaging

There is a fast imaging method to obtain high-quality reconstruction results, which is named Hadamard transform single-pixel imaging. It is known that the object is sampled and reconstructed through forward and inverse transformation respectively. However, the imaging effect is not ideal in the case of incomplete sampling, because the distribution of the key coefficients of Hadamard transform is not centralized. In this study, we proposed a single-pixel imaging of four-dimensional Walsh vector transform scheme based on zigzag scanning order. Key coefficients are available preferentially because of the four-dimensional Walsh vector transform with the characteristics of energy concentration. Based on the numerical simulations and experimental results, this method can produce clear reconstruction images, even if the number of sampling accounts for 20% of the complete sampling. In our work, the possibility of minimizing sampling times was provided without prior knowledge, which may play an important role in real-time video imaging.

New Method of Image Steganography Based on Particle Swarm Optimization Algorithm in Spatial Domain for High Embedding Capacity

Steganography is a form of technology utilised to safeguard secret data during communication in addition to data repository. Numerous researchers have endeavoured to enhance the performance of steganography techniques through the development of an effective algorithm for the selection of the optimal pixel location within the host image for the concealment of secret bits, for the enhancement of the embedding capacity of the secret data, and for maintaining the visual quality of the host image (stego image) in an accepted rate after the concealment of the secret data. Therefore, steganography is perceived as a challenging task. Thus, the current study proposes a new technique for image steganography based on particle swarm optimisation (PSO) algorithm by using pixel selection for the concealment of a secret image in spatial domain, for the purpose of high embedment capacity. The stego possesses a high level of resistance against a steganalytic attack due to the security provided via image steganography. The function of PSO algorithm is to choose an optimal pixel in grey scale host image for the concealment of secret bits, as the PSO has the ability to achieve an efficient fitness calculation that depends on the cost matrix by dividing the host and secret images into four parts. First of all, the secret bits are modified, which are then embedded within the host image. Several locations in the host image are determined through the order of scanning the host pixels and starting point of the scanning for better least significant bits LSBs of each pixel. The PSO algorithm was utilised to ascertain the ideal initiating point and scanning order. Experimental results show that (1) the average peak signal to noise ratio PSNR value in the benchmark technique based on genetic algorithm for five standard stego images is 45.13%, whereas the result obtained from the recommended technique is 56.60%. (2) The proposed technique has an advantage over the benchmark with a percentage of 33.34%, which encompasses all associated issues within the checklist scenario. Therefore, the performance of the recommended technique is superior over existing techniques.

选区顺序对激光直接制造TC4残余应力及变形的影响

选区顺序对激光直接制造TC4残余应力及变形的影响

Image integrity verification via reversible predictive hiding and elliptic curve Diffie-Hellman

This paper presents a medical image integrity verification algorithm which will be able to detect any changes made in the pixel value or the size of a medical image. At the same time, it also provides a secure way of transmitting images over the public domain. This algorithm can not only ensure integrity of the medical image, but also checks if the sender of the image is authentic, making it useful for archiving medical images and remote medical diagnosis. This algorithm essentially has three modules: hashing, data embedding and image encryption. Initially, hash signatures are extracted from the image and are embedded inside the medical image. In the embedding phase, the image is divided into blocks of uneven size. In the raster scanning order, the blocks are embedded with multiple pixels depending on its smoothness. At the receiver end, the signatures are extracted and it can be used to verify the integrity of the medical images. The effectiveness of the embedding algorithm is verified in terms of peak signal to noise ratio, structural similarity index metrics, correlation coefficient. L2 norm is used for verifying the integrity of the images.

