Search

This article will provide a perspective review of the most extensively investigated deep learning (DL) applications for musculoskeletal disease detection that have the best potential to translate into routine clinical practice over the next decade. Deep learning methods for detecting fractures, estimating pediatric bone age, calculating bone measurements such as lower extremity alignment and Cobb angle, and grading osteoarthritis on radiographs have been shown to have high diagnostic performance with many of these applications now commercially available for use in clinical practice. Many studies have also documented the feasibility of using DL methods for detecting joint pathology and characterizing bone tumors on magnetic resonance imaging (MRI). However, musculoskeletal disease detection on MRI is difficult as it requires multi-task, multi-class detection of complex abnormalities on multiple image slices with different tissue contrasts. The generalizability of DL methods for musculoskeletal disease detection on MRI is also challenging due to fluctuations in image quality caused by the wide variety of scanners and pulse sequences used in routine MRI protocols. The diagnostic performance of current DL methods for musculoskeletal disease detection must be further evaluated in well-designed prospective studies using large image datasets acquired at different institutions with different imaging parameters and imaging hardware before they can be fully implemented in clinical practice. Future studies must also investigate the true clinical benefits of current DL methods and determine whether they could enhance quality, reduce error rates, improve workflow, and decrease radiologist fatigue and burnout with all of this weighed against the costs. [ABSTRACT FROM AUTHOR]

Abstract The Waveflex semi-rigid-dynamic-internal-fixation system shows good short-term effects in the treatment of lumbar degenerative diseases, but there are few long-term follow-up studies, especially for recovery of sagittal balance. Fifty patients with lumbar degenerative diseases treated from January 2016 to October 2017 were retrospectively analysed: 25 patients treated with Waveflex semi-rigid-dynamic-internal-fixation system (Waveflex group) and 25 patients treated with double-segment PLIF (PLIF group). Clinical efficacy was evaluated by Visual Analogue Scale (VAS) and Oswestry Disability Index (ODI). Imaging data before surgery and at 3 months, 1 year, and 5 years postoperatively was used for imaging indicator assessment. Local disc degeneration of the cephalic adjacent segment (including disc height index (DHI), intervertebral foramen height (IFH), and range of motion (ROM)) and overall spinal motor function (including lumbar lordosis (LL), pelvic incidence (PI), sacral slope (SS), pelvic tilt (PT), and |PI-LL|) were analysed. Regarding clinical efficacy, comparison of VAS and ODI scores between the Waveflex and PLIF groups showed no significant preoperative or postoperative differences. The comparison of the objective imaging indicators showed no significant differences in the DHI, IFH, LL, |PI-LL|, and SS values between the Waveflex and PLIF groups preoperatively and 3 months postoperatively (P > 0.05). These values were significantly different at 1 and 5 years postoperatively (P

Purpose: To investigate the effect of postural changes on access for the OLIF of L2 to L5 in patients with degenerative lumbar scoliosis. Methods: Twenty-one individuals with degenerative lumbar scoliosis were chosen at random, 11 with left-sided convexity and 10 with left-sided concavity. Axial T2-weighted images were used to measure the following variables: (1) the distance between the left psoas major muscle and the abdominal aorta; (2) the angle of the surgical access; (3) the distance between the psoas major muscle attachment point and the vertebral body's transverse axis; (4) the region of the psoas major muscle above the vertebrae; and (5) the width-to-thickness ratio. A statistical analysis of the measured parameters was done. Results: The L2-5 segment in the supine position had a significantly longer window distance in the left convex and left concave groups than in the right lateral recumbent posture (P < 0.05). In all segments, the left concave group outperformed the left convex group, which was substantially higher in the right lateral recumbent posture than in the supine position (P < 0.05). After the position change, the spanning area was significantly higher compared to the same segment in the supine position. The psoas major muscle's morphology was stretched. Conclusions: The right lateral recumbent position limits access to OLIF for degenerative lumbar scoliosis, and the "safety window" for OLIF operation in the parietal region is smaller in the left convex group compared to the left concave group, posing a higher risk of intraoperative vascular and neurological injury. [ABSTRACT FROM AUTHOR]

Aim. Using the ultrasound characteristics of the carotid and femoral arteries and a number of laboratory blood parameters, develop an algorithm for non-invasive diagnosis of coronary atherosclerosis.Material and methods. The study included 216 patients (53% men) aged 24-87 years (mean age, 61,5±10,73 years) who were admitted and examined in the hospital of the National Medical Research Center for Therapy and Preventive Medicine in the period of 2016-2019. All patients underwent diagnostic coronary angiography, duplex carotid and femoral ultrasound and biochemical blood tests. In accordance with coronary angiography, 3 groups of patients were formed: 1) without coronary atherosclerosis, 2) with subclinical and 3) severe coronary atherosclerosis, the examination results of which formed the basis of the developed diagnostic algorithm.Results. A stepwise algorithm for non-invasive detection of coronary atherosclerosis has been developed, which includes biochemical blood tests at stage I (glucose, high-sensitivity C-reactive protein, creatinine and adiponectin), a visual scale (VS) at stage II, a combination of clinical and paraclinical parameters (Celermajer test and left atrium antero-­posterior diameter) at stage III and allows to identify patients with varying degrees of coronary onvolvement (including subclinical). The sequential passage of algorithm steps by the patient increases the detection rate of coronary atherosclerosis of any degree by 12,2 times, and by 13,8 times in case of severe involvement.Conclusion. To rationale the use of a combined panel of available investigations, presented as a stepwise diagnostic algorithm, the proposed algorithm should be validated on an independent cohort or in prospective observation of the initial cohort of patients.

