Quality Of Medical Images Research Articles

Image Reconstruction Quality and Parameters in Tomosynthesis

The effect of tomosynthesis reconstruction parameters on the quality of medical X-ray images is considered.

Medical Image Quality Assessment Using CSO Based Deep Neural Network.

This manuscript proposed a hybrid method of Deep Neural Network (DNN) and Cuckoo Search Optimization (CSO) with No-Reference Image Quality Assessment (NR-IQA) for achieving high accuracy, low computational complexity, flexibility and etc. of a medical image. NR-IQA is proposed due to till now there is no perfect reference image for finding the quality of real time medical imaging. It is an effective method for assessing the real-world medical images. The proposed method takes the distorted image as an input and estimate the quality of the image without the assistance of reference image. The techniques CSO and DNN with NR-IQA produces the quality of the image with high quality score and low Mean Square Error (MSE). Also, the proposed method is used to improve the quality score thereby improving the quality of the image. So that the resultant image has good visual properties which is useful for the analysis of further medical proceedings. The simulation result shows that the proposed system improves the quality score by 8% when compared to the other existing systems. The SROCC value can be increased as 6%, 14%, 6 and 2% for the different existing methods such as NR-BIQA, SBVQP-ML, PTQL/PTVC and NR-SIQA (3D) respectively.

Adaptive Real-Time Removal of Impulse Noise in Medical Images

Noise is an important factor that degrades the quality of medical images. Impulse noise is a common noise caused by malfunctioning of sensor elements or errors in the transmission of images. In medical images due to presence of white foreground and black background, many pixels have intensities similar to impulse noise and hence the distinction between noisy and regular pixels is difficult. Therefore, it is important to design a method to accurately remove this type of noise. In addition to the accuracy, the complexity of the method is very important in terms of hardware implementation. In this paper a low complexity de-noising method is proposed that distinguishes between noisy and non-noisy pixels and removes the noise by local analysis of the image blocks. All steps are designed to have low hardware complexity. Simulation results show that in the case of magnetic resonance images, the proposed method removes impulse noise with an acceptable accuracy.

Effect of morphological algorithms on medical imaging

Handling and improving the quality of medical images with the help of computer software is one of the important stages in the diagnosis and treatment. In this article, we focus on describing the new morphological algorithms by ITK (Insight Segmentation and Registration Toolkit). These morphological operators eliminate noise, detect good edges, and overcome the drawback of traditional edge detection methods.

Restoration of Full Data from Sparse Data in Low-Dose Chest Digital Tomosynthesis Using Deep Convolutional Neural Networks.

Chest digital tomosynthesis (CDT) provides more limited image information required for diagnosis when compared to computed tomography. Moreover, the radiation dose received by patients is higher in CDT than in chest radiography. Thus, CDT has not been actively used in clinical practice. To increase the usefulness of CDT, the radiation dose should reduce to the level used in chest radiography. Given the trade-off between image quality and radiation dose in medical imaging, a strategy to generating high-quality images from limited data is need. We investigated a novel approach for acquiring low-dose CDT images based on learning-based algorithms, such as deep convolutional neural networks. We used both simulation and experimental imaging data and focused on restoring reconstructed images from sparse to full sampling data. We developed a deep learning model based on end-to-end image translation using U-net. We used 11 and 81 CDT reconstructed input and output images, respectively, to develop the model. To measure the radiation dose of the proposed method, we investigated effective doses using Monte Carlo simulations. The proposed deep learning model effectively restored images with degraded quality due to lack of sampling data. Quantitative evaluation using structure similarity index measure (SSIM) confirmed that SSIM was increased by approximately 20% when using the proposed method. The effective dose required when using sparse sampling data was approximately 0.11mSv, similar to that used in chest radiography (0.1mSv) based on a report by the Radiation Society of North America. We investigated a new approach for reconstructing tomosynthesis images using sparse projection data. The model-based iterative reconstruction method has previously been used for conventional sparse sampling reconstruction. However, model-based computing requires high computational power, which limits fast three-dimensional image reconstruction and thus clinical applicability. We expect that the proposed learning-based reconstruction strategy will generate images with excellent quality quickly and thus have the potential for clinical use.

