Transverse View Research Articles

Deep Learning for Detecting Supraspinatus Calcific Tendinopathy on Ultrasound Images.

The aim of the study was to evaluate the feasibility of convolutional neural network (CNN)-based deep learning (DL) algorithms to dichotomize shoulder ultrasound (US) images with or without supraspinatus calcific tendinopathy (SSCT). This was a retrospective study pertaining to US examinations that had been performed by 18 physiatrists with 3-20 years of experience. 133,619 US images from 7836 consecutive patients who had undergone shoulder US examinations between January 2017 and June 2019 were collected. Only images with longitudinal or transverse views of supraspinatus tendons (SSTs) were included. During the labeling process, two physiatrists with 6-and 10-year experience in musculoskeletal US independently classified the images as with or without SSCT. DenseNet-121, a pre-trained model in CNN, was used to develop a computer-aided system to identify US images of SSTs with and without calcifications. Testing accuracy, sensitivity, and specificity calculated from the confusion matrix was used to evaluate the models. A total of 2462 images were used for developing the DL algorithm. The longitudinal-transverse model developed with a CNN-based DL algorithm was better for the diagnosis of SSCT when compared with the longitudinal and transverse models (accuracy: 91.32%, sensitivity: 87.89%, and specificity: 94.74%). The developed DL model as a computer-aided system can assist physicians in diagnosing SSCT during the US examination.

Metal Artefact Reduction Sequences (MARS) in Magnetic Resonance Imaging (MRI) after Total Hip Arthroplasty (THA)

BackgroundIn the past, radiographic imaging was of minor relevance in the diagnosis of periprosthetic joint infections (PJI). Since metal artefact reduction sequences (MARS) are available, magnetic resonance imaging (MRI) has become a promising diagnostic tool for the evaluation of hip arthroplasty implants. The purpose of the present study was to evaluate the efficacy of MARS-MRI in comparison to established diagnostic tools to distinguish between aseptic failure and PJI.MethodsFrom July 2018 to September 2019, 33 patients classified as having an aseptic joint effusion were recruited into the study. The group included 22 women and 11 men with a mean age of 70.4 ± 13.7 (42–88) years. In the same period, 12 patients were classified as having a PJI. The group consisted of 9 women and 3 men with a mean age of 72.5 ± 10.6 (54–88) years. MARS-MRI was conducted using the optimized parameters at 1.5 T in a coronal and axial STIR (short-tau-inversion recovery), a non-fat-saturated T2 in coronal view and a non-fat-saturated T1 in transverse view in 45 patients with painful hip after total hip arthroplasty (THA). Normally distributed continuous data were shown as mean ± standard deviation (SD) and compared using student's t-test. Non-normally distributed continuous data were shown as mean and compared using the Mann–Whitney U test.ResultsSynovial layering and muscle edema were significant features of periprosthetic joint infection, with sensitivities of 100% and specifities of 63.0—75.0%. The combined specifity and sensitivity levels of synovial layering and muscular edema was 88.0% and 90.0%. Granulomatous synovitis was a significant feature for aseptic failure, with 90.0% sensitivity and 57.0% specifity.ConclusionMARS-MRI is as suitable as standard diagnostic tools to distinguish between aseptic failure and PJI in patients with THA. Further studies with larger patient numbers have to prove whether MARS-MRI could be integral part of PJI diagnostic.

Ultrasonography and biochemical studies of hepatobiliary system in buffaloes

The present work was undertaken to study the ultrasonographic and clinico-biochemical parameters of hepatobiliary system in apparently healthy adult non gravid buffaloes. The present study revealed that, the hepatic parenchyma was homogenously coarse echogenic, interspersed with anechoic bands of hepatic vessels, with sharp margin and was hyperechoic relative to right renal cortex. It was imaged from just behind the last rib to the 6th intercostal space. The gallbladder was visualized between the 12th to 9th intercostal spaces, seen as a pear-shaped fluid-filled anechoic structure with hyperechoic wall, restricted to one to two intercostal spaces. The portal vein was seen as a stellate, branching, anechoic structure with hyperechoic wall and the hepatic vein as anechoic tubular structure with anechoic wall within the hepatic parenchyma. The caudal vena cava was observed as a triangular anechoic structure in transverse view and tubular in longitudinal view. It is concluded that the ultrasonography is a useful tool for the non-invasive examination of liver. Its sonographic appearance and parameters measured in healthy buffaloes can serve as reference values for the diagnosis of pathological changes in liver.

