Comparison of Vendor-Independent Software Tools for Liver Proton Density Fat Fraction Estimation at 1.5 T.

(1) Background: Open-source software tools are available to estimate proton density fat fraction (PDFF). (2) Methods: We compared four algorithms: complex-based with graph cut (GC), magnitude-based (MAG), magnitude-only estimation with Rician noise modeling (MAG-R), and multi-scale quadratic pseudo-Boolean optimization with graph cut (QPBO). The accuracy and reliability of the methods were evaluated in phantoms with known fat/water ratios and a patient cohort with various grades (S0-S3) of steatosis. Image acquisitions were performed at 1.5 Tesla (T). (3) Results: The PDFF estimates showed a nearly perfect correlation (Pearson r = 0.999, p < 0.001) and inter-rater agreement (ICC = from 0.995 to 0.999, p < 0.001) with true fat fractions. The absolute bias was low with all methods (0.001-1%), and an ANCOVA detected no significant difference between the algorithms in vitro. The agreement across the methods was very good in the patient cohort (ICC = 0.891, p < 0.001). However, MAG estimates (-2.30% ± 6.11%, p = 0.005) were lower than MAG-R. The field inhomogeneity artifacts were most frequent in MAG-R (70%) and GC (39%) and absent in QPBO images. (4) Conclusions: The tested algorithms all accurately estimate PDFF in vitro. Meanwhile, QPBO is the least affected by field inhomogeneity artifacts in vivo.

Spontaneous and Treatment-Related Changes of Serum Calcitonin in Medullary Thyroid Cancer: Long-Term Experience in a Patient With Multiple Endocrine Neoplasia Type 2B.

PURPOSE: Medullary thyroid carcinoma (MTC) in MEN2B syndrome is associated with germline RET mutation. Patients harboring de novo mutations are usually diagnosed at more advanced disease stages. We present a young woman with Met918Th mutation diagnosed with stage IV MTC at age 10 years. METHODS: The disease progressed despite total thyroidectomy and multiple surgical interventions for cervical lymph node recurrences, leading to distant metastases in the fifth year after the initial diagnosis. Subsequently, she underwent five different types of tyrosine kinase inhibitor (TKI) treatments. The 17-year disease course was divided into periods defined by four surgical interventions and sequential treatment intervals with four multikinase (sunitinib, vandetanib, cabozantinib, and lenvatinib) and one RET-selective TKI (selpercatinib). Tumor growth for different phases of spontaneous development and drug treatment intervals was characterized by changes in serial log-transformed calcitonin measurements (n = 114). RESULTS: Three operations (one for calcitonin-producing adrenal pheochromocytoma) were associated with drops in calcitonin levels. All of the nonselective TKIs were stopped due to adverse effects. As reflected by the negative calcitonin doubling rate, the best treatment response was observed with selpercatinib, which was associated with an initial large drop followed by a decreasing calcitonin trajectory over 514 days without any major side effects. CONCLUSION: This case of MEN2B medullary thyroid cancer with long-term survival presents how the effectiveness of different treatment modalities can be estimated using log-transformed calcitonin levels. Furthermore, our experience supports the view that serial calcitonin measurements may be more sensitive than radiological follow-up in advanced MTC. Our patient also represents a new case of rarely reported calcitonin-producing pheochromocytomas.

The Calculation and Evaluation of an Ultrasound-Estimated Fat Fraction in Non-Alcoholic Fatty Liver Disease and Metabolic-Associated Fatty Liver Disease.