Objective To analyze the imaging parameters of unlateralbiportal endoscopic (UBE) in the treatment of lumbar disc herniation (LDH),and to explore the correlation between the imaging parameters and the clinical effect. Methods A total of 58 patients with LDH who underwent UBE in the Department of Spine and Osteopathy of Zhuhai People′s Hospital from March 2021 to March 2022 were retrospectively analyzed. All patients underwent UBE. The preoperative radiographic data of the patients were collected, and the relevant radiographic parameters (lumbar lordosis, disc wedging, slip percentage, height of intervertebral space, Dural SAC area, grade of disc degeneration) were analyzed with the baseline data. The visual analogue scale (VAS) and Oswestry Disability Index were evaluated and their relationship with imaging parameters was analyzedat preoperative and postoperative follow-up. MacNab method was used to evaluate the curative effect at the last follow-up. Results The preoperative dural sac area was negatively correlated with VAS score and ODI score (all P＜0. 05) . The scores of VAS and ODI were significantly decreased at three months after operation and at the last follow-up (P＜0. 05) . At the last follow-up, the effective rate was 94. 8% (55/58) . Multiple linear regression analysis showed that the improvement of dural sac area, degeneration grade had a linear regression relationship with the improvement of VAS and ODI score, the greater the dural sac area improved, the lower the degeneration grade, the better the effect. Conclusion UBE is effective in the treatment of LDH. The radiographic parameters are affected by age, body mass index (BMI) and history of underlying diseases. The improvement of dural sac area and Pfirrmann grade of intervertebral disc are one of the factors affecting the clinical effect. [ABSTRACT FROM AUTHOR]

Abstract Background Diffusion tensor imaging is a promising technique for determining the responsible lesion of cervical radiculopathy, but the selection and delineation of the region of interest (ROI) affect the results. This study explored the impact of different ROI sketching methods on the repeatability and consistency of DTI measurement values in patients with cervical spondylotic radiculopathy (CSR). Methods This retrospective study included CSR patients who underwent DTI imaging. The images were analyzed independently by two radiologists. Four delineation methods were used: freehand method, maximum roundness, quadrilateral method, and multi-point averaging method. They re-examined the images 6 weeks later. The intra-class correlation coefficient (ICC) was used to investigate the consistency between the two measurements and the reproducibility between two radiologists. Results Forty-two CSR patients were included in this study. The distribution of the compressed nerve roots was five C4, eight C5, sixteen C6, eleven C7, and two C8. No differences were found among the four methods in fractional anisotropy (FA) or apparent diffusion coefficient (ADC), irrespective of radiologists (all P>0.05). Similar results were observed between the first and second measurements (all P>0.05), but some significant differences were observed for radiologist 2 for the four-small rounds method (P=0.033). The freehand and single largest circle methods were the two methods with the highest ICC between the two measurements and the two radiologists (all ICC >0.90). Conclusion The freehand and single largest circle methods were the most consistent methods for delineating DTI ROI in patients with CSR.

Simple Summary: Early and accurate assessment of endometrial cancer (EC) aggressiveness is of utmost importance for correct treatment in affected patients. However, features of EC aggressiveness are currently assessable only after surgery. The aim of the present study was to investigate the role of machine learning (ML)-based classification using 18F-FDG PET parameters in preoperatively characterizing and predicting features of EC aggressiveness. Precisely, a signature integrating the most conventional PET parameters and clinical data was built. As a result, the described approach allowed the characterization and prediction of the investigated features of EC aggressiveness, demonstrating how advanced PET image analysis based on conventional quantitative parameters and ML can complement qualitative analysis, supporting the non-invasive preoperative stratification and treatment management of EC patients, in an interpretable and applicable way. Purpose: to investigate the preoperative role of ML-based classification using conventional 18F-FDG PET parameters and clinical data in predicting features of EC aggressiveness. Methods: retrospective study, including 123 EC patients who underwent 18F-FDG PET (2009–2021) for preoperative staging. Maximum standardized uptake value (SUVmax), SUVmean, metabolic tumour volume (MTV), and total lesion glycolysis (TLG) were computed on the primary tumour. Age and BMI were collected. Histotype, myometrial invasion (MI), risk group, lymph-nodal involvement (LN), and p53 expression were retrieved from histology. The population was split into a train and a validation set (80–20%). The train set was used to select relevant parameters (Mann-Whitney U test; ROC analysis) and implement ML models, while the validation set was used to test prediction abilities. Results: on the validation set, the best accuracies obtained with individual parameters and ML were: 61% (TLG) and 87% (ML) for MI; 71% (SUVmax) and 79% (ML) for risk groups; 72% (TLG) and 83% (ML) for LN; 45% (SUVmax; SUVmean) and 73% (ML) for p53 expression. Conclusions: ML-based classification using conventional 18F-FDG PET parameters and clinical data demonstrated ability to characterize the investigated features of EC aggressiveness, providing a non-invasive way to support preoperative stratification of EC patients. [ABSTRACT FROM AUTHOR]

Background: Diffusion tensor imaging is a promising technique for determining the responsible lesion of cervical radiculopathy, but the selection and delineation of the region of interest (ROI) affect the results. This study explored the impact of different ROI sketching methods on the repeatability and consistency of DTI measurement values in patients with cervical spondylotic radiculopathy (CSR).Methods: This retrospective study included CSR patients who underwent DTI imaging. The images were analyzed independently by two radiologists. Four delineation methods were used: freehand method, maximum roundness, quadrilateral method, and multi-point averaging method. They re-examined the images 6 weeks later. The intra-class correlation coefficient (ICC) was used to investigate the consistency between the two measurements and the reproducibility between two radiologists.Results: Forty-two CSR patients were included in this study. The distribution of the compressed nerve roots was five C4, eight C5, sixteen C6, eleven C7, and two C8. No differences were found among the four methods in fractional anisotropy (FA) or apparent diffusion coefficient (ADC), irrespective of radiologists (all P>0.05). Similar results were observed between the first and second measurements (all P>0.05), but some significant differences were observed for radiologist 2 for the four-small rounds method (P=0.033). The freehand and single largest circle methods were the two methods with the highest ICC between the two measurements and the two radiologists (all ICC >0.90).Conclusion: The freehand and single largest circle methods were the most consistent methods for delineating DTI ROI in patients with CSR. [ABSTRACT FROM AUTHOR]

Purpose To provide imaging evidence of the feasibility and clinical efficacy of a new full lamina back shift spinal canal enlargement technique. Methods A retrospective analysis was conducted on 64 patients with multisegment cervical spondylotic myelopathy caused by cervical stenosis. Of these, 32 patients underwent the new full lamina back shift spinal canal enlargement technique (as observation group) and 32 patients underwent single open‐door miniature titanium plate internal fixation (as control group). The computed tomography (CT) data of both groups were imported into Mimics 17.0 software to measure the median sagittal diameter and cross‐sectional area of the spinal canal. Photoshop CS5 was employed to measure the drift distance of the spinal cord on MR images to perform a comparative study of the imaging parameters from the two groups. Results The T2‐weighted MR images in both groups showed continuous recovery of the cerebrospinal fluid signal in the C3–C7 range. The enlarged spinal canal cross‐sectional area (mm2) of each segment after the new full lamina back shift spinal canal enlargement technique was 130.90 ± 20.52 (C3), 180.81 ± 18.86 (C4), 240.48 ± 35.43 (C5), 145.93 ± 36.94 (C6), and 153.16 ± 36.28 (C7), and the enlarged median sagittal diameter (mm) was 5.31 ± 1.13 (C3), 8.8 ± 1.28 (C4), 10.28 ± 1.68 (C5), 9.46 ± 1.48 (C6), and 9.22 ± 1.12 (C7). Both parameters were significantly superior to single open‐door miniature titanium plate internal fixation (P 0.05). Conclusion The new full lamina back shift spinal canal enlargement technique achieved a thorough spinal canal decompression effect on imaging while ensuring a reasonable spinal drift distance and few surgical complications. The clinical curative effect of the new technique was precise.

