Pixels In Region Of Interest Research Articles

A robust watermarking approach against high‐density salt and pepper noise (RWSPN) to enhance medical image security

AbstractThis paper proposes a robust watermarking method against high‐density salt and pepper noise attacks. Automatic region of interest (ROI) detection, embedding encoded data, removing different densities of noise, data extraction by omitting, and labeling the noisy pixels of Region of Non‐Interest (RONI), and decoding the extracted data using ROI pixel information as a key, are various steps of the presented scheme. The automatic ROI detection method separates the RONI from ROI with four vertices of the smallest rectangle, for the embedding process. The encoded watermark data is embedded into the least significant bits of RONI in four neighbour pixels. Adaptive Removal of high‐density Salt and pepper Noise method can enhance image quality and reduce the effect of salt and pepper noise attacks. The embedded information is preserved from destruction if the host image is impaired through the power of robustness. The best results are obtained through the action of extracting the watermark from RONI pixels, utilizing the same ROI detection method. Omitting and labeling the noisy pixels of the RONI will ensure healthy extracted watermark data, leading to decreased Bit Error Rate (BER) values. Finally, these data are interpreted using the key of ROI pixels, and the watermark data is decoded and retrieved. Due to salt and pepper noise obliterating pixel bits and their corresponding transform coefficients in the transform domain, the spatial domain is employed to enhance robustness against such attacks. The results show the high performance of the presented scheme. The average BER value for five MRI databases in a 97% salt and pepper noise attack is 38.6.

A multidimensional chaotic image encryption algorithm based on the region of interest

Most image encryption algorithms encrypt the whole image, but only part of the data is important in the image. In this paper, we propose a multidimensional chaotic image encryption algorithm based on the region of interest (ROI). The histogram of oriented gradients (HOG) feature extraction and support vector machine (SVM) are used to separate the region of interest from the whole image. Then, the region of interest pixels is messed up by using the improved Henon sequence, Joseph sequence and the region of interest pixels are diffused by using the unified chaotic sequence to hide the sensitive information in the image, so as to achieve the purpose of private protection. Furthermore, the improved logistic sequence is used to hide the edge information of the target image to achieve the tradeoff between the secrecy of information and the complexity of encryption. A series of analyses are carried out including key space analysis, key sensitivity analysis, statistical analysis, information entropy analysis, analysis of the fixed-point obscuration analysis, quality analysis and image decryption for our encryption algorithm. Through experiments and comparisons, the proposed algorithm has good performance in encrypting image and coping with various invasions. The image encryption algorithm based on ROI has a good performance of security, moreover through the main encryption of ROI can effectively shorten the encryption time, so as to achieve the compromise of security and computational complexity.

Differentiation of Vascular Characteristics Using Contrast-Enhanced Ultrasound Imaging

Ultrasound contrast imaging has been used to assess tumour growth and regression by assessing the flow through the macro- and micro-vasculature. Our aim was to differentiate the blood kinetics of vessels such as veins, arteries and microvasculature within the limits of the spatial resolution of contrast-enhanced ultrasound imaging. The highly vascularised ovine ovary was used as a biological model. Perfusion of the ovary with SonoVue was recorded with a Philips iU22 scanner in contrast imaging mode. One ewe was treated with prostaglandin to induce vascular regression. Time-intensity curves (TIC) for different regions of interest were obtained, a lognormal model was fitted and flow parameters calculated. Parametric maps of the whole imaging plane were generated for 2 × 2 pixel regions of interest. Further analysis of TICs from selected locations helped specify parameters associated with differentiation into four categories of vessels (arteries, veins, medium-sized vessels and micro-vessels). Time-dependent parameters were associated with large veins, whereas intensity-dependent parameters were associated with large arteries. Further development may enable automation of the technique as an efficient way of monitoring vessel distributions in a clinical setting using currently available scanners.