Employment of the appropriate range of sawtooth-shaped-function illumination intensity to improve the image quality

The medical image quality determines the accuracy of diagnosis. To acquire higher signal-to-noise ratio (SNR) and higher quality image, based on the sawtooth-shaped-function (SSF) illumination technique and frame-accumulation algorithm, this paper presents a method of selecting the appropriate range of SSF illumination intensity. By avoiding the weak light intensity and strong light intensity near the camera saturation area, an appropriate range of illumination intensity is gotten, where the SNR of image is higher. The experimental results show that, compared with the previous SSF illumination and constant illumination methods, this proposed method can obtain the image with higher SNR and gray-scale resolution that is to obtain higher quality image. Moreover, this illumination method provides a reference for improving the medical image quality from the point of image acquisition.

Image Reconstruction Parameters in Dual Energy Radiography

The effect of reconstruction parameters used in dual energy radiography on the quality of medical X-ray imaging is considered.

Brain MRI imaging mechanism based on deep visual information perception and dementia degree induction

The traditional medical image recognition methods are limited by image resolution, image brightness and color processing parameters, and image quality evaluation is low. In particular, the incomplete visual information of medical images and the disorder of color structure make the complexity of human visual perception and recognition significantly increased and the accuracy is poor. In order to solve the above problems, this paper is based on the mechanism of deep brain information perception and dementia induced brain magnetic resonance imaging (BMI-DVDI). On the one hand, based on the depth fusion of the visual information system, the medical image depth vision system and its perception model with high precision and low complexity are designed for the two damage of medical image quality and the perception of visual information. On the other hand, the dementia model is designed by means of matrix representation of dementia image signal, screening of dementia sensing brain signal and image reconstruction. The model is helpful to solve the problems of image signal deformation, measurement precision of signal degree and reconstruction of image enhancement in brain magnetic resonance imaging. This model enhances the accuracy of brain diseases such as dementia. Then, we combine the sensing algorithm with the degree of dementia in the brain, and apply it to the MRI of the brain. Finally, through simulation experiments and nuclear magnetic resonance imaging experiments, the space complexity, time complexity, system execution efficiency and image quality evaluation are compared. The result is that the proposed algorithm has excellent performance.

Повышение качества медицинских изображений.

Повышение качества медицинских изображений.

Contrast Enhancement of Medical Images through Adaptive Genetic Algorithm (AGA) over Genetic Algorithm (GA) and Particle Swarm Optimization (PSO)

Assessment of images after processing is a significant step for determining how good the images are being analyzed. Quality of image is usually estimated with the help of image quality metrics. Unfortunately, most of the commonly used metrics cannot sufficiently portray the visual aspect of the enhanced image. In this proposed system, an approach for medical image enhancement is presented. Here adaptive genetic algorithm is proposed for medical image contrast enhancement. Initially, the chromosomes having gene value of the image gray levels have been generated. After that the fitness function will be calculated for each generated chromosome based on the image edge and their overall intensity values. The selected best chromosomes which have the high fitness value will be given to crossover and mutation operation. In GA the adaptive property is introduced by including adaptive crossover and mutation operations. The proposed method is compared with two different types of optimization algorithms such as Genetic algorithm (GA) and Particle swarm optimization (PSO) that ensure accuracy and quality of medical images in proposed adaptive genetic algorithm (AGA). The experimental solutions are got with the help of metrics like PSNR, SDME, MSE, SSIM, MSSIM, AD, MD, NAE, PSO and SC which proves the proposed algorithm, produces better results as compared to the existing algorithms.