The existing slender target detection methods based on optical satellite images are greatly affected by the satellite perspective and the solar perspective. Due to limited data sources, it is difficult to implement a fully data-driven approach. This work introduces the imaging parameters of optical satellite images, which greatly reduces the influence of the satellite perspectives and the solar perspectives, and reduces the demand for the amount of data. We improve the oriented bounding box (OBB) detector based on faster R-CNN (region convolutional neural networks) and propose an imaging parameters-considered detector (IPC-Det) which is more suitable for our task. Specifically, in the first stage, the umbra and the shadow are extracted by horizontal bounding box (HBB), respectively, and then the matching of the umbra and the shadow is realized according to the imaging parameters. In the second stage, the paired umbra and shadow features are used to complete the classification and regression, and the target is obtained by OBB. In experiments, after introducing imaging parameters, our detection accuracy is improved by 3.9% (up to 87.5%), proving that this work is a successful attempt to introduce imaging parameters for slender target detection. [ABSTRACT FROM AUTHOR]

Introduction: Micro-CT scanner with a resolution of about 5 micrometers is one of the modalities used to create three-dimensional/two-dimensional images of urinary stones. This study aimed to optimize imaging parameters in micro-computed tomography (CT) scanner based on the signal-to-noise ratio (SNR) of urinary stones for the analysis of stone composition. Material and Methods: In this study,eight micro-CT scanning protocols were applied to five urinary stones taken from different patients. Each scanning protocol had different voltage, current, and exposure parameters. The reconstructed images were then analyzed based on image brightness and SNR. The optimized imaging parameters which were chosen were that having high SNR because the high-quality image has high SNR. Results: The results showed that two groups of urinary stones had the same mean Hounsfield Units (HU) value in the third scanning protocols (i.e., 65 kV, 123 µA, and 850 ms). Mean HU values in group one (i.e., stones numbered 1, 3, and 4) were reported as 790, 760, and 720, respectively. The second group (i.e., stones numbered 2 and 5) had mean HU values of -514 and -343, respectively. The imaging parameters (i.e., 75 kV, 106 µA, and 600 ms) had high SNR (25-34) for the first group. The SNR (12.8-13.25) was for the second group at imaging parameters (i.e., 85 kV, 94 µA, and 500 ms). Conclusion: Based on the SNR, the two optimal imaging parameters for the first and second groups were reported as 75 kV, 106 µA, and 600 ms, as well as 85 kV, 94 µA, and 500 ms, respectively.

Abstract Background The objective of this study was to evaluate the clinical effect and correlation between preoperative imaging parameters and the clinical effect of endoscopic transforaminal decompression for lumbar spinal stenosis. Methods In this prospective study, 87 patients from Shanxi Province People’s Hospital met the criteria for lumbar spinal stenosis and were recruited from June 2014 to January 2016. These patients underwent endoscopic transforaminal decompression. The clinical symptoms were evaluated by VAS, ODI, and claudication at 3 and 6 months after surgery. The overall clinical efficacy was evaluated using the MacNab score. Yellow ligament thickness and area of the dural sac were examined by MRI. Bony vertebral canal area, real spinal canal area, nerve root canal bony area, nerve root canal real area, distance between the articular joints, and vertebral canal sagittal diameter were examined by CT. The soft tissue invasion ratio of the vertebral canal and the invasion ratio of the nerve root canal were calculated. Correlations between imaging parameters and age, sex, and clinical efficacy were examined. Results The MacNab scores were excellent in 47% of cases, good in 34%, generally good in 8%, and poor in 11%. VAS, ODI, and claudication were significantly improved compared with the preoperative values (P

• Pretreatment biomarkers were correlated with novel imaging parameters obtained during treatment in head and neck cancer patients. • CD44 showed a strong correlation with the dose–response matrix parameter indicative of radioresistance. • The data provides further evidence that cancer stem cells are likely responsible for local tumor recurrence. We have developed a novel imaging analysis procedure that is highly predictive of local failure after chemoradiation in head and neck cancer. In this study we investigated whether any pretreatment biomarkers correlated with key imaging parameters. Pretreatment biopsy material was available for 28 patients entered into an institutional trial of adaptive radiotherapy in which FDG-PET images were collected weekly during treatment. The biopsies were immunohistochemically stained for CD44, EGFR, GLUT1, ALDH1, Ki-67 and p53 and quantified using image analysis. Expression levels were correlated with previously derived imaging parameters, the pretreatment SUV max and the dose response matrix (DRM). The different parameters of the SUV max and DRM did not correlate with each other. We observed a positive and highly significant (p = 0.0088) correlation between CD44 expression and volume of tumor with a DRM greater than 0.8. We found no correlation between any DRM parameter and GLUT1, p53, Ki-67 and EGFR or ALDH1. GLUT1 expression did correlate with the maximum SUV 0 and the volume of tumor with an SUV 0 greater than 20. The pretreatment SUV max and DRM are independent imaging parameters that combine to predict local recurrence. The significant correlation between CD44 expression, a known cancer stem cell (CSC) marker, and volume of tumor with a DRM greater than 0.8 is consistent with concept that specific foci of cells are responsible for tumor recurrence and that CSCs may be randomly distributed in tumors in specific niches. Dose painting these small areas may lead to improved tumor control. [ABSTRACT FROM AUTHOR]