We aimed to develop a non-linear regression model that could predict the fat fraction of the liver (UEFF), similar to magnetic resonance imaging proton density fat fraction (MRI-PDFF), based on quantitative ultrasound (QUS) parameters. We measured and retrospectively collected the ultrasound attenuation coefficient (AC), backscatter-distribution coefficient (BSC-D), and liver stiffness (LS) using shear wave elastography (SWE) in 90 patients with clinically suspected non-alcoholic fatty liver disease (NAFLD), and 51 patients with clinically suspected metabolic-associated fatty liver disease (MAFLD). The MRI-PDFF was also measured in all patients within a month of the ultrasound scan. In the linear regression analysis, only AC and BSC-D showed a significant association with MRI-PDFF. Therefore, we developed prediction models using non-linear least squares analysis to estimate MRI-PDFF based on the AC and BSC-D parameters. We fitted the models on the NAFLD dataset and evaluated their performance in three-fold cross-validation repeated five times. We decided to use the model based on both parameters to calculate UEFF. The correlation between UEFF and MRI-PDFF was strong in NAFLD and very strong in MAFLD. According to a receiver operating characteristics (ROC) analysis, UEFF could differentiate between <5% vs. ≥5% and <10% vs. ≥10% MRI-PDFF steatosis with excellent, 0.97 and 0.91 area under the curve (AUC), accuracy in the NAFLD and with AUCs of 0.99 and 0.96 in the MAFLD groups. In conclusion, UEFF calculated from QUS parameters is an accurate method to quantify liver fat fraction and to diagnose ≥5% and ≥10% steatosis in both NAFLD and MAFLD. Therefore, UEFF can be an ideal non-invasive screening tool for patients with NAFLD and MAFLD risk factors.

Evaluation of Artificial Intelligence-Calculated Hepatorenal Index for Diagnosing Mild and Moderate Hepatic Steatosis in Non-Alcoholic Fatty Liver Disease.

Background and Objectives: This study aims to evaluate artificial intelligence-calculated hepatorenal index (AI-HRI) as a diagnostic method for hepatic steatosis. Materials and Methods: We prospectively enrolled 102 patients with clinically suspected non-alcoholic fatty liver disease (NAFLD). All patients had a quantitative ultrasound (QUS), including AI-HRI, ultrasound attenuation coefficient (AC,) and ultrasound backscatter-distribution coefficient (SC) measurements. The ultrasonographic fatty liver indicator (US-FLI) score was also calculated. The magnetic resonance imaging fat fraction (MRI-PDFF) was the reference to classify patients into four grades of steatosis: none < 5%, mild 5-10%, moderate 10-20%, and severe ≥ 20%. We compared AI-HRI between steatosis grades and calculated Spearman's correlation (rs) between the methods. We determined the agreement between AI-HRI by two examiners using the intraclass correlation coefficient (ICC) of 68 cases. We performed a receiver operating characteristics (ROC) analysis to estimate the area under the curve (AUC) for AI-HRI. Results: The mean AI-HRI was 2.27 (standard deviation, ±0.96) in the patient cohort. The AI-HRI was significantly different between groups without (1.480 ± 0.607, p < 0.003) and with mild steatosis (2.155 ± 0.776), as well as between mild and moderate steatosis (2.777 ± 0.923, p < 0.018). AI-HRI showed moderate correlation with AC (rs = 0.597), SC (rs = 0.473), US-FLI (rs = 0.5), and MRI-PDFF (rs = 0.528). The agreement in AI-HRI was good between the two examiners (ICC = 0.635, 95% confidence interval (CI) = 0.411-0.774, p < 0.001). The AI-HRI could detect mild steatosis (AUC = 0.758, 95% CI = 0.621-0.894) with fair and moderate/severe steatosis (AUC = 0.803, 95% CI = 0.721-0.885) with good accuracy. However, the performance of AI-HRI was not significantly different (p < 0.578) between the two diagnostic tasks. Conclusions: AI-HRI is an easy-to-use, reproducible, and accurate QUS method for diagnosing mild and moderate hepatic steatosis.

Impact of a Targeted Project for Shortening of Imaging Diagnostic Waiting Time in Patients with Suspected Oncological Diseases in Hungary.