Increased nuchal translucency can be ascertained using transverse planes

The detection of increased nuchal translucency is crucial for the assessment risk of aneuploidies and other fetal anomalies. This study aimed to investigate the ability of a transverse view of the fetal head to detect increased fetal nuchal translucency at 11 to 13 weeks of gestation. This was a prospective study enrolling a nonconsecutive series of women who attended our outpatient clinic from January 2020 to April 2021 for combined screening and were examined by operators certified by the Fetal Medicine Foundation. In each patient, nuchal translucency measurements were obtained both from a median sagittal view and from a transverse view. A second sonologist blinded to the results of the first examination obtained another measurement to assess intermethod and interobsever reproducibility. A total of 1023 women were enrolled. An excellent correlation was found between sagittal and transverse nuchal translucency measurements, with a mean difference of 0.01 mm (95% confidence interval,-0.01 to 0.02). No systematic difference was found between the 2 techniques. The inter-rater reliability (intraclass correlation coefficient, 0.957; 95% confidence interval, 0.892-0.983) and intrarater reliability (intraclass correlation coefficient, 0.976; 95% confidence interval, 0.941-0.990) of axial measurements were almost perfect. Transverse measurements of 3.0 mm identified all cases with sagittal measurements of ≥3.0 with a specificity of 99.7%; transverse measurements of >3.2 mm identified all cases with sagittal measurements of 3.5 mm with a specificity of 99.7%. The time required to obtain transverse nuchal translucency measurements was considerably shorter than for sagittal measurements, particularly when the fetus had an unfavorable position. When the sonogram is performed by an expert sonologist, the difference in nuchal translucency measurement obtained with a transverse or sagittal plane is minimal. Increased nuchal translucency can be reliably identified by using transverse views, and in some cases, this may technically be advantageous.

PO-718-08 PREDICTING TISSUE CONDUCTANCE CHANGES AND ABLATION LESION PATTERNS USING A QUASI-DYNAMIC PULSED FIELD ELECTROPORATION NUMERICAL MODEL FOR CARDIAC ABLATION

Pulsed field ablation (PFA) is proposed as a useful modality in AF treatment, but factors determining the field distribution, and lesion transmurality and geometry have not been fully characterized. To understand the physics behind the tissue response to PSA, we propose a quasi-dynamic PFA model that can predict the tissue conductivity when electroporation equilibrium is achieved, and identify tissue regions that undergo fully irreversible electroporation (IRE) Model parameters include: tissue conductivity at baseline (0.2 S/m) and after complete electroporation (0.8 S/m), critical depth of the electric field for reversible electroporation (RE, 200 V/m and IRE, 450 V/m), and the applied pulsed voltage amplitude. The iterative model yields the steady-state solution with a tissue conductivity map, clearly identifying regions of IRE. The model employed a circular catheter with 9 - 3 mm electrodes fired sequentially using a 1500 V and 3000 V pulse. The steady-state IRE regions are shown (figure) with an estimated IRE lesion surface area and volume of 780 mm2 and 1411 mm3, respectively, at 1500 V, and 1178 mm2 and 2760 mm3, respectively, at 3000 V. Lesion discontinuity is observed at 5.0 mm depth with 1500 V and 7.2 mm with 3000 V. Transverse views through the lesions show maximum lesion depth at each voltage amplitude. The tissue conductivity profiles through predicted lesions show the formation of gaps in the lesion pattern. (Figure 1) The proposed quasi-dynamic model yields steady-state tissue conductivity maps, shows full IRE tissue regions, and confirms larger lesions with higher pulse amplitudes. Future work will predict time-based tissue response given specific pulse duration and delivery of sequential pulses and compare numerical and experimental results.

