Region Of Interest Size Research Articles

Inter-software reliability and agreement for follicular and luteal morphometric and echotextural ultrasonographic parameters in beef cattle

This study aimed to compare the inter-software and inter-observer reliability and agreement for the assessment of follicular and luteal morphometry and echotexture parameters in beef crossbreed females (3/8 Bos taurus indicus and 5/8 Bos taurus taurus). B-mode and color Doppler ultrasonographic ovarian images were obtained at specific time points of estradiol-progesterone-based protocols for timed artificial insemination (TAI). Sonograms were analyzed by two observers using a licensed (IASP1) and an open access (IASP2) software package. A total of 292 snap-shot sonograms were analyzed for morphometric parameters and 504 for echotexture parameters. inter-software reliability was judged moderate to excellent (ICC or CCC=0.73–0.98), whereas inter-observer reliability for morphometric parameters was deemed good to excellent (ICC or CCC=0.82–0.98). A small percentage (up to 10.95 %) of measured parameters fell outside the limits of inter-software and inter-observer agreement. For echotexture parameters, inter-software reliability varied widely (ICC or CCC=0.16–0.95) based on the size of regions of interest (ROI), while inter-observer reliability ranged from moderate to excellent (ICC or CCC= 0.71–0.97). The highest inter-software reliability for pixel value and heterogeneity value was observed for the corpus luteum (ICCs=0.81–0.95; P>0.05), followed by the peripheral follicular antrum (ICCs=0.75–0.78; P<0.05). However, lower reliability was determined for the follicular wall (ICCs=0.08–0.33; P<0.0001) and perifollicular stroma (ICCs=0.16–0.46; P<0.05). In conclusion, both software packages showed high reproducibility for morphometric measurements, while echotexture measurements were more challenging to replicate based on ROI sizes. Caution is advised when selecting ROI sizes for echotexture measurements in bovine ovaries.

A novel method for developing contrast-detail curves from clinical patient images based on statistical low-contrast detectability

Purpose. To develop a method to extract statistical low-contrast detectability (LCD) and contrast-detail (C-D) curves from clinical patient images. Method. We used the region of air surrounding the patient as an alternative for a homogeneous region within a patient. A simple graphical user interface (GUI) was created to set the initial configuration for region of interest (ROI), ROI size, and minimum detectable contrast (MDC). The process was started by segmenting the air surrounding the patient with a threshold between −980 HU (Hounsfield units) and −1024 HU to get an air mask. The mask was trimmed using the patient center coordinates to avoid distortion from the patient table. It was used to automatically place square ROIs of a predetermined size. The mean pixel values in HU within each ROI were calculated, and the standard deviation (SD) from all the means was obtained. The MDC for a particular target size was generated by multiplying the SD by 3.29. A C-D curve was obtained by iterating this process for the other ROI sizes. This method was applied to the homogeneous area from the uniformity module of an ACR CT phantom to find the correlation between the parameters inside and outside the phantom, for 30 thoracic, 26 abdominal, and 23 head images. Results. The phantom images showed a significant linear correlation between the LCDs obtained from outside and inside the phantom, with R2 values of 0.67 and 0.99 for variations in tube currents and tube voltages. This indicated that the air region outside the phantom can act as a surrogate for the homogenous region inside the phantom to obtain the LCD and C-D curves. Conclusion. The C-D curves obtained from outside the ACR CT phantom show a strong linear correlation with those from inside the phantom. The proposed method can also be used to extract the LCD from patient images by using the region of air outside as a surrogate for a region inside the patient.

Quantitative image analysis pipeline for detecting circulating hybrid cells in immunofluorescence images with human-level accuracy.