188 Impact of Adult Level 1 Trauma Designation on the Timeliness of Pediatric Emergency Department Computed Tomography Scans

Timeliness in the provision of medical care is an essential aspect of patient safety and health care quality. Given the unpredictable nature of the emergency department (ED) with regards to variation in disease acuity and patient volume, it is essential to gather crucial information in a prompt fashion to aid medical decisionmaking and patient disposition. The adult emergency department of our institution was verified as a Level 1 Trauma Center by the American College of Surgeons in December 2015, 24 years after our pediatric emergency department was recognized as a Level 1 Trauma Center. One of the metrics used as an operational benchmark in the ED is imaging interval (time from order of an imaging test to the availability of results). There is a single computed tomography (CT) scanner located in the emergency department, which is utilized by both the adult and pediatric sections. With an anticipated increase in the adult ED patient volume following its new status, we sought to assess the impact of adult Level 1 Trauma designation on the timeliness of pediatric non-trauma CT scans in an emergency department. Retrospective chart review of patients in the pediatric emergency department who had CT scans was performed. We compared CT scan orders of patients over the same 6-month period prior to and after the adult ED Level 1 Trauma verification, and assessed the time from order to completion of CT scans. The months with historically highest trauma volumes were chosen. All pediatric trauma activations were excluded, as well as duplicate orders and scans performed during EMR downtime. Unpaired student t-tests and Chi-squared tests were used for statistical analyses. There were 218 consecutive CT scans analyzed in 2015, and 193 consecutive CT scans in the same time frame in 2016. About half of the CT scans in both years were head imaging studies, and more males were scanned compared to females. The average time for CT completion prior to adult ED Level 1 verification was 39.5 minutes, compared to 55.5 minutes in the same time frame the next year (p<0.0001). Also, the proportion of children receiving CT scans within 60 minutes from order decreased from 81% in 2015 to 65% in 2016. (p=0.0002; 95% CI 7.4 - 24.4). The new designation of adult Level 1 trauma status in our ED was associated with a significant delay (by more than 40%) in performance of non-trauma pediatric CT scans, and decrease in the likelihood of children receiving CT scans within 60 minutes. These findings may have significant clinical and economic ramifications for pediatric patients and hospitals, and represent the potential "collateral damage" of an institution receiving an upgrade of their adult trauma center to Level 1 status.

Digital Image Steganography Using LSB Substitution, PVD, and EMD

To protect from pixel difference histogram (PDH) analysis and RS analysis, two hybrid image steganography techniques by appropriate combination of LSB substitution, pixel value differencing (PVD), and exploiting modification directions (EMD) have been proposed in this paper. The cover image is traversed in raster scan order and partitioned into blocks. The first technique operates on 2 × 2 pixel blocks and the second technique operates on 3 × 3 pixel blocks. For each block, the average pixel value difference, d, is calculated. If d value is greater than 15, the block is in an edge area, so a combination of LSB substitution and PVD is applied. If d value is less than or equal to 15, the block is in a smooth area, so a combination of LSB substitution and EMD is applied. Each of these two techniques exists in two variants (Type 1 and Type 2) with respect to two different range tables. The hiding capacities and PSNR of both the techniques are found to be improved. The results from experiments prove that PDH analysis and RS analysis cannot detect these proposed techniques.

Efficient Directory Mutations in a Full-Path-Indexed File System

Full-path indexing can improve I/O efficiency for workloads that operate on data organized using traditional, hierarchical directories, because data is placed on persistent storage in scan order. Prior results indicate, however, that renames in a local file system with full-path indexing are prohibitively expensive. This article shows how to use full-path indexing in a file system to realize fast directory scans, writes, and renames. The article introduces a range-rename mechanism for efficient key-space changes in a write-optimized dictionary. This mechanism is encapsulated in the key-value Application Programming Interface (API) and simplifies the overall file system design. We implemented this mechanism in B ε -trees File System (BetrFS), an in-kernel, local file system for Linux. This new version, BetrFS 0.4, performs recursive greps 1.5x faster and random writes 1.2x faster than BetrFS 0.3, but renames are competitive with indirection-based file systems for a range of sizes. BetrFS 0.4 outperforms BetrFS 0.3, as well as traditional file systems, such as ext4, Extents File System (XFS), and Z File System (ZFS), across a variety of workloads.