Use of Point-of-care upper airway ultrasound during preanesthetic airway assessment to predict difficult intubation in adult patients with obesity

BackgroundUltrasound is a simple, noninvasive complementary tool for preoperative airway assessment. This study compared airway ultrasound and conventional airway assessments to predict difficult intubation. MethodsAirway assessment was performed in 99 patients scheduled for bariatric surgery using the following two parameters: El-Ganzouri score (EGS) and oblique transverse ultrasound view. The following measurements were gathered with ultrasound: pre-epiglottic space depth (pre-E), distance from the epiglottis to the mid-vocal cords (E-VC), mean distance between the skin and epiglottis (mDSE), distance between the skin and hyoid bone (DSHB), and tongue volume. After induction of anesthesia, direct laryngoscopy was performed and Cormack–Lehane (CL) classification was reported.CL grades 1 and 2 were classified as the easy laryngoscopy (EL) group, Grades 3 and 4 were classified as having a difficult laryngoscopy (DL). The crucial performance measure was the E-ability VC's to anticipate difficult intubation. Association between these factors and the CL categorization was a significant indication. ResultsIn 12.12 percent of cases, hard intubation was seen. The DL group had a lower E-VC than the EL group (0.9 ± 0.28 cm vs. 1.52 ± 0.3 cm; P -value less than0.001). The DL cutoff point was <1.22 cm, with 91.7% sensitivity and 83.9% specificity. The CL classification indicated a strongly positive association with the EGS and strongly negative association with the E-VC measurement (r = 0.700 and –0.828, consecutively; P-value less than 0.001), moderately positive correlation with mDSE and DSHB (r = 0.433 and 0.566, consecutively; P-value less than 0.001), and weakly positive correlation with pre-E and tongue volume (r = 0.331 and 0.344, respectively; P-value less than 0.001). ConclusionsUltrasound as a complementary measure to anticipate difficult intubation is beneficial with high sensitivity and low specificity in avoiding the hazards of unanticipated, difficult intraoperative laryngeal visualization.

Feasibility study in assessment of congenital cardiovascular malformation by recent technique of fetal cardiac MR imaging

BackgroundAbnormalities of the cardiovascular system are the most common congenital diseases in the fetus and the first cause of infant mortality. Echocardiography is still the method of choice to visualize the fetal cardiac cardiovascular abnormalities, yet cardiovascular magnetic resonance (CMR) is relatively unaffected by maternal and fetal conditions such as maternal obesity, uterine myoma, twins, oligohydramnios, fetal position and rib calcification, which particularly impair sonographic visualization of the fetal heart. Fetal cardiac MR imaging is a novel MRI technique which can provide valuable information that could add to the prenatal diagnosis and evaluation of cardiac and most of extra-cardiac anomalies. In this work, we aimed to highlight the advantage of FCMRI over fetal echocardiography in assessment of fetal congenital cardiac anomalies.ResultsFifty-Two fetuses with suspected or diagnosed congenital cardiac anomalies. All cases underwent detailed history taking, underwent fetal echocardiography using suitable curvilinear probe and performed according to standard protocol, then fetal cardiac MR was done with Balanced fast field echo, Black blood single shot, white blood Cine, Real-time 3D dynamic sequences, images were acquired in the transverse, four-chamber, short-axis, coronal and oblique sagittal views. Findings of fetal echocardiography were compared with that of cardiac MRI and with standard post-natal echocardiography. Prenatal Echo and fetal cardiac MRI showed significant moderate agreement between the two modalities in the detection of different congenital cardiac anomalies, Kappa test: 0.500; p value 0.021. Fetal MRI had a significant role in detection of extra cardiac anomalies in most cases. Comparing to gold standard post-natal echo. Accuracy of fetal CMRI is 95.5% and of fetal Echo is 86.4% regarding overall cardiac anomalies:ConclusionFetal cardiac MR imaging as an adjunct to fetal echocardiography may provide valuable information that could add to the prenatal diagnosis and evaluation of cardiac and most of extra cardiac anomalies.

Comparison of halo and compression signs assessed by a high frequency ultrasound probe for the diagnosis of Giant Cell Arteritis.

To evaluate the diagnosis performances of halo and compression signs alone and combined, assessed by a high frequency 22-MHz probe, and test their agreement in giant cell arteritis (GCA). In this cross-sectional study on patients suspected with GCA, halo sign was defined as hypo or iso-echogenic circumferential aspect of the vessel wall in transverse or longitudinal view; and compression sign was defined as visibility of the vessel wall upon transducer-imposed compression of the artery. Agreement of the two signs was tested using the Cohen's kappa statistic. A total of 80 patients (50% women) were included with a mean age of 74.4years. Twenty participants (25%) were ultimately treated for GCA. Halo and compression signs have respective prevalences of 35% and 48%, with respective sensitivity and specificity of 80% and 80% for the halo sign; and 85% and 65% for the compression sign. The kappa coefficient for the global agreement of the two signs was 0.67 (95% confident interval: 0.54-0.85). Combination of the two signs give a sensitivity of 80% and a specificity of 81.7%. Halo and compression signs assessed by a high frequency probe, show a good level of agreement for the diagnosis of GCA and improve ultrasound specificity when combined together.