Circulating hybrid cells (CHCs) are a newly discovered, tumor-derived cell population found in the peripheral blood of cancer patients and are thought to contribute to tumor metastasis. However, identifying CHCs by immunofluorescence (IF) imaging of patient peripheral blood mononuclear cells (PBMCs) is a time-consuming and subjective process that currently relies on manual annotation by laboratory technicians. Additionally, while IF is relatively easy to apply to tissue sections, its application to PBMC smears presents challenges due to the presence of biological and technical artifacts. To address these challenges, we present a robust image analysis pipeline to automate the detection and analysis of CHCs in IF images. The pipeline incorporates quality control to optimize specimen preparation protocols and remove unwanted artifacts, leverages a β-variational autoencoder (VAE) to learn meaningful latent representations of single-cell images, and employs a support vector machine (SVM) classifier to achieve human-level CHC detection. We created a rigorously labeled IF CHC data set including nine patients and two disease sites with the assistance of 10 annotators to evaluate the pipeline. We examined annotator variation and bias in CHC detection and provided guidelines to optimize the accuracy of CHC annotation. We found that all annotators agreed on CHC identification for only 65% of the cells in the data set and had a tendency to underestimate CHC counts for regions of interest (ROIs) containing relatively large amounts of cells (>50,000) when using the conventional enumeration method. On the other hand, our proposed approach is unbiased to ROI size. The SVM classifier trained on the β-VAE embeddings achieved an F1 score of 0.80, matching the average performance of human annotators. Our pipeline enables researchers to explore the role of CHCs in cancer progression and assess their potential as a clinical biomarker for metastasis. Further, we demonstrate that the pipeline can identify discrete cellular phenotypes among PBMCs, highlighting its utility beyond CHCs.

An integrated machine learning-based model for joint diagnosis of ovarian cancer with multiple test indicators

ObjectiveTo construct a machine learning diagnostic model integrating feature dimensionality reduction techniques and artificial neural network classifiers to develop the value of clinical routine blood indexes for the auxiliary diagnosis of ovarian cancer.MethodsPatients with ovarian cancer clearly diagnosed in our hospital were collected as a case group (n = 185), and three groups of patients with other malignant otolaryngology tumors (n = 138), patients with benign otolaryngology diseases (n = 339) and those with normal physical examination (n = 92) were used as an overall control group. In this paper, a fully automated segmentation network for magnetic resonance images of ovarian cancer is proposed to improve the reproducibility of tumor segmentation results while effectively reducing the burden on radiologists. A pre-trained Res Net50 is used to the three edge output modules are fused to obtain the final segmentation results. The segmentation results of the proposed network architecture are compared with the segmentation results of the U-net based network architecture and the effect of different loss functions and region of interest sizes on the segmentation performance of the proposed network is analyzed.ResultsThe average Dice similarity coefficient, average sensitivity, average specificity (specificity) and average hausdorff distance of the proposed network segmentation results reached 83.62%, 89.11%, 96.37% and 8.50, respectively, which were better than the U-net based segmentation method. For ROIs containing tumor tissue, the smaller the size, the better the segmentation effect. Several loss functions do not differ much. The area under the ROC curve of the machine learning diagnostic model reached 0.948, with a sensitivity of 91.9% and a specificity of 86.9%, and its diagnostic efficacy was significantly better than that of the traditional way of detecting CA125 alone. The model was able to accurately diagnose ovarian cancer of different disease stages and showed certain discriminative ability for ovarian cancer in all three control subgroups.ConclusionUsing machine learning to integrate multiple conventional test indicators can effectively improve the diagnostic efficacy of ovarian cancer, which provides a new idea for the intelligent auxiliary diagnosis of ovarian cancer.

Unbiased centroiding of point targets close to the Cramer Rao limit.

Systematic errors affecting center-of-gravity (CoG) measurements may occur from coarse sampling of the point-spread-function (PSF) or from signal truncation at the boundaries of the region-of-interest (ROI). For small ROI and PSF widths, these effects are shown to become dominant, but this can be mitigated by introducing novel unbiased estimators that are largely free of systematic error and perform particularly well for low photon numbers. Analytical expressions for the estimator variances, comprising contributions from photon shot noise, random pixel noise, and residual systematic error, are derived and verified by Monte Carlo simulations. The accuracy and computational speed of the unbiased estimators are compared to those of other common estimators, including iteratively weighted CoG, thresholded CoG, iterative least squares fitting, and two-dimensional Gaussian regression. Each estimator is optimized with respect to ROI size and PSF radius and its error compared to the theoretical limit defined by the Cramer Rao lower bound (CRLB). The unbiased estimator with full systematic error correction operating on a small ROI [3×3] emerges as one of the most accurate estimators while requiring significantly less computing effort than alternative algorithms.

Deep-Interior: A new pathway to interior tomographic image reconstruction via a weighted backprojection and deep learning.