Segmented cell analyses to measure redox states of autofluorescent NAD(P)H, FAD & Trp in cancer cells by FLIM

Multiphoton FLIM microscopy offers many opportunities to investigate processes in live cells, tissue and animal model systems. For redox measurements, FLIM data is mostly published by cell mean values and intensity-based redox ratios. Our method is based entirely on FLIM parameters generated by 3-detector time domain microscopy capturing autofluorescent signals of NAD(P)H, FAD and novel FLIM-FRET application of Tryptophan and NAD(P)H-a2%/FAD-a1% redox ratio. Furthermore, image data is analyzed in segmented cells thresholded by 2 × 2 pixel Regions of Interest (ROIs) to separate mitochondrial oxidative phosphorylation from cytosolic glycolysis in a prostate cancer cell line. Hundreds of data points allow demonstration of heterogeneity in response to intervention, identity of cell responders to treatment, creating thereby different sub-populations. Histograms and bar charts visualize differences between cells, analyzing whole cell versus mitochondrial morphology data, all based on discrete ROIs. This assay method allows to detect subtle differences in cellular and tissue responses, suggesting an advancement over means-based analyses.

Providing uncertainty estimates of the Sentinel-2 top-of-atmosphere measurements for radiometric validation activities

ABSTRACTAs part of the Sentinel-2 mission, a Radiometric Uncertainty Tool (RUT) has been recently released to the community. This tool estimates the Sentinel-2 radiometric uncertainty associated with each pixel in the top-of-atmosphere (TOA) reflectance factor images provided by the European Space Agency (ESA). The use of such information enables users to assess the “fitness for purpose” of the data to their specific application. The work described here summarises the efforts and results of integrating the RUT for radiometric validation activities for the Sentinel-2 mission. Starting from the results provided by the RUT, the focus will be on providing a methodology to calculate the uncertainty associated with the mean TOA reflectance factor in a Region of Interest (ROI). Two different methods – one simple method directly using the RUT and a more rigorous one based on Monte Carlo method (MCM) propagation – are proposed and compared. These two methods focus on the effect of the spectral, spatial and temporal correlation of the errors in different ROI pixels and the impact of correlation on the uncertainty associated with the mean TOA reflectance factor. The study has also considered the impact of uncertainty contributions not included in the first version of the RUT.

Development of a Computer-aided Diagnosis System to Distinguish between Benign and Malignant Mammary Tumors in Dynamic Magnetic Resonance Images: Automatic Detection of the Position with the Strongest Washout Effect in the Tumor

We propose a computer-aided diagnostic (CAD) system that uses time-intensity curves to distinguish between benign and malignant mammary tumors. Many malignant tumors show a washout pattern in time-intensity curves. Therefore, we designed a program that automatically detects the position with the strongest washout effect using the technique, such as the subtraction technique, which extracts only the washout area in the tumor, and by scanning data in 2×2 pixel region of interest (ROI). Operation of this independently developed program was verified using a phantom system that simulated tumors. In three cases of malignant tumors, the washout pattern detection rate in images with manually set ROI was ≤6%, whereas the detection rate with our novel method was 100%. In one case of a benign tumor, when the same method was used, we checked that there was no washout effect and detected the persistent pattern. Thus, the distinction between benign and malignant tumors using our method was completely consistent with the pathological diagnoses made. Our novel method is therefore effective for differentiating between benign and malignant mammary tumors in dynamic magnetic resonance images.

Supervised Vessels Classification Based on Feature Selection

Arterial-venous classification of retinal blood vessels is important for the automatic detection of cardiovascular diseases such as hypertensive retinopathy and stroke. In this paper, we propose an arterial-venous classification (AVC) method, which focuses on feature extraction and selection from vessel centerline pixels. The vessel centerline is extracted after the preprocessing of vessel segmentation and optic disc (OD) localization. Then, a region of interest (ROI) is extracted around OD, and the most efficient features of each centerline pixel in ROI are selected from the local features, grey-level co-occurrence matrix (GLCM) features, and an adaptive local binary patten (A-LBP) feature by using a max-relevance and min-redundancy (mRMR) scheme. Finally, a feature-weighted K-nearest neighbor (FW-KNN) algorithm is used to classify the arterial-venous vessels. The experimental results on the DRIVE database and INSPIRE-AVR database achieve the high accuracy of 88.65% and 88.51% in ROI, respectively.