Ultrasound-assisted spinal anesthesia: A randomized comparison between midline and paramedian approaches

Study objectiveNeuraxial ultrasonography can improve the technical performance of spinal anesthesia. However, there are no data regarding the optimal approach for ultrasound-assisted spinal anesthesia. This study aimed to compare the midline and paramedian approaches for ultrasound-assisted spinal anesthesia in adult orthopedic patients. DesignA single-center, prospective, randomized controlled trial. SettingOperating room. PatientsOne hundred and twelve patients undergoing orthopedic surgery were included. InterventionsPatients were randomized to undergo either the midline or paramedian approach for preprocedural ultrasound-assisted spinal anesthesia. MeasurementsThe primary outcome was the number of needle passes required for successful dural puncture. Secondary outcomes were the number of needle insertions, first pass/attempt success rate, procedural time, and the incidence of periprocedural complications. Main resultsThe median number of needle passes was significantly lower in the paramedian group (1 [IQR, 1–2]) than in the midline group (3 [2–6]; P < 0.001). The paramedian approach showed higher first pass/attempt success rates compared with the midline group (58.9% vs. 21.4%; 91.1% vs. 53.6%; both, P < 0.001). Total procedure times were significantly shorter in the paramedian group than in the midline group (113 [72.5–146.5] vs. 196 [138–298.5] seconds; P < 0.001). The quality of sonographic images was graded as good in 94.6% of paramedian sagittal oblique views and 54.5% of transverse median views. No significant intergroup differences were found in the incidence of periprocedural complications. ConclusionsCompared with the midline approach, the ultrasound-assisted paramedian approach showed improved efficacy in terms of the number of needle manipulations, first pass success rates, and procedural time. These results suggest that the paramedian approach may be more efficient for preprocedural ultrasound-assisted spinal anesthesia. Trial registration numberNCT03491943

Improving Image Quality for Single-Angle Plane Wave Ultrasound Imaging With Convolutional Neural Network Beamformer.

Ultrafast ultrasound imaging based on plane wave (PW) compounding has been proposed for use in various clinical and preclinical applications, including shear wave imaging and super resolution blood flow imaging. Because the image quality afforded by PW imaging is highly dependent on the number of PW angles used for compounding, a tradeoff between image quality and frame rate occurs. In the present study, a convolutional neural network (CNN) beamformer based on a combination of the GoogLeNet and U-Net architectures was developed to replace the conventional delay-and-sum (DAS) algorithm to obtain high-quality images at a high frame rate. RF channel data are used as the inputs for the CNN beamformers. The outputs are in-phase and quadrature data. Simulations and phantom experiments revealed that the images predicted by the CNN beamformers had higher resolution and contrast than those predicted by conventional single-angle PW imaging with the DAS approach. In in vivo studies, the contrast-to-noise ratios (CNRs) of carotid artery images predicted by the CNN beamformers using three or five PWs as ground truths were approximately 12 dB in the transverse view, considerably higher than the CNR obtained using the DAS beamformer (3.9 dB). Most tissue speckle information was retained in the in vivo images produced by the CNN beamformers. In conclusion, only a single PW at 0° was fired, but the quality of the output image was proximal to that of an image generated using three or five PW angles. In other words, the quality-frame rate tradeoff of coherence compounding could be mitigated through the use of the proposed CNN for beamforming.

An Artificial Intelligence Ultrasound Platform for Screening and Staging of Thyroid Cancer