In recent years, deep learning strategies have been combined with either the filtered backprojection or iterative methods or the direct projection-to-image by deep learning only to reconstruct images. Some of these methods can be applied to address the interior reconstruction problems for centered regions of interest (ROIs) with fixed sizes. Developing a method to enable interior tomography with arbitrarily located ROIs with nearly arbitrary ROI sizes inside a scanning field of view (FOV) remains an openquestion. To develop a new pathway to enable interior tomographic reconstruction for arbitrarily located ROIs with arbitrary sizes using a single trained deep neural network model. The method consists of two steps. First, an analytical weighted backprojection reconstruction algorithm was developed to perform domain transform from divergent fan-beam projection data to an intermediate image feature space, , for an arbitrary size ROI at an arbitrary location inside the FOV. Second, a supervised learning technique was developed to train a deep neural network architecture to perform deconvolution to obtain the true image from the new feature space . This two-step method is referred to as Deep-Interior for convenience. Both numerical simulations and experimental studies were performed to validate the proposed Deep-Interior method. The results showed that ROIs as small as a diameter of 5 cm could be accurately reconstructed (similarity index 0.985 ± 0.018 on internal testing data and 0.940 ± 0.025 on external testing data) at arbitrary locations within an imaging object covering a wide variety of anatomical structures of different body parts. Besides, ROIs of arbitrary size can be reconstructed by stitching small ROIs without additionaltraining. The developed Deep-Interior framework can enable interior tomographic reconstruction from divergent fan-beam projections for short-scan and super-short-scan acquisitions for small ROIs (with a diameter larger than 5 cm) at an arbitrary location inside the scanning FOV with high quantitative reconstruction accuracy.

Plasma IL‐6 levels are Negatively Associated with MRI‐Based Water Exchange Rate Across the Blood‐Brain Barrier

AbstractBackgroundChanges in blood‐brain barrier (BBB) integrity are associated with cognitive decline in aging. Inflammatory processes are thought to contribute to BBB dysfunction but this link has been difficult to measure in humans using non‐invasive, in‐vivo methods. Recently, a novel magnetic resonance imaging (MRI) technique, diffusion‐prepared arterial spin labeling (DP‐ASL), has been shown to provide a means to test the water exchange rate across the BBB (kw) as well as arterial transit time (ATT) and cerebral blood flow (CBF). Here, we tested the association between these DP‐ASL measures and a specific plasma marker of inflammation, IL‐6, in older adults without dementia.MethodWe tested the association between kw, ATT, and CBF and a marker of peripheral inflammation in a sample of 58 older adults without dementia. Regions of interest (ROIs) included the putamen, caudate, and pallidum. Plasma IL‐6 levels were assessed using Simoa assays. IL‐6 was natural log transformed and data collected ≤ 1 year from the MRI visit were used for subsequent analyses. Multiple linear regression models, controlling for age, sex, and ROI size, tested the impact of plasma IL‐6 on kw, ATT, and CBF metrics. Both linear and quadratic effects of IL‐6 on the outcomes of interest were examined.ResultThere was a negative linear association between IL‐6 and kw in the putamen. As plasma IL‐6 levels increased, kw values decreased. In contrast, no relationships were observed between plasma IL‐6 and ATT or CBF in any of the ROIs tested.ConclusionOur results suggest that the water exchange rate across the BBB in the putamen was most uniquely associated with the peripheral proinflammatory cytokine, IL‐6. These findings could be indicative of the involvement of peripheral inflammation in reducing glymphatic clearance which would manifest as reduced water exchange rate across the BBB. The lack of association with ATT and CBF suggest that peripheral cytokines could act on a mechanism more closely related to water channel function rather than overall brain perfusion.

Recommendations on Use of DIC in Experimental Geotechnics

Abstract Image-based deformation measurement techniques, like digital image correlation (DIC), are used to measure the displacement of soil particles and thereby to investigate the behavior of geo-structure in experimental geotechnics. The precision of the measurement from the DIC technique depends on the position of the camera and light panels, the color temperature of light, size of the region of interest (RoI), type and relative density of sand, seeding ratio, type of camera, and strain rate used in experiments. In the paper, the effect of above-mentioned parameters is investigated experimentally in 1-g model tests. The studies show that the precision in DIC measurements is dependent on the position of camera and light panels, size of RoI, texture of the sand, and seeding ratio. However, the relative density of sand and strain rate do not significantly affect the results from the DIC analysis. The efficacy in the measurement of the large deformations and suitability of smartphone cameras using DIC are also addressed. Last, we developed recommendations to obtain precise results from DIC in experimental geotechnics.

Impact of Region-of-Interest Size on the Diagnostic Performance of Shear Wave Elastography in Differentiating Thyroid Nodules.