Human Visual System-based Unequal Error Protection for Robust Video Coding

To increase the overall visual quality of the video services without increasing data rate, a human visual system-based video coding, founded on a hierarchy of the video stream in different levels of importance, is developed. Determining these importance levels takes in count three classification criteria: the position of current image in the group of images (image level), the importance of the motion vectors of macroblocks in the current image (macroblock level) and belonging or not of a pixel in a spatial region of interest (pixel level). At the end of this classification process, an interpolation between the results of the three-level selection allows to establish an index of importance for each macroblock of the image to be encoded. This index determines the type of channel coding to be applied to the corresponding macroblock. Tests have shown that the technique presented in this paper achieves better results in PSNR and SSIM (structural similarity) than an equal error protection technique.

Region-based hybrid medical image watermarking for secure telemedicine applications

In this paper we propose a novel region based hybrid medical image watermarking (MIW) scheme to ensure authenticity, integrity and confidentiality of medical images. In this scheme a digital medical image is partitioned into region of interest (ROI) and the region of non interest (RONI). To detect and localize ROI tampering with high accuracy pixel wise positional and relational bits are calculated. Positional bit is calculated with respect to MSBs, row and column of the pixel. Relational bit shows the relation between MSBs. Two original LSBs of each ROI pixel are replace by their corresponding positional and relational bits. Original LSBs of ROI pixels are concatenated and embedded in RONI for ROI recovery in the case of tampering. Multiple watermarks i.e. electronic patient record (EPR), hospitals logo and LSBs of ROI are embedded simultaneously as a robust watermark in RONI using IWT-SVD hybrid transform. The proposed scheme is blind and free from false positive detection. Various experiments have been carried out on different medical imaging modalities to evaluate the performance of the proposed scheme in terms of imperceptibility, robustness, tamper detection, localization, recovery and computation time. ROI tampering is detected and recovered with high accuracy. Thus, the proposed scheme is effective in telemedicine applications.

Elastography for Predicting and Localizing Nodal Metastases during Endobronchial Ultrasound

Background: Elastography is a relatively new technology that can generate images reflective of tissue stiffness (elasticity). Neoplastic tissue is usually stiffer than normal structures. Objectives: The aim of this study was to evaluate the feasibility and utility of elastography when combined with convex-probe endobronchial ultrasound (CP-EBUS) for predicting and localizing metastatic lymph nodes during endobronchial ultrasound with transbronchial needle aspiration (EBUS-TBNA). Methods: Consecutive results of endobronchial elastography of lymph nodes performed using EBUS- TBNA were prospectively collected and retrospectively analyzed. Elastography images were acquired as JPEG images and also recorded as video clips. Stiff area ratios [(stiff areas as blue pixels) / (lymph node areas as region of interest pixels)] for each lymph node determined by elastography were collated with the results of pathological diagnosis. We also performed elastography of surgically resected lymph nodes and compared image findings with pathological sections. Results: We evaluated 49 lymph nodes in 21 patients by CP-EBUS. There were 16 metastatic nodes (10 lung cancer metastases and 6 metastases from extrathoracic malignancies). Mean stiff area ratios were significantly greater for metastatic lymph nodes (0.478) than for benign nodes (0.216; p = 0.0002). Using a cutoff value of 0.311 for stiff area ratios, the sensitivity and specificity for predicting metastatic disease were 0.81 and 0.85, respectively. The stiff area was histologically compatible with metastatic distribution in surgically resected lymph nodes. Conclusions: Endobronchial elastography is feasible for lymph nodes when combined with CP-EBUS. Stiff area ratios are useful for predicting metastatic lymph nodes, which may be an efficient guide for TBNA.

Prognostic value of combined visualization of MR diffusion and perfusion maps in glioblastoma.