<h3>Purpose/Objective(s)</h3> Thyroid cancer is one of the most rapidly increasing cancer in the US, largely due to increased detection. BRAF mutation (V600E) is common in papillary thyroid carcinoma (PTC) and is associated with poor prognosis. The purpose of this study was to establish a multimodal artificial intelligence (AI) ultrasound platform that consists of radiomics, topological data analysis (TDA), TI-RADS features, and deep learning (DL) to predict malignancy and pathological outcome in patients with papillary thyroid cancer. <h3>Materials/Methods</h3> Between 2010 and 2021, 464 patients (511 nodules) underwent fine needle biopsy at our institution. The median age was 49 years. The dataset was divided into an internal training (103 malignant, 259 benign nodules), an internal validation (51 malignant and 98 benign nodules), and an external validation (270 malignant and 50 benign nodules) datasets. High-resolution ultrasonography (sagittal and transverse views) was performed preoperatively for all patients. We employed multimodal machine learning (ML) methods: radiomics (a high-throughput quantitative analysis of image features), TDA (a rapidly growing ML pipeline that analyzes geometric relationships between data points in images), and deep learning (where algorithms run data through layers of neural network to achieve prediction). One hundred and four quantitative radiomics features, 476 topological algebraic-geometric features), and TI-RADS features were obtained for all patients. Linear discriminant analysis (LDA) along with the Pearson correlation coefficient (0.85 threshold) were employed for feature extraction. Support vector machine (SVM) was chosen as the machine learning classifier. <h3>Results</h3> For malignancy prediction, radiomics, TDA, TI-RADS, and deep learning model achieves an accuracy of 88.7% (0.87 AUC), 81.5% (0.81 AUC), 80% (0.76 AUC), and 87.4% (0.92 AUC), respectively. Our multimodal AI platform which utilizes radiomics, TDA, TI-RADS, and deep learning achieves a 98.7% accuracy (0.99 AUC), which is significantly improved when compared to individual model (radiomics, TDA, TI-RADS: p<0.001, deep learning: p=0.002). On the external validation dataset, our model achieves a 91.4% accuracy (0.94 AUC) for malignancy prediction. Using multimodal that consists of radiomics, TDA, and TI-RADS, an accuracy of 93% accuracy (0.93 AUC), 89% accuracy (0.88 AUC), and 98% (0.96 AUC) is achieved for T, N, and extrathyroidal extension, respectively. Employing a 5-fold cross validation, we achieve an accuracy of 96% (0.97 AUC) for predicting BRAF mutation. <h3>Conclusion</h3> We established a robust AI platform for predicting thyroid malignancy, pathological stage, and BRAF mutation using routine ultrasonography. This low-cost reproducible approach would allow developing of personalized treatment planning, especially in countries where resources are limited.

Virtual Compared With In-Clinic Transvaginal Ultrasonography for Ovarian Reserve Assessment.

To evaluate noninferiority of virtual transvaginal ultrasonography compared with in-clinic ultrasonography for ovarian reserve assessment. We conducted a single-site, head-to-head crossover trial. Participants performed self-administered virtual transvaginal ultrasonography at home, guided by a remote-certified ultrasound technologist, then underwent transvaginal ultrasonography in-clinic with another ultrasound technologist. Participants were women in the greater Boston area interested in evaluating ovarian reserve and recruited through social media, health care referrals, and professional networks. The uterus and ovaries were captured in sagittal and transverse views. These randomized recordings were reviewed by two or three independent, blinded reproductive endocrinologists. The primary outcome was noninferiority of the rate of clinical quality imaging produced at home compared with in clinic. Sample size was selected for greater than 90% power, given the 18% noninferiority margin. Secondary outcomes included antral follicle count equivalency and net promoter score superiority. Fifty-six women were enrolled from December 2020 to May 2021. Participants varied in age (19-35 years), BMI (19.5-33.9), and occupation. Ninety-six percent of virtual and 98% of in-clinic images met "clinical quality." The difference of -2.4% (97.5% CI lower bound -5.5%) was within the noninferiority margin (18%). Antral follicle counts were equivalent across settings, with a difference in follicles (0.23, 95% CI -0.36 to 0.82) within the equivalence margin (2.65). Virtual examinations had superior net promoter scores (58.1 points, 97.5% CI of difference 37.3-79.0, P<.01), indicating greater satisfaction with the virtual experience. Virtual transvaginal ultrasonography remotely guided by an ultrasonography technologist is noninferior to in-clinic transvaginal ultrasonography for producing clinical quality images and is equivalent for estimating antral follicle count. Virtual transvaginal ultrasonography had superior patient satisfaction and has potential to significantly expand patient access to fertility care. This study was sponsored by Turtle Health. ClinicalTrials.gov, NCT04687189.

CIRURGIA ENDOSCÓPICA INTERLAMINAR DA COLUNA VERTEBRAL GUIADA POR ULTRASSOM PORTÁTIL: UMA NOVA TÉCNICA

ABSTRACT This paper describes the technique of interlaminar endoscopic surgery guided by a portable ultrasound device. This innovation allows endoscopic surgery to be performed without the use of real-time radiography, which is associated with a higher risk of radiation damage. The portable wireless ultrasound device used for this technique, which has not been yet described in the world literature for minimally invasive surgeries, can be used as an imaging tool to delimit the interlaminar space in minimally invasive surgeries via both transverse and sagittal views. Level of evidence I; Quality of evidence A.