This study investigated the impact of different region-of-interest (ROI) sizes (Max, 1 mm, and 2 mm) on shear wave elastography (SWE) in differentiating between malignant and benign thyroid nodules. The study cohort comprised 129 thyroid nodules (50 malignant, 79 benign) and 78 normal subjects. Diagnostic efficacy was assessed through pairwise comparisons of area under the curve (AUC) values in receiver operating characteristic analysis by using DeLong's test. Our results indicated significant differences in all SWE elasticity metrics between the groups, with malignant nodules exhibiting higher values than benign nodules (p < 0.05). Smaller ROIs (1 and 2 mm) were found to outperform the max ROI in terms of diagnostic accuracy, particularly for the Emax and Emin elasticity metrics. Emax(1mm) had the highest diagnostic accuracy, with an AUC of 0.883, sensitivity of 74.0%, and specificity of 86.1%. This study underscores the significant influence of ROI size selection on the diagnostic performance of SWE, offering valuable insights for future research and clinical applications in thyroid nodule assessment.

Myofascial Trigger Point Identification in B-Mode Ultrasound: Texture Analysis Versus a Convolutional Neural Network Approach.

Myofascial pain syndrome (MPS) is one of the most common causes of chronic pain and affects a large portion of patients seen in specialty pain centers as well as primary care clinics. Diagnosis of MPS relies heavily on a clinician's ability to identify the presence of a myofascial trigger point (MTrP). Ultrasound can help, but requires the user to be experienced in ultrasound. Thus, this study investigates the use of texture features and deep learning strategies for the automatic identification of muscle with MTrPs (i.e., active and latent MTrPs) from normal (i.e., no MTrP) muscle. Participants (n=201) were recruited from Toronto Rehabilitation Institute, and ultrasound videos of their trapezius muscles were acquired. This new data set consists of 1344 images (248 active, 120 latent, 976 normal) collected from these videos. For texture analysis, several features were investigated with varying parameters (i.e., region of interest size, feature type and pixel pair relationships). Convolutional neural networks (CNN) were also applied to observe the performance of deep learning approaches. Performance was evaluated based on the classification accuracy, micro F1-score, sensitivity, specificity, positive predictive value and negative predictive value. The best CNN approach was able to differentiate between muscles with and without MTrPs better than the best texture feature approach, with F1-scores of 0.7299 and 0.7135, respectively. The results of this study reveal the challenges associated with MTrP identification and the potential and shortcomings of CNN and radiomics approaches in detail.

T1rho MR properties of human patellar cartilage: correlation with indentation stiffness and biochemical contents.

Cartilage degeneration involves structural, compositional, and biomechanical alterations that may be detected non-invasively using quantitative MRI. The goal of this study was to determine if topographical variation in T1rho values correlates with indentation stiffness and biochemical contents of human patellar cartilage. Cadaveric patellae from unilateral knees of 5 donors with moderate degeneration were imaged at 3-Telsa with spiral chopped magnetization preparation T1rho sequence. Indentation testing was performed, followed by biochemical analyses to determine water and sulfated glycosaminoglycan contents. T1rho values were compared to indentation stiffness, using semi-circular regions of interest (ROIs) of varying sizes at each indentation site. ROIs matching the resected tissues were analyzed, and univariate and multivariate regression analyses were performed to compare T1rho values to biochemical contents. Grossly, superficial degenerative change of the cartilage (i.e., roughened texture and erosion) corresponded with regions of high T1rho values. High T1rho values correlated with low indentation stiffness, and the strength of correlation varied slightly with the ROI size. Spatial variations in T1rho values correlated positively with that of the water content (R2 = 0.10, p < 0.05) and negatively with the variations in the GAG content (R2 = 0.13, p < 0.01). Multivariate correlation (R2 = 0.23, p < 0.01) was stronger than either of the univariate correlations. These results demonstrate the sensitivity of T1rho values to spatially varying function and composition of cartilage and that the strength of correlation depends on the method of data analysis and consideration of multiple variables.

2D shear wave elastography for the assessment of quadriceps entheses-a methodological study.