Purpose To provide imaging evidence of the feasibility and clinical efficacy of a new full lamina back shift spinal canal enlargement technique. Methods: A retrospective analysis was conducted on 64 patients with multisegment cervical spondylotic myelopathy caused by cervical stenosis. Of these, 32 patients underwent the new full lamina back shift spinal canal enlargement technique (as observation group) and 32 patients underwent single open‐door miniature titanium plate internal fixation (as control group). The computed tomography (CT) data of both groups were imported into Mimics 17.0 software to measure the median sagittal diameter and cross‐sectional area of the spinal canal. Photoshop CS5 was employed to measure the drift distance of the spinal cord on MR images to perform a comparative study of the imaging parameters from the two groups. Results: The T2‐weighted MR images in both groups showed continuous recovery of the cerebrospinal fluid signal in the C3–C7 range. The enlarged spinal canal cross‐sectional area (mm2) of each segment after the new full lamina back shift spinal canal enlargement technique was 130.90 ± 20.52 (C3), 180.81 ± 18.86 (C4), 240.48 ± 35.43 (C5), 145.93 ± 36.94 (C6), and 153.16 ± 36.28 (C7), and the enlarged median sagittal diameter (mm) was 5.31 ± 1.13 (C3), 8.8 ± 1.28 (C4), 10.28 ± 1.68 (C5), 9.46 ± 1.48 (C6), and 9.22 ± 1.12 (C7). Both parameters were significantly superior to single open‐door miniature titanium plate internal fixation (P < 0.05). No significant difference was detected in the drift distance of the spinal cord between the two groups (P > 0.05). Conclusion: The new full lamina back shift spinal canal enlargement technique achieved a thorough spinal canal decompression effect on imaging while ensuring a reasonable spinal drift distance and few surgical complications. The clinical curative effect of the new technique was precise. [ABSTRACT FROM AUTHOR]

Yuluo Rong,* Yongjun Luo,* Wei Liu, Fangyi Gong, Pengyu Tang, Weihua Cai Department of Orthopaedics, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, Jiangsu, China *These authors contributed equally to this work Objective: To investigate the early and mid-term efficacy and safety of the bridge-type ROI-C interbody fusion cage system in the treatment of cervical spondylosis with osteoporosis during anterior cervical discectomy and fusion (ACDF). Patients and methods: The clinical data from 24 cervical spondylosis patients with osteoporosis treated with ACDF were retrospectively analyzed. All patients were treated with ROI-C cage. Double-energy X-ray absorptiometry (DEXA) was used to measure the bone mineral density (BMD). The cervical sagittal radiographic parameters were measured and compared using X-ray including C2–C7 Cobb angle, segmental angle (SA), cervical vertebral bow depth, and height of operation segment (HOS). Postoperative dysphagia was recorded according to the Bazaz score. The Japanese Orthopedic Association (JOA) scores and Neck Disability Index (NDI) scores were used to evaluate the clinical outcomes at different time points. Odom and Vaccaro criteria were used to assess the surgical effects and to evaluate the fusion of the bone graft. Results: The mean duration of the postoperative follow-up was 27.4±5.7 months (ranging from 21 to 36 months). The JOA scores and NDI scores at 3 months post operation and at the time of final follow-up were significantly different from those before surgery (P0.05). Conclusion: The ROI-C cage system is safe and effective for use in patients undergoing anterior cervical spondylosis surgery for osteoporosis treatment. It results in a positive effect on bone graft fusion, is able to effectively improve cervical curvature, restores intervertebral height, and reduces the incidence of postoperative dysphagia. The clinical effects were positive at the early and middle postoperative stages. Keywords: cervical spondylosis, ROI-C, osteoporosis, imaging parameters, dysphagia

BACKGROUND: In the treatment of single- and double-level cervical spondylosis, anterior cervical discectomy and fusion and hybrid surgery can restore local cervical lordosis, and play a role in the overall sequence of the cervical spine, but the role of this conclusion in three-level surgery is still unclear. OBJECTIVE: To compare sagittal alignment, radiological parameters and clinical outcomes between three-level hybrid surgery and anterior cervical discectomy and fusion on cervical spondylotic myelopathy. METHODS: Totally 93 patients with three-level cervical spondylotic myelopathy, who were treated in the Pingdingshan First People’s Hospital from February 2007 to September 2013, were enrolled in this study. Among them, 32 patients underwent anterior cervical discectomy and fusion (anterior cervical discectomy and fusion group); 36 patients underwent single-level disc replacement and double-level hybrid surgery (HS1 group); 25 cases underwent doublelevel disc replacement and single-level hybrid surgery (HS2 group). X-ray films of cervical spine were taken during the follow-up. Sagittal alignment of cervical spine (C2-C7 cervical lordosis, 2-C7 sagittal vertical axis, and T1 slope) and anatomical parameters (range of motion, upper and lower adjacent range of motion, operated-segment lordosis, and adjacent segment degeneration) were analyzed. Neck disability index and Japanese Orthopedic Association score were used to assess clinical outcomes. This study was approved by the Ethics Committee of Pingdingshan First People’s Hospital. RESULTS AND CONCLUSION: Ninety-three patients were followed up for a mean of (76.4±9.0) months. (1) There was no significant difference in parameter values of cervical sagittal alignment among the three groups in the final follow-up (P > 0.05). There was no significant difference in the value of each parameter among the three groups at the final follow-up compared with the change before the operation (P > 0.05). (2) There was no significant difference in anatomical parameter values of the cervical spine at the final follow-up among the three groups (P > 0.05), and there was no significant difference in the values of the parameters among the three groups at the final follow-up compared with the preoperative changes (P > 0.05). (3) The neck disability index and Japanese Orthopedic Association score at the final follow-up of the three groups were significantly improved compared with preoperatively (P < 0.05), but there was no significant difference among the three groups (P > 0.05). (4) Correlation analysis showed that there was no clear correlation of cervical spine anatomical parameters and neck disability index with Japanese Orthopedic Association score between postoperative and final follow-up. (5) The results showed that for three-level cervical spondylotic myelopathy, anterior cervical discectomy and fusion and hybrid surgery have achieved a clear mid-term and long-term effect, and there is no clear correlation between anatomical parameters and clinical outcomes, so artificial disc prosthesis implantation is not necessary. [ABSTRACT FROM AUTHOR]

The existing slender target detection methods based on optical satellite images are greatly affected by the satellite perspective and the solar perspective. Due to limited data sources, it is difficult to implement a fully data-driven approach. This work introduces the imaging parameters of optical satellite images, which greatly reduces the influence of the satellite perspectives and the solar perspectives, and reduces the demand for the amount of data. We improve the oriented bounding box (OBB) detector based on faster R-CNN (region convolutional neural networks) and propose an imaging parameters-considered detector (IPC-Det) which is more suitable for our task. Specifically, in the first stage, the umbra and the shadow are extracted by horizontal bounding box (HBB), respectively, and then the matching of the umbra and the shadow is realized according to the imaging parameters. In the second stage, the paired umbra and shadow features are used to complete the classification and regression, and the target is obtained by OBB. In experiments, after introducing imaging parameters, our detection accuracy is improved by 3.9% (up to 87.5%), proving that this work is a successful attempt to introduce imaging parameters for slender target detection.