Background and Objectives: Medical imaging is a key element in the clinical workup of patients with suspected oncological disease. In Hungary, due to the high number of patients, waiting lists for Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) were created some years ago. The Municipality of Budapest and Semmelweis University signed a cooperation agreement with an extra budget in 2020 (HBP: Healthy Budapest Program) to reduce the waiting lists for these patients. The aim of our study was to analyze the impact of the first experiences with the HBP. Material and Methods: The study database included all the CT/MRI examinations conducted at Semmelweis University with a referral diagnosis of suspected oncological disease within the first 13 months of the HBP (6804 cases). In our retrospective, two-armed, comparative clinical study, different components of the waiting times in the oncology diagnostics pathway were analyzed. Using propensity score matching, we compared the data of the HBP-funded patients (n = 450) to those of the patients with regular care provided by the National Health Insurance Fund (NHIF) (n = 450). Results: In the HBP-funded vs. the NHIF-funded patients, the time interval from the first suspicion of oncological disease to the request for imaging examinations was on average 15.2 days shorter (16.1 vs. 31.3 days), and the mean waiting time for the CT/MRI examination was reduced by 13.0 days (4.2 vs. 17.2 days, respectively). In addition, the imaging medical records were prepared on average 1.7 days faster for the HBP-funded patients than for the NHIF-funded patients (3.4 vs. 5.1 days, respectively). No further shortening of the different time intervals during the subsequent oncology diagnostic pathway (histological investigation and multidisciplinary team decision) or in the starting of specific oncological therapy (surgery, irradiation, and chemotherapy) was observed in the HBP-funded vs. the NHIF-funded patients. We identified a moderately strong negative correlation (r = -0.5736, p = 0.0350) between the CT/MR scans requested and the active COVID-19 case rates during the pandemic waves. Conclusion: The waiting lists for diagnostic CT/MR imaging can be effectively shortened with a targeted project, but a more comprehensive intervention is needed to shorten the time from the radiological diagnosis, through the decisions of the oncoteam, to the start of the oncological treatment.

Tissue attenuation imaging and tissue scatter imaging for quantitative ultrasound evaluation of hepatic steatosis.

We aimed to assess the feasibility of ultrasound-based tissue attenuation imaging (TAI) and tissue scatter distribution imaging (TSI) for quantification of liver steatosis in patients with nonalcoholic fatty liver disease (NAFLD). We prospectively enrolled 101 participants with suspected NAFLD. The TAI and TSI measurements of the liver were performed with a Samsung RS85 Prestige ultrasound system. Based on the magnetic resonance imaging proton density fat fraction (MRI-PDFF), patients were divided into ≤5%, 5-10%, and ≥10% of MRI-PDFF groups. We determined the correlation between TAI, TSI, and MRI-PDFF and used multiple linear regression analysis to identify any association with clinical variables. The diagnostic performance of TAI, TSI was determined based on the area under the receiver operating characteristic curve (AUC). The intraclass correlation coefficient (ICC) was calculated to assess interobserver reliability. Both TAI (rs = 0.78, P < .001) and TSI (rs = 0.68, P < .001) showed significant correlation with MRI-PDFF. TAI overperformed TSI in the detection of both ≥5% MRI-PDFF (AUC = 0.89 vs 0.87) and ≥10% (AUC = 0.93 vs 0.86). MRI-PDFF proved to be an independent predictor of TAI (ß = 1.03; P < .001), while both MRI-PDFF (ß = 50.9; P < .001) and liver stiffness (ß = -0.86; P < .001) were independent predictors of TSI. Interobserver analysis showed excellent reproducibility of TAI (ICC = 0.95) and moderate reproducibility of TSI (ICC = 0.73). TAI and TSI could be used successfully to diagnose and estimate the severity of hepatic steatosis in routine clinical practice.

Focal Liver Lesion MRI Feature Identification Using Efficientnet and MONAI: A Feasibility Study.