A Novel Edge Preserving Improved Adaptive Wavelet Filter (EPIAWF) for Speckle Removal in Ultrasound Images

Background: The paper proposes a new method for improving the quality of ultrasound medical images using a two stage wavelet based denoising approach. The first stage utilizes an adaptive wavelet shrinkage function with a novel interscale measure based variance estimate and a new subband adaptive threshold for reducing the speckle present in ultrasound images. Methods: The performance of the first stage is further improved in the second by employing an error measure estimate. Thus the proposed denoising approach is optimized in the mean square error sense and results in improved noise removal as well as in the edge preservation of the reconstructed image. It also improves the visual quality of the ultrasound images. Next, quantitative evaluation is also done with performance metrics like Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Structural Similarity Index Measure (SSIM), Equivalent Number of Looks (ENL) and Edge preservation Index (EPI). Results: The performance of the proposed approach is measured and compared with the existing approaches and found to be better in improving the quality of ultrasound images. Keywords: Ultrasound images, speckle, adaptive filter, undecimated, wavelet, edge preservation.

Medical images Compression using convolutional neural network with LWT

In compression of medical image using convolutional neural network trained with the back-propagation algorithm and lefted wavelet transformation is proposed to compress high quality medical images. It gives much better result as compared to feed-forward neural network . Medical image processing process is one of the most important section of research in medical applications in digital medical information. In this new approach , a three hidden layer convolutional network (CNN) is applied directly as the main compression algorithm to compress an MRI, X-ray, computer tomography images. After training with sufficient sample images, the compression process will be carried out on the target image. The coupling weights and activation values of each neuron in the hidden layer will be stored after training. Compression is achieved by using smaller number of hidden neurons as compare to the number of image pixels due to lesser information being stored. experimental results proves that the anticipated algorithm is superior to another algorithm in both lossy and lossless compression for all medical images tested Experimental results show that the CNN is able to achieve comparable compression performance to popular existing medical image compression schemes such as JPEG2000 and JPEG-LS. The Wavelet-SPIHT algorithm provides PSNR very important values for MRI and CT scan images.

Adoption of high technology medical imaging and hospital quality and efficiency: Towards a conceptual framework.

Measuring the value of medical imaging is challenging, in part, due to the lack of conceptual frameworks underlying potential mechanisms where value may be assessed. To address this gap, this article proposes a framework that builds on the large body of literature on quality of hospital care and the classic structure-process-outcome paradigm. The framework was also informed by the literature on adoption of technological innovations and introduces 2 distinct though related aspects of imaging technology not previously addressed specifically in the literature on quality of hospital care: adoption (a structural hospital characteristic) and use (an attribute of the process of care). The framework hypothesizes a 2-part causality where adoption is proposed to be a central, linking factor between hospital structural characteristics, market factors, and hospital outcomes (ie, quality and efficiency). The first part indicates that hospital structural characteristics and market factors influence or facilitate the adoption of high technology medical imaging within an institution. The presence of this technology, in turn, is hypothesized to improve the ability of the hospital to deliver high quality and efficient care. The second part describes this ability throughout 3 main mechanisms pointing to the importance of imaging use on patients, to the presence of staff and qualified care providers, and to some elements of organizational capacity capturing an enhanced clinical environment. The framework has the potential to assist empirical investigations of the value of adoption and use of medical imaging, and to advance understanding of the mechanisms that produce quality and efficiency in hospitals.

A New Blind Medical Image Watermarking Based on Weber Descriptors and Arnold Chaotic Map

Protecting the personal patient’s information in distributed health infrastructures seems to be a crucial task. As a solution of this issue, image watermarking is widely used to secure and prevent the content alteration. Moreover, it is necessary to find a new blind watermarking technique way that could decrease the latency and preserve the quality of medical images. This paper presents a new watermarking technique for medical image. Our technique consists in combining the DCT transform, Weber descriptors (WDs) and Arnold chaotic map. This combination brings three significant steps. First, the watermark image is scrambled using Arnold chaotic map. Second, the DCT is performed on each medical image block, and the watermark data are embedded in the DCT middle- band coefficients of each block. Finally, a new embedding and extracting technique is proposed, based on WDs without any loss by selecting the right coefficients. We improve the robustness of the proposed algorithm against several scenarios of attacks such as noising, filtering and JPEG compression. The obtained results make our algorithm eligible to be practicable.