Photoacoustic imaging (PAI) has been widely investigated by researchers from a range of areas. According to the differences between excitation laser sources, PAI can be categorized into two main types, namely the time-domain and frequency-domain PAI. Although the frequency-domain approach is more portable and economic than the other alternative, the low intensity of excitation source may lead to a lower signal-to-noise ratio (SNR). This paper aiming to propose the suitable scheme for image reconstruction, we focus on the model-based PAI system and make great efforts to reduce the impact of noise algorithmically. Three regularization algorithms, i.e., Least Square QR-factorization, Tikhonov, and Total Variation minimization by Augmented Lagrangian and alternating direction algorithms (TVAL3) are studied. By choosing three important parameters as criteria, i.e., the peak SNR, image quality index, and time consumption during different situations, the most effective regularization algorithm amongst has been selected. Based on simulation results and detailed discussions, TVAL3 algorithm performs better than the other two for model-based signal reconstruction. The result is pivotal for effective PAI in high quality and highly efficient biomedical tomography and microimaging.

BACKGROUND: MRI has high sensitivity to thoracic myelopathy, which can assess the spinal cord injury by morphology and magnitude of cervical spinal cord compression. Additionally, it is a valuable tool for the prognosis evaluation of thoracic spinal stenosis. OBJECTIVE: To explore the value of quantitative MRI T2WI parameters in predicting surgical outcome of thoracic ossification of the ligamentum flavum, and to establish the prediction model of poor prognosis, so as to provide reference for prognosis evaluation. METHODS: From January 2010 to January 2019 at Cangzhou Central Hospital, clinical and imaging data of 87 cases of thoracic ossification of the ligamentum flavum treated by thoracic laminectomy were reviewed retrospectively. According to the JOA recovery rate at 6-month follow-up, the patients were divided into good recovery group (≥ 50%) and poor recovery group (＜ 50%). Age, sex, duration of disease, JOA score, Sato type of ossification, maximum spinal cord compression, cross-sectional area, distribution of hyperintense signal, signal intensity ratio, intramedullary signal size, local kyphosis, kyphosis correction, number of decompressed levels and incidence of cerebrospinal fluid were compared between two groups. Univariate analysis was used to analyze indicators with significant differences. Receiver operating characteristic curve was plotted to analyze prognosis. Areas under the curve and cut-off values were recorded. The independent predictors of poor recovery were estimated through multivariate logistic regression analysis and the prediction model was established. RESULTS AND CONCLUSION: (1) The duration of disease, JOA score, maximum spinal cord compression, cross-sectional area, signal intensity ratio and intramedullary signal size showed significant difference between good recovery and poor recovery groups (P ＜ 0.05). (2) Receiver operating characteristic curve analysis showed that the area under the curve of the duration of disease, JOA score, maximum spinal cord compression, cross-sectional area, signal intensity ratio and intramedullary signal size was 0.670, 0.733, 0.647, 0.715, 0.753 and 0.765 respectively. The cut-off value was duration of 13 months, score 4, 29.8%, 0.25 cm², 1.593 and 13.64 mm respectively. The duration of disease and maximum spinal cord compression had low discrimination power (the area under the curve ＜ 0.7) in predicting poor recovery, whereas the JOA score, cross-sectional area, signal intensity ratio and intramedullary signal size had moderate discrimination power (the area under the curve 0.7-0.9). The area under the curve indicates good ability of signal intensity ratio and intramedullary signal size in combination (the area under the curve=0.791). (3) Logistic multivariate regression analysis showed that JOA score, cross-sectional area and combination of signal intensity ratio and intramedullary signal size were independent risk factors of poor recovery. A predicting model was built according to the result of the logistic regression analysis. It was shown that the area under the curve of this model was 0.890, which was significantly higher than that of the JOA score, cross-sectional area and combination of signal intensity ratio and intramedullary signal size (P ＜ 0.05). (4) Combination of signal intensity ratio and intramedullary signal size had higher predictive ability than other MRI parameters. JOA score, together with quantitative MRI T2WI parameters may have a better predictive value for the risk of poor recovery in patients with thoracic ossification of the ligamentum flavum. [ABSTRACT FROM AUTHOR]

Acid Sphingomyelinase Deficiency (ASMD), or Niemann-Pick type A/B disease, is a rare lipid storage disorder leading to accumulation of sphingomyelin and its precursors primarily in macrophages. The disease has a broad phenotypic spectrum ranging from a fatal infantile form with severe neurological involvement (the infantile neurovisceral type) to a primarily visceral form with different degrees of pulmonary, liver, spleen and skeletal involvement (the chronic visceral type). With the upcoming possibility of treatment with enzyme replacement therapy, the need for biomarkers that predict or reflect disease progression has increased. Biomarkers should be validated for their use as surrogate markers of clinically relevant endpoints. In this review, clinically important endpoints as well as biochemical and imaging markers of ASMD are discussed and potential new biomarkers are identified. We suggest as the most promising biomarkers that may function as surrogate endpoints in the future: diffusion capacity measured by spirometry, spleen volume, platelet count, low-density lipoprotein cholesterol, liver fibrosis measured with a fibroscan, lysosphingomyelin and walked distance in six minutes. Currently, no biomarkers have been validated. Several plasma markers of lipid-laden cells, fibrosis or inflammation are of high potential as biomarkers and deserve further study. Based upon current guidelines for biomarkers, recommendations for the validation process are provided. • ERT for ASMD is under development, necessitating biomarkers to monitor disease manifestations and response to therapy. • Currently most biochemical and imaging markers in ASMD have not been properly validated as biomarkers of the disease. • The best evaluated potential biomarkers for ASMD are diffusion capacity, spleen volume, platelet count, LDL cholesterol, liver fibrosis measured with fibroscan and LSM. • New markers that may be predictive of irreversible disease should be urgently explored to help decide when to initiate treatment. [ABSTRACT FROM AUTHOR]