We analyzed whether the combined visualization of decreased apparent diffusion coefficient (ADC) values and increased cerebral blood volume (CBV) in perfusion imaging can identify prognosis-related growth patterns in patients with newly diagnosed glioblastoma. Sixty-five consecutive patients were examined with diffusion and dynamic susceptibility-weighted contrast-enhanced perfusion weighted MRI. ADC and CBV maps were co-registered on the T1-w image and a region of interest (ROI) was manually delineated encompassing the enhancing lesion. Within this ROI pixels with ADC values <the 30th percentile (ADCmin), pixels with CBV values >the 70th percentile (CBVmax) and the intersection of pixels with ADCmin and CBVmax were automatically calculated and visualized. Initially, all tumors with a mean intersection greater than the upper quartile of the normally distributed mean intersection of all patients were subsumed to the first growth pattern termed big intersection (BI). Subsequently, the remaining tumors' growth patterns were categorized depending on the qualitative representation of ADCmin, CBVmax and their intersection. Log-rank test exposed a significantly longer overall survival of BI (n = 16) compared to non-BI group (n = 49) (p = 0.0057). Thirty-one, four and 14 patients of the non-BI group were classified as predominant ADC-, CBV- and mixed growth group, respectively. In a multivariate Cox regression model, the BI-, CBV- and mixed groups had significantly lower adjusted hazard ratios (p-value, α(Bonferroni) < 0.006) when compared to the reference group ADC: 0.29 (0.0027), 0.11 (0.038) and 0.33 (0.0059). Our study provides evidence that the combination of diffusion and perfusion imaging allows visualization of different glioblastoma growth patterns that are associated with prognosis. A possible biological hypothesis for this finding could be the interpretation of the ADCmin fraction as the invasion-front of tumor cells while the CBVmax fraction might represent the vascular rich tumor border that is "trailing behind" the invasion-front in the ADC group.

Automatic ROI construction for analyzing time-signal intensity curve in dynamic contrast-enhanced MR imaging of the breast.

Our purpose in this study was to construct a 3-dimensional (3D) region of interest (ROI) for analyzing the time-signal intensity curve (TIC) semi-automatically in dynamic contrast-enhanced magnetic resonance (DCE-MR) imaging of the breast. DCE-MR breast imaging datasets were acquired by a 3.0-Tesla MR system with the use of a 3D fast gradient echo sequence. The essential idea in the new method was to analyze each pixel and to construct an ROI made up of pixels with similar TICs. First, an analyst selected a starting point in the contrast media-enhanced tumor. Second, we calculated Pearson's correlation coefficients (CCs) between the TIC in the starting coordinate selected by the analyst and the TIC in the other coordinates. Third, ROI pixels were selected if their CC threshold satisfied a level of coefficient variation of the ROI determined by prior research performed in our institution. We made a retrospective review of patients who underwent breast DCE-MR examination for pre-operative diagnosis. To confirm the feasibility of the resulting 3D-ROI from TIC analysis, we compared Fischer's score obtained from 3D-ROI by applying a new method to a score obtained from a manually selected 2-dimensional (2D) ROI which was used during routine clinical examination. The Fischer's scores obtained from both the automatically selected 3D-ROI and the manually selected 2D-ROI showed almost equivalent results. Thus, we considered that the new method was comparable to the conventional method. Furthermore, the new method has the potential to be used for evaluation of the extent of tumors.

Different wavelengths selection methods for identification of early blight on tomato leaves by using hyperspectral imaging technique

Identification of early blight on tomato leaves by using hyperspectral imaging technique based on different effective wavelengths selection methods (successive projections algorithm, SPA; x-loading weights, x-LW; gram-schmidt orthogonaliza-tion, GSO) was studied in the present paper. Hyperspectral images of seventy healthy and seventy infected tomato leaves were obtained by hyperspectral imaging system across the wavelength range of 380-1023 nm. Reflectance of all pixels in region of interest (ROI) was extracted by ENVI 4. 7 software. Least squares-support vector machine (LS-SVM) model was established based on the full spectral wavelengths. It obtained an excellent result with the highest identification accuracy (100%) in both calibration and prediction sets. Then, EW-LS-SVM and EW-LDA models were established based on the selected wavelengths suggested by SPA, x-LW and GSO, respectively. The results showed that all of the EW-LS-SVM and EW-LDA models performed well with the identification accuracy of 100% in EW-LS-SVM model and 100%, 100% and 97. 83% in EW-LDA model, respectively. Moreover, the number of input wavelengths of SPA-LS-SVM, x-LW-LS-SVM and GSO-LS-SVM models were four (492, 550, 633 and 680 nm), three (631, 719 and 747 nm) and two (533 and 657 nm), respectively. Fewer input variables were beneficial for the development of identification instrument. It demonstrated that it is feasible to identify early blight on tomato leaves by using hyperspectral imaging, and SPA, x-LW and GSO were effective wavelengths selection methods.