Measuring image sharpness using modulation transfer function in magnetic resonance imaging

Background/Objectives: The purpose of this study was to propose a method to maintain the objectivity and validity of measuring image sharpness by changes of ETL using MTF in MRI quantitatively and provide fundamental data for future evaluation and management of magnetic resource imaging quality.Methods/Statistical analysis: We conducted phantom test using ACR MRI phantom. ImageJ and OriginPro programs were used for MTF measurement. For MTF measurement using edge method, after achieving ESF by using ImageJ, LSF was calculated by differentiation in OriginPro. Finally, MTF value was obtained through conversion. Image sharpness was defined based on 50% of MTF value.Findings: Results of sharpness measurement by ETL increase revealed that MTF 50% was decreased when ETL was increased. Sharpness comparison between 1ch head coil and 8ch brain coil at 1.5T showed that it was higher for 1ch head coil, although the difference between the two was not statistically significant. However, sharpness of 1ch head coil at 3.0T MRI was found to be significantly higher than that of 8ch brain coil at 3.0T MRI.Improvements/Applications: This study confirms the theoretical concept that MTF measured by ACR standard phantom can be used as a quantitative evaluation method for spatial resolution in the magnetic resonance medical image quality management. It can be considered a meaningful objective evaluation method.

Enhanced brain image retrieval using carrier frequency offset compensated orthogonal frequency division multiplexing for telemedicine applications

AbstractWireless technology plays a vital role in the Electronic health (e‐health) applications such as telemedicine and remote patient monitoring. One of the major challenges of telemedicine applications is ensuring adequate quality of service and realizing precise diagnosis. Telemedicine employs image fusion to enhance the quality of medical image for better diagnosis. However, the transmission of medical image needs high data rate and bandwidth. Due to the robustness to multipath fading, high data rate transmission capability, high flexibility, high bandwidth efficiency, and easy implementation using fast Fourier transform, an orthogonal frequency division multiplexing (OFDM) is the suitable technique for medical image transmission in telemedicine applications This paper is focused on improving the quality of the fused brain image transmission over OFDM based telemedicine service using comb type pilot based carrier frequency offset compensation. The simulation results show that the proposed method offers better performance in terms of mean squared error, Euclidean distance, peak signal to noise ratio, bit error rate, carrier to interference ratio, and computational complexity.

Gold Nanoparticles by Laser Ablation for X-Ray Imaging and Protontherapy Improvements.

Gold nanoparticles, 5-20 nm in diameter, were generated with a pulsed Nd: YAG laser at 1010 W/cm2 at solution concentrations ranging between 1-100 mg/ml. The incremental X-ray contrast imaging using gold nanoparticles was investigated and measured. The study was performed with the aim to enhance the massive absorption coefficient of X-ray radiation in the tumor for medical image quality and to improve traditional X-ray radiotherapy or proton therapy. A simulation of proton therapy improvement was conducted using a human ocular melanoma model, placed 3 cm behind the eye lens, and testing 60 MeV protons. Calculations suggest that the local injection of a solution containing Au-NPs may increase the proton energy released in the tumor above 50%, with the dose in the surrounding tissues leading to an increased probability of tissue healing. A discussion on recent patents in the ambit of the preparation and use of Au nanoparticles in medical imaging and therapy is presented. Au nanoparticles were characterized using optical absorbance, X-ray fluorescence, SEM, and TEM microscopies. Biocompatible nanoparticle solutions were injected intravenously into tail veins of mice followed by X-ray imaging using 20-45 keV photons to evaluate the uptake and the clearance by different organs of the nanoparticles. Diagnostic X-ray images of mice in which the Au-NPs were injected showed high spatial resolution contrast of different organs having high up-take. A calculation of the dose released by X-rays, electrons and protons to the tumor site demonstrates that an increment of the order of 50% can be obtained using adapt solution concentration. The use of Au-NPs in biocompatible solutions injected in living organism permits their blood transport up to different organs. The NPs can be employed as contrast medium to enhance the medical image resolution and to prepare the cancer tissues to be exposed to ionization radiations in order to enhance the dose released to the tumor cells. This effect permits to reduce the total dose given to the patient and to increase the dose released to the tumor cells with respect to healthy ones.