Background: The objective of this study was to evaluate the clinical effect and correlation between preoperative imaging parameters and the clinical effect of endoscopic transforaminal decompression for lumbar spinal stenosis.Methods: In this prospective study, 87 patients from Shanxi Province People's Hospital met the criteria for lumbar spinal stenosis and were recruited from June 2014 to January 2016. These patients underwent endoscopic transforaminal decompression. The clinical symptoms were evaluated by VAS, ODI, and claudication at 3 and 6 months after surgery. The overall clinical efficacy was evaluated using the MacNab score. Yellow ligament thickness and area of the dural sac were examined by MRI. Bony vertebral canal area, real spinal canal area, nerve root canal bony area, nerve root canal real area, distance between the articular joints, and vertebral canal sagittal diameter were examined by CT. The soft tissue invasion ratio of the vertebral canal and the invasion ratio of the nerve root canal were calculated. Correlations between imaging parameters and age, sex, and clinical efficacy were examined.Results: The MacNab scores were excellent in 47% of cases, good in 34%, generally good in 8%, and poor in 11%. VAS, ODI, and claudication were significantly improved compared with the preoperative values (P < 0.01). A significant difference was observed between the 71-81 year age group and the other age groups (P < 0.05). There were good correlations between clinical efficacy and vertebral canal sagittal diameter, distance between the articular joints, soft tissue invasion ratio of the vertebral canal, and invasion ratio of the nerve root canal.Conclusion: Treatment of lumbar spinal stenosis by endoscopic transforaminal decompression can achieve good clinical results. This operation is less effective in patients older than 71 years of age. There were positive correlations between clinical efficacy and the vertebral canal sagittal diameter, the articular joints, soft tissue invasion ratio of the vertebral canal, and invasion ratio of the nerve root canal. [ABSTRACT FROM AUTHOR]

GPR has been widely acknowledged as an effective and efficient technique for imaging the subsurface. But the process of constructing 3D GPR images (C-scans) is still subjective and mainly relies upon the operator's knowledge and experience. This study reviews the parameters that affect GPR imaging quality: namely, profile spacing (PS), slice thickness (ST) and interpolations. Feature characteristics that have a crucial influence on imaging quality were also identified. Through conducting 25 carefully designed empirical experiments on concrete as well as subsurface structures, the relationship between 3D imaging parameters and feature characteristics were observed. A general workflow was derived for GPR C-scan generation, which is analogous to the typical signal processing steps used in 2D radargram signal processing (Jol, 2009). Empirical values in workflow were based on the retrieval of known ground-truth data and comparison with the processed images, i.e. the closest to reality. Unlike 2D processing, the workflow for 3D is feature-oriented and case-specific, and the proposed workflow gives guidelines on suitable ranges for 3 major parameters when used in a variety of applications. It was identified that feature shapes and the ratios of feature size to radar footprint are of vital importance. With the proposed flowchart, the often vague "survey experience" is parametrized and standardized, and the upper and lower limits governing the generation of objective and trustworthy 3D GPR images are defined. This workflow for GPR 3D slice imaging also paves the way for GPR feature extraction and change detection commonly adopted in remote sensing. • Empirical experiments that quantified the relationship between GPR imaging parameters and feature characteristics. • Key parameters for 3D GPR imaging parameters are profiles spacing, slice thickness and interpolation methods. • Feature size and GPR wavelength are considered for determining suitable range defined for each imaging parameters. • Create a standardized workflow for 3D GPR depth slice generation. [ABSTRACT FROM AUTHOR]

Due to the specific working mechanism of alternately transmitting and receiving signals between multiple adjacent swaths, scanning synthetic aperture radar (ScanSAR) will cause periodical wavelike modulation of the intensity image along azimuth direction, which is known as scalloping. Conventional descalloping methods are achieved by proper azimuth antenna pattern (AAP) correction and multi-looking techniques but are limited by the accuracy of Doppler centroid estimating and the thermal noise. Another type of method extracts and suppresses the scalloping texture on the image, but the scanning parameters of ScanSAR are insufficiently considered. The period of scalloping on the image is related to the period of switching between subswathes. While the harmonics can be calculated by the period of a periodic signal, an improved frequency filtering method combined with imaging parameters is proposed. The scalloping modulation model of ScanSAR combined with imaging parameters is constructed, and the harmonics of scalloping texture are accurately calculated and filtered. For the low signal to noise ratio (SNR) image, the antenna pattern is modified according to SNR to avoid scalloping caused by noise. For non-uniform scenes, scalloping suppression is achieved by using the scalloping features acquired by uniform scenes. To separate the non-uniform scenes from uniform scenes, we still use the characteristics of harmonics caused by scalloping. Our descalloping method achieves accurate suppression of scalloping without sea-land segmentation and ship mask and the residual scalloping is reduced from 1.0 to 0.3−0.5 dB. The residual scalloping and statistical characteristics of the image are analyzed to demonstrate the effectiveness of the proposed method.

Photoacoustic imaging (PAI) has been widely investigated by researchers from a range of areas. According to the differences between excitation laser sources, PAI can be categorized into two main types, namely the time-domain and frequency-domain PAI. Although the frequency-domain approach is more portable and economic than the other alternative, the low intensity of excitation source may lead to a lower signal-to-noise ratio (SNR). This paper aiming to propose the suitable scheme for image reconstruction, we focus on the model-based PAI system and make great efforts to reduce the impact of noise algorithmically. Three regularization algorithms, i.e., Least Square QR-factorization, Tikhonov, and Total Variation minimization by Augmented Lagrangian and alternating direction algorithms (TVAL3) are studied. By choosing three important parameters as criteria, i.e., the peak SNR, image quality index, and time consumption during different situations, the most effective regularization algorithm amongst has been selected. Based on simulation results and detailed discussions, TVAL3 algorithm performs better than the other two for model-based signal reconstruction. The result is pivotal for effective PAI in high quality and highly efficient biomedical tomography and microimaging. [ABSTRACT FROM AUTHOR]