Medical Image Watermarking Technique for Accurate Tamper Detection in ROI and Exact Recovery of ROI.

In telemedicine while transferring medical images tampers may be introduced. Before making any diagnostic decisions, the integrity of region of interest (ROI) of the received medical image must be verified to avoid misdiagnosis. In this paper, we propose a novel fragile block based medical image watermarking technique to avoid embedding distortion inside ROI, verify integrity of ROI, detect accurately the tampered blocks inside ROI, and recover the original ROI with zero loss. In this proposed method, the medical image is segmented into three sets of pixels: ROI pixels, region of noninterest (RONI) pixels, and border pixels. Then, authentication data and information of ROI are embedded in border pixels. Recovery data of ROI is embedded into RONI. Results of experiments conducted on a number of medical images reveal that the proposed method produces high quality watermarked medical images, identifies the presence of tampers inside ROI with 100% accuracy, and recovers the original ROI without any loss.

An Improved Iris Localization Algorithm Based on Curve Fitting

Iris localization is to detect outer-and-inner boundaries of iris in an iris image. In the paper, an improved algorithm was proposed to quickly and effectively locate outer-and-inner boundaries. As for this algorithm, the first is to block an iris image and extract its sub-image blocks which cover pupil; the second is to set a binary threshold of pupil by adopting the method of Maximum Variance between Clusters; the third is to get the value outer-boundary-points of iris, on the basis of gray gradient of key Regions-of-interest; the last is to select some characteristic pixels in regions of interest respectively and fit outer-and-inner boundaries of iris according to curve fitting.

Apartitioned Approach to Real Time Lane Detection at Different Weather Conditions

A partitioned approach to real time lane detection is proposed based on the ARM core microprocessor S3C6410. With the help of the dedicated camera interface in S3C6410, the original image can be converted to RGB format and got window-cut in hardware, leaving the target region of interest (ROI). The pixels in ROI are partitioned into two parts to deal with some hostile weather conditions when lane markings in far field are hard to be distinguished from the homogenous road surface. Hough transform is applied into the top part to utilize lane continuum, and the pixel in bottom part is detected in some fixed search bars to reduce computation complexity. Experiments show that the detection algorithm possesses real time performanceand good robustness at different weather conditions.

Influence of myocardial region of interest definition on quantitative analysis of planar 123I-mIBG images

In planar (123)I-mIBG myocardial imaging, definition of the heart region of interest (ROI) is a critical step in quantifying uptake. The present study evaluated the impact of changes in heart ROI size on quantitative results in subjects with good and poor uptake. Reference irregular whole-heart and square upper mediastinum ROIs were defined visually on 531 planar (123)I-mIBG images. Based on the reference heart ROI, an automated program created two other ROIs: one larger (+1 pixel) and one smaller (-1 pixel), the stated numbers representing the spacing intervals between each epicardial boundary pixel. Two additional smaller ROIs (-2 and -3 pixels) were drawn for the 100 images (19%) with a heart/mediastinum (H/M) ratio ≤1.30. The number of pixels, the counts per pixel, and the H/M ratio for each heart ROI were calculated and compared to that in the reference ROI. Washout rate and changes as a function of ROI size were also calculated for 110 subjects who had both early and late images. The mean changes in heart ROI size ranged from +19.0% for the +1 pixel to -44.4% for the -3 pixels ROI. For the +1 and -1 pixel ROIs, mean relative counts per pixel changes were -1.2% and +0.7%, respectively, with corresponding ranges of change in the H/M ratio of -0.12 to +0.05 and -0.05 to +0.11. For scans with H/M ratio ≤1.30, mean relative counts per pixel changes were 0, 0, -0.7%, and -1.4% for the four ROIs, with range of change in the H/M ratio from -0.13 to +0.05. Mean washout rates were almost identical for the reference ROIs (45.0%) and the +1 pixel and -1 pixel heart ROIs (44.9% and 45.1%, respectively). Significant changes in the size of the best visually defined heart ROI produce minimal, clinically inconsequential changes in the H/M ratio and washout rate, even in subjects with significantly reduced myocardial uptake of (123)I-mIBG.