To establish the scanning protocol for 2-dimensional shear wave elastography (SWE) on normal entheses by investigating the possible confounding factors that may increase the variability of measured elasticity. 30 normal quadriceps entheses were scanned using SWE to compare the stiffness and coefficient variation by changing the ultrasonic coupling gel thickness, knee position, region of interest size, and scanning plane. No significant difference in median shear wave velocity (SWV) was observed in different coupling gel thicknesses. The median SWV was higher in the knee flexion position than in the extended position (p < 0.001). Increased knee flexion led to stiffer quadriceps enthesis and higher SWV (ρ = 0.8, p < 0.001). The median SWV was higher when the diameter region of interest was 4.0mm than 2.0mm (p = 0.001). The median SWV was higher in the transverse plane than in the longitudinal plane (p < 0.001). Strong correlation was found between SWV and the degree of the shear wave to muscle fiber direction (ρ = 0.8, p < 0.001). The coefficient variation was lower in a gel thickness of 2.5cm, with an extended knee, a region of interest of 2.0mm, and a longitudinal plane (p > 0.05). For interobserver reliability for the proposed protocol, the intraclass correlation coefficients was 0.763. In this study, we determined supine position with the knee extended; using 2.0mm diameter region of interest and image acquisition at the longitudinal plane with thicker layer coupling gel seems most appropriate to reliably image healthy quadriceps entheses with SWE.

Influence of Measurement Depth and Acquisition Parameters on Shear Wave Speed and Shear Wave Dispersion in Certified Phantoms Using a Canon Aplio Clinical Ultrasound Scanner.

The aim of the work described here was to investigate the relative contribution of confounding factors on liver shear wave speed (SWS) and shear wave dispersion slope (SWDS) measurements in three certified phantoms using a Canon Aplio clinical ultrasound scanner. A Canon Aplio i800 i-series ultrasound system (Canon Medical Systems Corporation, Otawara, Tochigi, Japan) with i8CX1 convex array (center frequency=4 MHz) was used to examine dependencies caused by the depth, width and height of the acquisition box (AQB), the depth and size of the region of interest (ROI), the AQB angle and the pressure of the ultrasound probe on the surface of the phantom. Results revealed that depth is the most significant confounder in both SWS and SWDS measurements. AQB angle, height and width and ROI size exhibited minimal confounding effects on measurements. For SWS, the most consistent measurement depth is when the top of the AQB is placed between 2 and 4 cm, and the ROI is located between 3 and 7 cm deep. For SWDS, results indicate that measurement values significantly decrease with depth from the surface of the phantom until approximately 7 cm deep, and consequently no stable area of AQB placement or ROI depth exists. In contrast to SWS, the same ideal acquisition depth range cannot necessarily be applied to SWDS measurements because of a significant depth dependency.

Factors influencing the Shear Wave Elastography evaluation of the patellar tendon.

There is no clear standardization for tendon Shear Wave Elastography (SWE), and data regarding the factors that influence the correctness of evaluation are scarce. We aimed to determine the intra- and interobserver agreement in patellar tendon SWE, and to establish the influence of various factors on elasticity values. We recruited 37 healthy volunteers; SWE of the patellar tendon was performed by two examiners. The following factors were analyzed: probe frequency, degree of joint flexion, size of region of interest (ROI), distance of the color box from the probe footprint, utilization of coupling gel as standoff, and the effect of physical exercise on elastic modulus values. Highest overall interobserver [k=0.767, 95%CI (0.717-0.799), p<0.001] and intraobserver agreement [k=0.920 (0.909-0.929) for examiner 1, k=0.891 (0.875-0.905) for examiner 2] was obtained with the knee in the neutral position, using the L18-5 probe. With the knee flexed at 30º and 45º, the elasticity values were higher compared to the neutral position (p<0.001). With the probe immersed in 0.25 and 0.50 cm coupling gel, the median values were lower compared to the probe placed on the skin (p=0.001, p=0.018). The ROI dimensions and the placement of the SWE box at the level of the skin or at 0.5 cm below skin did not significantly influence the elastic modulus. After physical exercise, the elasticity values decreased in the proximal and middle portion of the tendon (p=0.002, p<0.001). The best results obtained in patellar tendon SWE were with the knee in neutral position, in the proximal or middle tendon, after 10 minutes of relaxation and with the probe placed directly on the skin with minimal pressure. The size and position of ROI do not significantly influence the examination.

Properties of the SSIM metric in medical image assessment: correspondence between measurements and the spatial frequency spectrum.