Purpose: The aim of this study was to determine if functional parameters extracted from the hybrid positron emission tomography/magnetic resonance imaging (PET/MRI) correlate with the immunohistochemical markers of breast cancer (BC) lesions, to assess their ability to predict BC subtype.Methods: This prospective study was approved by the institution’s Ethics Committee, and all patients provided written informed consent. A total of 50 BC patients at diagnosis underwent PET/MRI before pharmacological and surgical treatment. For each primary lesion, the following data were extracted: morphological data including tumour-node-metastasis stage and lesion size; apparent diffusion coefficient (ADC); perfusion data including forward volume transfer constant (Ktrans), reverse efflux volume transfer constant (Kep) and extravascular extracellular space volume (Ve); and metabolic data including standardized uptake value (SUV), lean body mass (SUL), metabolic tumour volume and total lesion glycolysis. Immunohistochemical reports were used to determine receptor status (oestrogen, progesterone, and human epidermal growth factor receptor 2), cellular differentiation status (grade), and proliferation index (Ki67) of the tumour lesions. Correlation studies (Mann-Whitney U test and Spearman’s test), receiver operating characteristic (ROC) curve analysis, and multivariate analysis were performed.Results: Association studies were performed to assess the correlations between imaging and histological prognostic markers of BC. Imaging biomarkers, which significantly correlated with biological markers, were selected to perform ROC curve analysis to determine their ability to discriminate among BC subtypes. SUVmax, SUVmean and SUL were able to discriminate between luminal A and luminal B subtypes (AUCSUVmean = 0.799; AUCSUVmax = 0.833; AUCSUL = 0.813) and between luminal A and nonluminal subtypes (AUCSUVmean = 0.926; AUCSUVmax = 0.917; AUCSUL = 0.945), and the lowest SUV and SUL values were associated with the luminal A subtype. Kepmax was able to discriminate between luminal A and luminal B subtypes (AUC = 0.779), and its highest values were associated with the luminal B subtype. Ktransmax (AUC = 0.881) was able to discriminate between luminal A and nonluminal subtypes, and the highest perfusion values were associated with the nonluminal subtype. In addition, ADC (AUC = 0.877) was able to discriminate between luminal B and nonluminal subtypes, and the lowest ADCmean values were associated with the luminal B subtype. Multivariate analysis was performed to develop a prognostic model, and the best predictive model included Ktransmax and SUVmax parameters.Conclusion: Using multivariate analysis of both PET and MRI parameters, a prognostic model including Ktransmax and SUVmax was able to predict the tumour subtype in 38 of 49 patients (77.6%, p < 0.001), with higher accuracy for the luminal B subtype (86.2%). [ABSTRACT FROM AUTHOR]

Ultrashort TE (UTE) is a relatively new MRI technique that allows for the visualization of tissue structures with short T2 components that show little or no signal on all conventional MR imaging sequences. This technique, to the best of our knowledge, has been described only in adults and uses a half excitation pulse and radial k-space data acquisition to produce echo times of less than 100 microseconds with no need for additional hardware modifications. We describe the feasibility of using a 2-D UTE sequence in vivo on a routine 1.5 T clinical MR scanner to depict various musculoskeletal structures in children. [ABSTRACT FROM AUTHOR]

Purpose To evaluate the imaging performance of an on-board helical kilovoltage computed tomography (kVCT) system mounted on a helical tomotherapy unit for various imaging parameters and setup conditions. Methods Images of a commonly used computed tomography (CT) image quality phantom were acquired while varying the selection of available parameters (anatomy, mode, body size) as well as phantom positioning and size. Image quality metrics (IQM) including noise, uniformity, contrast, CT number constancy, and spatial resolution were compared for parameter and setup variations.  Results The use of fine mode improved noise and contrast metrics by 20-30% compared to normal mode and by nearly a factor of two compared to the coarse mode for otherwise identical protocols. Uniformity, CT number constancy, and spatial resolution were also improved for fine mode. Thorax and pelvis anatomy protocols improved noise, uniformity, and contrast metrics by 10-20% compared to images acquired with head protocols. No significant differences in CT number constancy or spatial resolution were observed regardless of anatomy choice. Increasing body size (milliampere second (mAs)/rotation) improved each image quality metric. Vertical and lateral phantom shifts of up to ±6 cm degraded noise and contrast metrics by up to 30% relative to the isocenter while also worsening uniformity and CT number constancy. IQM were also degraded substantially with the use of annuli to increase the phantom diameter (32 cm vs. 20 cm). Despite variations in image characteristics among the investigated changes, most metrics were within manufacturer specifications when applicable. Conclusion This work demonstrates the dependence of image quality on parameter selection and setup conditions for a helical kVCT system utilized in image-guided and adaptive helical tomotherapy treatments. While the overall image quality is robust to variations in imaging parameters, care should be taken when selecting parameters as patient size increases or positioning moves from the isocenter to ensure adequate image quality is still achieved., Competing Interests: While this project received no external funding, data were collected on a Radixact system (Accuracy, Inc.) provided to UW-Madison under a research agreement (Bayouth, PI). Under this agreement, Accuracy, Inc. was provided the opportunity to review this article prior to submission to ensure that no trade secrets were divulged., (Copyright © 2022, Tegtmeier et al.)

Objectives: Half-Fourier single-shot turbo spin-echo (HASTE) can be applied to diffusion-weighted (DW) imaging of the head and neck. However, HASTE DW imaging has certain drawbacks, such as severe blurring and low signal-to-noise ratio (SNR). The purpose of this study was to evaluate the influence of imaging parameters on image quality using phantoms. We also evaluated the discrimination ability of the apparent diffusion coefficient (ADC) in HASTE DW imaging to assess whether the technique is applicable to head and neck lesions. Methods: The modulation transfer function (MTF), SNR, and ADC were compared using dedicated phantoms to evaluate the influence of matrix size (192 × 192 and 256 × 256) and receiver bandwidth (RBW 200, 400, 600, and 789 Hz/pixel) on HASTE DW images. Results: A wide RBW setting tended to improve the MTF, regardless of the matrix size and phase-encoding direction. In contrast, a wide RBW setting tended to impair the SNR, regardless of the matrix size. At the same RBW setting, the MTF and SNR for a matrix size of 192 × 192 were higher than those for a matrix size of 256 × 256. A wide RBW setting tended to improve the discrimination ability of ADCs among the substances, regardless of the matrix size. Conclusions: A wide RBW and small matrix size improved the MTF and SNR of HASTE DW images. A wide RBW also improved the discrimination ability of the ADC. Therefore, HASTE DWI should be performed with a wide RBW setting. [ABSTRACT FROM AUTHOR]

This paper discusses the role of a computer advisor in MR image acquisition, interpretation and the diagnosis of cerebral disease. The development of an image and statistical database for use in providing advice is described. [ABSTRACT FROM AUTHOR]