Pharmacognostic, phytochemical analysis and in-vitro antioxidant activity of Senecio edgeworthii Hook plant

The objective of the present study was to investigate pharmacognostic, phytochemical and evaluate in vitro antioxidant properties of Senecio edgeworthii hook Plant. The study of leaf macroscopic features, microscopic features was undertaken and physicochemical parameters were evaluated using standard procedures as per WHO guidelines. In vitro antioxidant activity of S. edgeworthii plant was determined by application of standard methods. In microscopy of S. edgeworthii , leaf is having Upper epidermis, cuticle, multicellular covering trichomes. The transverse view of the lamina shows a single layer of closely packed palisade cells below the upper epidermis. In powder characteristics analysis, lignified xylem parenchyma, lignified fibres, stomata, epidermis, cork cell, xylem vessels, spiral vessels and trichomes were observed. A low amount of total ash acid insoluble ash and water-soluble ash indicate that the inorganic matter and non-physiological matter such as silica is less S. edgeworthii plant. The estimation of total phenolic content in the examined extracts showed SE-Chloro and SE-alcohol extract contains high phenolic content than SE-Oil and SE-Pet ether. The results of antioxidant evaluation based on the three models (DPPH, H 2 O 2 , and NO) used in this study revealed that chloroform and alcohol extract and flower oil of S. edgeworthii plant possess interesting antioxidant activity. The pharmacognostic, phytochemical analysis can contribute to the development of the quality control norms for this species. This plant may possess considerable antioxidant activities.

Cardiac SPECT radiomic features repeatability and reproducibility: A multi-scanner phantom study

BackgroundThe aim of this work was to assess the robustness of cardiac SPECT radiomic features against changes in imaging settings, including acquisition, and reconstruction parameters. MethodsFour commercial SPECT and SPECT/CT cameras were used to acquire images of a static cardiac phantom mimicking typical myorcardial perfusion imaging using 185 MBq of 99mTc. The effects of different image acquisition and reconstruction parameters, including number of views, view matrix size, attenuation correction, as well as image reconstruction related parameters (algorithm, number of iterations, number of subsets, type of post-reconstruction filter, and its associated parameters, including filter order and cut-off frequency) were studied. In total, 5,063 transverse views were reconstructed by varying the aforementioned factors. Eighty-seven radiomic features including first-, second-, and high-order textures were extracted from these images. To assess reproducibility and repeatability, the coefficient of variation (COV), as a widely adopted metric, was measured for each of the radiomic features over the different imaging settings. ResultsThe Inverse Difference Moment Normalized (IDMN) and Inverse Difference Normalized (IDN) features from the Gray Level Co-occurrence Matrix (GLCM), Run Percentage (RP) from the Gray Level Co-occurrence Matrix (GLRLM), Zone Entropy (ZE) from the Gray Level Size Zone Matrix (GLSZM), and Dependence Entropy (DE) from the Gray Level Dependence Matrix (GLDM) feature sets were the only features that exhibited high reproducibility (COV ≤ 5%) against changes in all imaging settings. In addition, Large Area Low Gray Level Emphasis (LALGLE), Small Area Low Gray Level Emphasis (SALGLE) and Low Gray Level Zone Emphasis (LGLZE) from GLSZM, and Small Dependence Low Gray Level Emphasis (SDLGLE) from GLDM feature sets turned out to be less reproducible (COV > 20%) against changes in imaging settings. The GLRLM (31.88%) and GLDM feature set (54.2%) had the highest (COV < 5%) and lowest (COV > 20%) number of the reproducible features, respectively. Matrix size had the largest impact on feature variability as most of the features were not repeatable when matrix size was modified with 82.8% of them having a COV > 20%. ConclusionThe repeatability and reproducibility of SPECT/CT cardiac radiomic features under different imaging settings is feature-dependent. Different image acquisition and reconstruction protocols have variable effects on radiomic features. The radiomic features exhibiting low COV are potential candidates for future clinical studies.

The AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Abdomen and/or Retroperitoneum.

The AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Abdomen and/or Retroperitoneum.

The AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Neonatal and Infant Spine.

The AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Neonatal and Infant Spine.

3D Reconstruction of Carotid Artery from Ultrasound Images.

3D reconstruction is an important area in computer vision, which can be applied to assist in medical diagnosis. Compared to observing 2D ultrasound images, 3D models are more suitable for diagnostic interpretation. In this paper, we describe an approach for 3D reconstruction of the carotid artery utilizing ultrasound images from the transverse and longitudinal views. We implement a human-computer interface to ensure the accuracy of the segmentation results by involving superpixels and ellipse fitting techniques. This approach is expected to achieve better accuracy to assist diagnostics in the future.

A Generalized Approach for Automatic 3-D Geometry Assessment of Blood Vessels in Transverse Ultrasound Images Using Convolutional Neural Networks.

Accurate 3-D geometries of arteries and veins are important clinical data for diagnosis of arterial disease and intervention planning. Automatic segmentation of vessels in the transverse view suffers from the low lateral resolution and contrast. Convolutional neural networks are a promising tool for automatic segmentation of medical images, outperforming the traditional segmentation methods with high robustness. In this study, we aim to create a general, robust, and accurate method to segment the lumen-wall boundary of healthy central and peripheral vessels in large field-of-view freehand ultrasound (US) datasets. Data were acquired using the freehand US, in combination with a probe tracker. A total of ±36 000 cross-sectional images, acquired in the common, internal, and external carotid artery ( N = 37 ), in the radial, ulnar artery, and cephalic vein ( N = 12 ), and in the femoral artery ( N = 5 ) were included. To create masks (of the lumen) for training data, a conventional automatic segmentation method was used. The neural networks were trained on: 1) data of all vessels and 2) the carotid artery only. The performance was compared and tested using an open-access dataset. The recall, precision, DICE, and intersection over union (IoU) were calculated. Overall, segmentation was successful in the carotid and peripheral arteries. The Multires U-net architecture performs best overall with DICE = 0.93 when trained on the total dataset. Future studies will focus on the inclusion of vascular pathologies.

Ultrasonography of the spinal cord and the collection of cerebrospinal fluid from the Atlanto-occipital space in goat.

Ultrasonography, a non-invasive and useful technique, is used for the examination of Atlanto-occipital space structural visualization. The collection of cerebrospinal fluid is more accurate and easier under ultrasound-guided procedure. In this study, longitudinal and transverse views of the Atlanto-occipital space were scanned and their different structural dimensions were measured in sixty healthy Beetal goats. In longitudinal plane, gap between skin and arachnoidea ranged from 8.71 to 10.21mm (mean±SD, 9.76±0.44mm). Depth of the subarachnoid gap dorsal and ventral to the spinal cord ranged from 2.14 to 3.23mm (mean±SD, 2.81±0.33mm) and from 6.09 to 7.68 (mean±SD, 7.02±0.45mm) respectively. Spinal cord diameter varied from 3.76 to 5.26mm (mean±SD, 4.57±0.44mm) and entire dural sac diameter varied from 12.59 to 15.69mm (mean±SD, 14.37±0.74mm). The spinal cord can be seen only in longitudinal plane over a distance of 1.81 to 2.93mm (mean±SD, 2.46±0.35mm). While in the transverse plane, gap between the skin and arachnoidea ranged from 11.01 to 13.11mm (mean±SD, 12.39±0.54mm). Depth of the subarachnoid space dorsal and ventral to spinal cord varied from 5.05 to 6.13mm (mean±SD, 5.59±0.34mm) and 4.12 to 5.25 (mean±SD, 4.65±0.29mm) respectively. Spinal cord diameter ranged from 4.45 to 5.90mm (mean±SD, 5.24±0.44mm) and entire dural sac diameter varied from 14.68 to 16.96mm (mean±SD, 15.58±0.57mm). These standard measurements will be the reference values in healthy Beetal goats. Cerebrospinal fluid was colourless with the quantity of 2-4ml (mean±SD, 3±0.89ml). It was neither turbid nor coagulate. The white blood cell count was 10/µl and red blood cells were not present. Furthermore, total protein and glucose were also measured, which ranged from 23.5 to 28mg/dl (mean±SD, 25.78±2.32mg/dl) and 38-50mg/dl (mean±SD, 43.33±4.60mg/dl) respectively. Ziehl-Neelsen Staining and gram staining were negative.