Liver tumors constitute a major part of the global disease burden, often making regular imaging follow-up necessary. Recently, deep learning (DL) has increasingly been applied in this research area. How these methods could facilitate report writing is still a question, which our study aims to address by assessing multiple DL methods using the Medical Open Network for Artificial Intelligence (MONAI) framework, which may provide clinicians with preliminary information about a given liver lesion. For this purpose, we collected 2274 three-dimensional images of lesions, which we cropped from gadoxetate disodium enhanced T1w, native T1w, and T2w magnetic resonance imaging (MRI) scans. After we performed training and validation using 202 and 65 lesions, we selected the best performing model to predict features of lesions from our in-house test dataset containing 112 lesions. The model (EfficientNetB0) predicted 10 features in the test set with an average area under the receiver operating characteristic curve (standard deviation), sensitivity, specificity, negative predictive value, positive predictive value of 0.84 (0.1), 0.78 (0.14), 0.86 (0.08), 0.89 (0.08) and 0.71 (0.17), respectively. These results suggest that AI methods may assist less experienced residents or radiologists in liver MRI reporting of focal liver lesions.

Diagnosis of focal liver lesions with deep learning-based multi-channel analysis of hepatocyte-specific contrast-enhanced magnetic resonance imaging.

BACKGROUND: The nature of input data is an essential factor when training neural networks. Research concerning magnetic resonance imaging (MRI)-based diagnosis of liver tumors using deep learning has been rapidly advancing. Still, evidence to support the utilization of multi-dimensional and multi-parametric image data is lacking. Due to higher information content, three-dimensional input should presumably result in higher classification precision. Also, the differentiation between focal liver lesions (FLLs) can only be plausible with simultaneous analysis of multi-sequence MRI images. AIM: To compare diagnostic efficiency of two-dimensional (2D) and three-dimensional (3D)-densely connected convolutional neural networks (DenseNet) for FLLs on multi-sequence MRI. METHODS: We retrospectively collected T2-weighted, gadoxetate disodium-enhanced arterial phase, portal venous phase, and hepatobiliary phase MRI scans from patients with focal nodular hyperplasia (FNH), hepatocellular carcinomas (HCC) or liver metastases (MET). Our search identified 71 FNH, 69 HCC and 76 MET. After volume registration, the same three most representative axial slices from all sequences were combined into four-channel images to train the 2D-DenseNet264 network. Identical bounding boxes were selected on all scans and stacked into 4D volumes to train the 3D-DenseNet264 model. The test set consisted of 10-10-10 tumors. The performance of the models was compared using area under the receiver operating characteristic curve (AUROC), specificity, sensitivity, positive predictive values (PPV), negative predictive values (NPV), and f1 scores. RESULTS: The average AUC value of the 2D model (0.98) was slightly higher than that of the 3D model (0.94). Mean PPV, sensitivity, NPV, specificity and f1 scores (0.94, 0.93, 0.97, 0.97, and 0.93) of the 2D model were also superior to metrics of the 3D model (0.84, 0.83, 0.92, 0.92, and 0.83). The classification metrics of FNH were 0.91, 1.00, 1.00, 0.95, and 0.95 using the 2D and 0.90, 0.90, 0.95, 0.95, and 0.90 using the 3D models. The 2D and 3D networks' performance in the diagnosis of HCC were 1.00, 0.80, 0.91, 1.00, and 0.89 and 0.88, 0.70, 0.86, 0.95, and 0.78, respectively; while the evaluation of MET lesions resulted in 0.91, 1.00, 1.00, 0.95, and 0.95 and 0.75, 0.90, 0.94, 0.85, and 0.82 using the 2D and 3D networks, respectively. CONCLUSION: Both 2D and 3D-DenseNets can differentiate FNH, HCC and MET with good accuracy when trained on hepatocyte-specific contrast-enhanced multi-sequence MRI volumes.

Laterality of deep white matter hyperintensities correlates with basilar artery bending and vertebral artery dominance.

AIM: To investigate whether vertebrobasilar geometry contributes to the presence, severity, and laterality of white matter hyperintensities (WMH). METHODS: We retrospectively reviewed 290 cerebral scans of patients who underwent time-of-flight and fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) between 2017 and 2018. WMH were counted, localized, and grouped according to laterality on the FLAIR sequence. A 3D mesh of the posterior circulation was reconstructed (with ITK SNAP software) and the morphology of the vertebrobasilar system analyzed with an in-house software written in Python. RESULTS: Patients were assigned into a group with WMH (n=204) and a group without WMH (n=86). The severity of WMH burden was mainly affected by age and hypertension, while the localization of the WMH (or laterality) was mainly affected by the vertebrobasilar system morphology. Basilar artery morphology only affected the parieto-occipital region significantly if both posterior communicating arteries were hypoplastic or absent. The dominant vertebral artery and basilar artery curve had an opposite directional relationship. CONCLUSIONS: An unequal vertebral artery flow is an important hemodynamic contributor to basilar bending. Increased basilar artery curvature and increased infratentorial WMH burden may signal inadequate blood flow and predict cerebrovascular events.