In the paper, a preliminary real-time corona discharge imaging system is introduced, and the effects of the imaging parameters, such as electric field intensities and their frequencies, on the resulting images are investigated. From experiments using metal phantoms and plant leaves, it is found that no insulators can be captured by the corona discharge, and the effect of imaging parameters on the image quality is clearly demonstrated. For the future application of this real-time corona discharge imaging system, a composite phantom made of four different metal substances is investigated, along with the various combinations of imaging parameters, and the corona discharge images of a fresh plant leaf are captured as a test version of biomedical objects. The ionisation current, which is important in biomedical applications, is measured in each imaging procedure and estimated by curve-fitting the measured values to the previously known theory of ionisation. [ABSTRACT FROM AUTHOR]

Objectives: To investigate a correlation between 68Ga-DOTATOC PET/MR imaging parameters such as arterial and venous contrast enhancement, diffusion restriction, and maximum standardized uptake value (SUVmax) with histopathological tumor grading in patients with neuroendocrine neoplasms (NEN).Material and Methods: A total of 26 patients with newly diagnosed, therapy-naive neuroendocrine neoplasms (NEN) were enrolled in this prospective study and underwent 68Ga-DOTATOC PET/MRI. Images were evaluated regarding NEN lesion number and location, predominant tumor signal intensity on precontrast T1w and T2w images and on postcontrast arterial and portal venous phase T1w images, apparent diffusion coefficient (ADC) and SUVmax. Histopathological tumor grading was assessed and related to PET/MRI features using Pearson's correlation coefficient and Fisher's exact t-test. A binary logistic regression analysis was performed to evaluate a potential relation with an aggressive tumor biology and odds ratios (OR) were calculated.Results: There was a moderate negative correlation between arterial contrast enhancement and tumor grading (r=-0.35, p = 0.005), while portal venous enhancement showed a weak positive correlation with the Ki-67 index (r = 0.28, p = 0.008) and a non-significant positive correlation with tumor grading (r = 0.19, p = 0.063). Features that were significantly associated with an aggressive tumor biology were the presence of liver metastases (OR 2.6, p = 0.042), T1w hyperintensity in comparison to muscle (OR 12.7, p = 0.0001), arterial phase hyperenhancement (OR 1.4, p = 0.001), diffusion restriction (OR 2.8, p = 0.02) and SUVmax above the hepatic level (OR 7.0, p = 0.001).Conclusion: The study reveals that PET/MRI features might be useful for prediction of NEN grading and thus provide a preliminary assessment of tumor aggressiveness. [ABSTRACT FROM AUTHOR]

The field of radiomics is at the forefront of personalized medicine. However, there is concern that high variation in imaging parameters will impact robustness of radiomic features and subsequently the performance of the predictive models built upon them. Therefore, our review aims to evaluate the impact of imaging parameters on the robustness of radiomic features. We also provide insights into the validity and discrepancy of different methodologies applied to investigate the robustness of radiomic features. We selected 47 papers based on our predefined inclusion criteria and grouped these papers by the imaging parameter under investigation: (i) scanner parameters, (ii) acquisition parameters and (iii) reconstruction parameters. Our review highlighted that most of the imaging parameters are disruptive parameters, and shape along with First order statistics were reported as the most robust radiomic features against variation in imaging parameters. This review identified inconsistencies related to the methodology of the reviewed studies such as the metrics used for robustness, the feature extraction techniques, the reporting style, and their outcome inclusion. We hope this review will aid the scientific community in conducting research in a way that is more reproducible and avoids the pitfalls of previous analyses., (Copyright © 2021 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Due to the specific working mechanism of alternately transmitting and receiving signals between multiple adjacent swaths, scanning synthetic aperture radar (ScanSAR) will cause periodical wavelike modulation of the intensity image along azimuth direction, which is known as scalloping. Conventional descalloping methods are achieved by proper azimuth antenna pattern (AAP) correction and multi-looking techniques but are limited by the accuracy of Doppler centroid estimating and the thermal noise. Another type of method extracts and suppresses the scalloping texture on the image, but the scanning parameters of ScanSAR are insufficiently considered. The period of scalloping on the image is related to the period of switching between subswathes. While the harmonics can be calculated by the period of a periodic signal, an improved frequency filtering method combined with imaging parameters is proposed. The scalloping modulation model of ScanSAR combined with imaging parameters is constructed, and the harmonics of scalloping texture are accurately calculated and filtered. For the low signal to noise ratio (SNR) image, the antenna pattern is modified according to SNR to avoid scalloping caused by noise. For non-uniform scenes, scalloping suppression is achieved by using the scalloping features acquired by uniform scenes. To separate the non-uniform scenes from uniform scenes, we still use the characteristics of harmonics caused by scalloping. Our descalloping method achieves accurate suppression of scalloping without sea-land segmentation and ship mask and the residual scalloping is reduced from 1.0 to 0.3–0.5 dB. The residual scalloping and statistical characteristics of the image are analyzed to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Objectives: To investigate a correlation between 68 Ga-DOTATOC PET/MR imaging parameters such as arterial and venous contrast enhancement, diffusion restriction, and maximum standardized uptake value (SUVmax) with histopathological tumor grading in patients with neuroendocrine neoplasms (NEN)., Material and Methods: A total of 26 patients with newly diagnosed, therapy-naive neuroendocrine neoplasms (NEN) were enrolled in this prospective study and underwent 68 Ga-DOTATOC PET/MRI. Images were evaluated regarding NEN lesion number and location, predominant tumor signal intensity on precontrast T1w and T2w images and on postcontrast arterial and portal venous phase T1w images, apparent diffusion coefficient (ADC) and SUVmax. Histopathological tumor grading was assessed and related to PET/MRI features using Pearson's correlation coefficient and Fisher's exact t-test. A binary logistic regression analysis was performed to evaluate a potential relation with an aggressive tumor biology and odds ratios (OR) were calculated., Results: There was a moderate negative correlation between arterial contrast enhancement and tumor grading (r=-0.35, p = 0.005), while portal venous enhancement showed a weak positive correlation with the Ki-67 index (r = 0.28, p = 0.008) and a non-significant positive correlation with tumor grading (r = 0.19, p = 0.063). Features that were significantly associated with an aggressive tumor biology were the presence of liver metastases (OR 2.6, p = 0.042), T1w hyperintensity in comparison to muscle (OR 12.7, p = 0.0001), arterial phase hyperenhancement (OR 1.4, p = 0.001), diffusion restriction (OR 2.8, p = 0.02) and SUVmax above the hepatic level (OR 7.0, p = 0.001)., Conclusion: The study reveals that PET/MRI features might be useful for prediction of NEN grading and thus provide a preliminary assessment of tumor aggressiveness., (Copyright © 2021 Elsevier B.V. All rights reserved.)