In radiological imaging, the acquisition of the required diagnostic image quality under optimized conditions is important. Although techniques based on structural similarity (SSIM) have been investigated, concerns have been raised regarding their application to medical images. This study aims to clarify the properties of SSIM as an image quality index in medical images, focusing on digital radiography and verifying the correspondence between the evaluation results obtained by SSIM and the frequency spectrum. The analysis target was chest X-ray images of a human-body phantom. Various types of processing were applied to the images, and several regions of interest (ROIs) were used in local areas for analysis. The SSIM was measured using unprocessed data as a reference while changing the calculation parameters, and the spatial frequency spectrum of each local region was analyzed. Thus, a significant effect of ROI size was observed when calculating the SSIM. This indicates that larger the ROI size leads to SSIM values closer to 1 for all analysis conditions. In addition, a relationship between the size of the ROI in the analysis and the frequency components is demonstrated. It was shown that careful attention should be paid to the structures included in the ROI, and parameter settings should be reconsidered. Furthermore, when using SSIM to assess medical images, a multiscale SSIM method obtained by changing the ROI size would be useful.

Impact of Region-of-Interest Delineation on Stability and Reproducibility of Liver SNR Measurements in 68 Ga-PSMA PET/CT.

Objective This study aims to assess the impact of various regions of interest (ROIs) and volumes of interest (VOIs) delineations on the reproducibility of liver signal-to-noise-ratio (SNRliver) measurements, as well as to find the most reproducible way to estimate it in gallium-68 positron emission tomography ( 68 Ga-PET) imaging. We also investigated the SNRliver-weight relationship for these ROIs and VOIs delineations. Methods A cohort of 40 patients (40 males; mean weight: 76.5 kg [58-115 kg]) with prostate cancer were included. 68 Ga-PET/CT imaging (mean injected activity: 91.4 MBq [51.2 MBq to 134.1 MBq] was performed on a 5-ring bismuth germanium oxide-based Discovery IQ PET/CT using ordered subset expectation maximization image reconstruction algorithm. Afterward, circular ROIs and spherical VOIs with two different diameters of 30 and 40 mm were drawn on the right lobe of the livers. The performance of the various defined regions was evaluated by the average standardized uptake value (SUV mean ), standard deviation (SD) of the SUV (SUV SD ), SNR liver , and SD of the SNR liver metrics. Results There were no significant differences in SUV mean among the various ROIs and VOIs ( p > 0.05). On the other hand, the lower SUV SD was obtained by spherical VOI with diameter of 30 mm. The largest SNR liver was obtained by ROI (30 mm). The SD of SNR liver with ROI (30 mm) was also the largest, while the lowest SD of SNR liver was observed for VOI (40 mm). There is a higher correlation coefficient between the patient-dependent parameter of weight and the image quality parameter of SNRliver for both VOI (30 mm) and VOI (40 mm) compared to the ROIs. Conclusion Our results indicate that SNR liver measurements are affected by the size and shape of the respective ROIs and VOIs. The spherical VOI with a 40 mm diameter leads to more stable and reproducible SNR measurement in the liver.

An ROI Optimization Method Based on Dynamic Estimation Adjustment Model

An important research direction in the field of traffic light recognition of autonomous systems is to accurately obtain the region of interest (ROI) of the image through the multi-sensor assisted method. Dynamic evaluation of the performance of the multi-sensor (GNSS, IMU, and odometer) fusion positioning system to obtain the optimum size of the ROI is essential for further improvement of recognition accuracy. In this paper, we propose a dynamic estimation adjustment (DEA) model construction method to optimize the ROI. First, according to the residual variance of the integrated navigation system and the vehicle velocity, we divide the innovation into an approximate Gaussian fitting region (AGFR) and a Gaussian convergence region (GCR) and estimate them using variational Bayesian gated recurrent unit (VBGRU) networks and a Gaussian mixture model (GMM), respectively, to obtain the GNSS measurement uncertainty. Then, the relationship between the GNSS measurement uncertainty and the multi-sensor aided ROI acquisition error is deduced and analyzed in detail. Further, we build a dynamic estimation adjustment model to convert the innovation of the multi-sensor integrated navigation system into the optimal ROI size of the traffic lights online. Finally, we use the YOLOv4 model to detect and recognize the traffic lights in the ROI. Based on laboratory simulation and real road tests, we verify the performance of the DEA model. The experimental results show that the proposed algorithm is more suitable for the application of autonomous vehicles in complex urban road scenarios than the existing achievements.