Injecting Mold Protection Method Based on Machine Vision

Aimed at solving the problem of mold damage caused by a foreign body in the mold before mold clamping, this paper proposes a solution, which applies image processing technology such as background updating and the difference image algorithm to solve it. Not only can it judge whether there is a foreign body in the mold but it can also detect whether the product is perfect by comparing the foreground image with the background image at the appropriate time (before mold clamping or after mold opening) and by calculating the qualified rate of pixel in all ROIs (Region of Interest). To eliminate the influence of vibration and of changes in brightness in the surrounding environment on the detecting results, this paper utilizes near infrared illumination technology and the background updating algorithm. In addition, the ROI is set to improve the detecting speed and accuracy.

Evaluation of Regional Brain Perfusion in Patients with Pseudoexfoliation Syndrome

Pseudoexfoliation syndrome (PES) is characterised by the accumulation of abnormal extracellular fibrillar material in ocular tissue, which develops with ageing, and can lead to ocular complications, including glaucoma, cataract, zonular damage, and keratopathy. It has been shown that extraocular accumulation of pseudoexfoliation matter can occur in tissues such as the heart, lung, liver, kidney, cerebral meninges, aorta, and adventitia layer of cerebral arteries. Acute vascular complications are seen commonly with the involvement of the systemic vasculature. Pseudoexfoliation syndrome has also been reported in dementia and stroke patients. In this study, the authors aimed to evaluate regional brain perfusion in pseudoexfoliation syndrome patients. Twenty-four patients with PES with glaucoma (14 female, 10 male) and a control group of 19 healthy volunteers (10 female, 9 male) were included in the study. There was no cerebrovascular or psychiatric complaint in the patient or control group, and all subjects were right-handed. Brain perfusion single-photon emission computed tomography (SPECT) images were taken by low-energy general purpose collimator of a single-headed gamma camera after intravenous injection of 25 mCi technetium-99m-hexamethylpropylene amine oxime (Tc99m-HMPAO). To obtain the corticocerebellar ratio by semiquantitative analysis, 3 × 3 pixel regions of interest were drawn on transaxial sections, and the counts of frontal, temporal, parietal, and occipital regions were divided by the counts of the cerebellar region bilaterally. The mean ages of the PES and control groups were 67.83 ± 9.03 and 64.52 ± 6.88 years, respectively. The mean corticocerebellar ratio in PES was decreased in frontal, temporal, parietal, and occipital lobes bilaterally (p value interval min: 0.001, max: 0.049). The results showed that there was a decrease in frontal, temporal, parietal, and occipital cortices perfusion in PES patients bilaterally.

Myocardial SPECT Images for Diagnosis of Pulmonary Hypertension and Right Ventricular Hypertrophy

The utility of (99m)Tc-tetrofosmin myocardial SPECT for assessment of pulmonary hypertension and right ventricular thickness has not been well studied. We hypothesized that a ratio of right ventricular activity to left ventricular activity (RV/LV uptake ratio) from SPECT myocardial perfusion images could identify the presence of increased right ventricular wall thickness and elevated systolic pulmonary artery pressure with or without the use of attenuation correction. We identified 33 patients with normal findings on stress (99m)Tc-tetrofosmin left ventricular myocardial perfusion imaging who had a complete 2-dimensional echocardiographic study within 3 wk of the SPECT study. Two 6 × 6 pixel regions of interest were placed in the right and left ventricular free walls of both non-attenuation-corrected and attenuation-corrected SPECT images. We examined the correlation of RV/LV uptake ratio with echocardiographic right ventricular free-wall thickness and with pulmonary artery systolic pressure. RV/LV uptake ratio, measured on non-attenuation-corrected images, correlated significantly with both pulmonary artery systolic pressure (r = 0.63 and P < 0.001) and right ventricular wall thickness (r = 0.6 and P < 0.001). Receiver-operating-characteristic analysis of the use of RV/LV uptake ratio to detect significant pulmonary hypertension showed that the area under the curve was 0.78 (95% confidence interval, 0.62-0.95). However, no significant correlation of RV/LV uptake ratio with pulmonary artery systolic pressure or right ventricular wall thickness was found on attenuation-corrected images. RV/LV uptake ratio measured on SPECT images can be used to identify patients with high pulmonary artery pressure or right ventricular hypertrophy.