Model observer for assessing digital breast tomosynthesis for multi-lesion detection in the presence of anatomical noise

Model observers are widely used in task-based assessments of medical image quality. The presence of multiple abnormalities in a single set of images, such as in multifocal multicentric breast cancer (MFMC), has an immense clinical impact on treatment planning and survival outcomes. Detecting multiple breast tumors is challenging as MFMC is relatively uncommon, and human observers do not know the number or locations of tumors a priori. Digital breast tomosynthesis (DBT), in which an x-ray beam sweeps over a limited angular range across the breast, has the potential to improve the detection of multiple tumors. However, prior studies of DBT image quality all focus on unifocal breast cancers. In this study, we extended our 2D multi-lesion (ML) channelized Hotelling observer (CHO) into a 3D ML-CHO that detects multiple lesions from volumetric imaging data. Then we employed the 3D ML-CHO to identify optimal DBT acquisition geometries for detection of MFMC. Digital breast phantoms with multiple embedded synthetic lesions were scanned by simulated DBT scanners of different geometries (wide/narrow angular span, different number of projections per scan) to simulate MFMC cases. With new implementations of 3D partial least squares (PLS) and modified Laguerre–Gauss (LG) channels, the 3D ML-CHO made detection decisions based upon the overall information from individual DBT slices and their correlations. Our evaluation results show that: (1) the 3D ML-CHO could achieve good detection performance with a small number of channels, and 3D PLS channels on average outperform the counterpart LG channels; (2) incorporating locally varying anatomical backgrounds and their correlations as in the 3D ML-CHO is essential for multi-lesion detection; (3) the most effective DBT geometry for detection of MFMC may vary when the task of clinical interest changes, and a given DBT geometry may not yield images that are equally informative for detecting MF, MC, and unifocal cancers.

Automated image quality assessment for chest CT scans.

Medical image quality needs to be maintained at standards sufficient for effective clinical reading. Automated computer analytic methods may be applied to medical images for quality assessment. For chest CT scans in a lung cancer screening context, an automated quality assessment method is presented that characterizes image noise and image intensity calibration. This is achieved by image measurements in three automatically segmented homogeneous regions of the scan: external air, trachea lumen air, and descending aorta blood. Profiles of CT scanner behavior are also computed. The method has been evaluated on both phantom and real low-dose chest CT scans and results show that repeatable noise and calibration measures may be realized by automated computer algorithms. Noise and calibration profiles show relevant differences between different scanners and protocols. Automated image quality assessment may be useful for quality control for lung cancer screening and may enable performance improvements to automated computer analysis methods.

Image quality assessment metric for frame accumulated image.

The medical image quality determines the accuracy of diagnosis, and the gray-scale resolution is an important parameter to measure image quality. But current objective metrics are not very suitable for assessing medical images obtained by frame accumulation technology. Little attention was paid to the gray-scale resolution, basically based on spatial resolution and limited to the 256 level gray scale of the existing display device. Thus, this paper proposes a metric, "mean signal-to-noise ratio" (MSNR) based on signal-to-noise in order to be more reasonable to evaluate frame accumulated medical image quality. We demonstrate its potential application through a series of images under a constant illumination signal. Here, the mean image of enough images was regarded as the reference image. Several groups of images by different frame accumulation and their MSNR were calculated. The results of the experiment show that, compared with other quality assessment methods, the metric is simpler, more effective, and more suitable for assessing frame accumulated images that surpass the gray scale and precision of the original image.