Single-shot extended-object super-oscillatory imaging

A point-spread function (PSF) that locally oscillates faster than its highest Fourier component can reconstruct the sub-diffraction information of objects in the far field without any near-field placements to break the diffraction limit. However, the spatial capacity of such super-oscillatory (SO) PSFs for carrying sub-diffraction information is restricted by high sidelobes surrounding a desired region of interest (ROI). Here, we propose generalized periodic SO masks without optimization to push the ROI borders away for imaging extended objects. Our imaging experiments without any image post-processing demonstrate single-shot extended-object SO imaging with extended ROIs more than ten times the size of some typical ROIs (around 2λ/NA) and a sub-wavelength resolution of 0.49λ. The SO sub-wavelength resolution (0.7 times the diffraction limit) remains robust to additive noise with a signal-to-noise ratio above 13 dB. Our method is applicable to lifting the ROI size restrictions for various SO applications such as high-density data storage, acoustic SO imaging, super-narrow frequency conversion, and temporal SO pulses.

Liver Fat Quantification With Ultrasound: Depth Dependence of Attenuation Coefficient.

The primary aim was to estimate the influence of various depths on ultrasound attenuation coefficient (AC) of multiple vendors in the liver. The secondary aim was to evaluate the impact of region of interest (ROI) size on AC measurements in a subset of participants. This Institutional Review Board (IRB)-approved Health Insurance Portability and Accountability Act (HIPAA)-compliant retrospective study was carried out in two centers using AC-Canon and AC-Philips algorithms and extracting AC-Siemens values from ultrasound-derived fat fraction algorithm. Measurements were performed positioning ROI upper edge (3 cm size) at 2, 3, 4, 5 cm from the liver capsule with AC-Canon and AC-Philips and at 1.5, 2, 3 cm with Siemens algorithm. In a subset of participants, measurements were obtained with 1 and 3 cm ROI size. Univariate and multivariate linear regression models and Lin's concordance correlation coefficient (CCC) were used for statistical analysis as appropriate. Three different cohorts were studied. Sixty-three participants (34 females; mean age: 51 ± 14 years) were studied with AC-Canon, 60 (46 females; mean age: 57 ± 11 years) with AC-Philips, and 50 (25 females; 61 ± 13 years) with AC-Siemens. There was a decrease in AC values per 1 cm increase in depth in all. In multivariable analysis, the coefficient was -0.049 (-0.060; -0.038 P < .001) with AC-Canon, -0.058 (-0.066; -0.049 P < .001) with AC-Philips and -0.081 (-0.112; -0.050 P < .001) with AC-Siemens. AC values with 1 cm ROI were significantly higher than those obtained with 3 cm ROI at all depths (P < .001) but the agreement between AC values obtained with different ROI size was excellent (CCC 0.82 [0.77-0.88]). There is depth dependence in AC measurement that affects results. A standardized protocol with fixed ROI's depth and size is needed.

Adrenal lesion classification with abdomen caps and the effect of ROI size

Accurate classification of adrenal lesions on magnetic resonance (MR) images are very important for diagnosis and treatment planning. The detection and classification of lesions in medical imaging heavily rely on several key factors, including the specialist's level of experience, work intensity, and fatigue of the clinician. These factors are critical determinants of the accuracy and effectiveness of the diagnostic process, which in turn has a direct impact on patient health outcomes. With the spread of artificial intelligence, the use of computer-aided diagnosis (CAD) systems in disease diagnosis has also increased. In this study, adrenal lesion classification was performed using deep learning on MR images. The data set used was obtained from the Department of Radiology, Faculty of Medicine, Selcuk University, and all adrenal lesions were identified and reviewed in consensus by two radiologists experienced with abdominal MR. Studies were carried out on two different data sets created by T1- and T2-weighted MR images. The data set consisted of 112 benign and 10 malignant lesions for each mode. Experiments were performed with regions of interest (ROIs) of different sizes to increase the working performance. Thus, the effect of the selected ROI size on the classification performance was assessed. In addition, instead of the convolutional neural network (CNN) models used in deep learning, a unique classification model structure called Abdomen Caps was proposed. When the data sets used in classification studies are manually separated for training, validation, and testing, different results are obtained with different data sets for each stage. To eliminate this imbalance, tenfold cross-validation was used in this study. The best results obtained were 0.982, 0.999, 0.969, 0.983, 0.998, and 0.964 for accuracy, precision, recall, F1-score, area under the curve (AUC) score, and kappa score, respectively.