Our initial experiences with mpMRI-ultrasound fusion-guided prostate biopsy / Kezdeti tapasztalataink az mpMR fúziós ultrahangvezérelt prosztatabiopsziával

Introduction: The past decade has seen some major changes in the diagnostics of prostate cancer. Progress in MR imaging has allowed us to better visualise prostate cancer and thus perform targeted biopsies of tumour suspect lesions. mpMRI-ultrasound fusion-guided prostate biopsy is a precise and cost-effective method to diagnose prostate cancer. Objective: The purpose of this study was to summarise our results in mpMRI-ultrasound fusion biopsy between 2017 and 2019 and compare them with the findings in the current literature. Method: Between 2017 and 2019, fully 40, mpMRI-ultrasound fusion biopsies were performed transperineally using the BioJet fusion system at Semmelweis University Urology Clinic. The MRI evaluations were done in line with the PI-RADS v2 guidelines. It was analysed whether the PI-RADS score, the location of the tumour, lesion size, the signs of extraprostatic extension, PSA/PSAD density and prostate volume have an influence on the outcome of mpMRI-ultrasound fusion biopsy. Results: Prostate cancer was diagnosed in 80% of the cases during targeted biopsies. The detection rate was 91%, 85%, and 20% for PI-RADS 5, 4 and 3 lesions, respectively. The detection rate was significantly higher for lesions located at the peripheral zone compared to the ones in the transitional zone (khi2(1) = 6.555, p = 0.010, Fisher-exact p = 0.017, V = 0.355). Signs of extraprostatic extension and higher PSAD correlated with better detection rate (khi2(1)= 7.704, p = 0.006, Fisher-exact p = 0.004, V = 0.355; and 0.47 ± 0.50 ng/ml2 vs. 0.18 ± 0.17 ng/ml2; Z = 3.447, p<0.001, respectively). The size of the lesions did not influence the outcome. The analysis showed a significant correlation between large prostate volumes and negative biopsies (50.9 ± 18.8 ml vs. 119.6 ± 91.6 ml; Z= ­3.505, p<0.001). Conclusions: The detection rate of prostate cancer with targeted biopsies was higher than the data found in the international literature. The PI-RADS score, the location of the tumour, MRI signs of extraprostatic extension, PSAD and prostate volume had an influence on the detection rate. Our findings may promote a better selection of the best candidates for targeted biopsies in the future.

Our initial experiences with MR arthrography / Kezdeti tapasztalataink az MR-artrográfiával.

INTRODUCTION: The gold standard method for shoulder surgery imaging is magnetic resonance imaging (MRI). The sensitivity of the conventional MRI for full thickness tears is way above 85%, for partial rotator cuff lesions this number is less than 40%. Diagnostic accuracy is equally low in detecting different variations of anterior labrum lesion and also in injury of the glenohumeral ligaments. Contrast agent given directly into the joint can improve these diagnostic inaccuracies. METHOD: We made the first shoulder joint related MR arthrography in February 2018 at the Semmelweis University. The studied period lasted for 17 months; during that time 29 examinations were performed. For the injection we used gadolinium-based contrast substance. RESULTS: In the case of 26 patients (89.9%), the contrast substance had a good position and distended well. From all of the shoulder MRI examinations, 9 (31%) surgical interventions were made, and 8 surgical findings had the same diagnosis as the radiologic description. CONCLUSION: The contrast substance given into the intraarticular space makes it easier to identify and detect the structures in the joint. The injection under X-ray screening is safe, it is absorbed from the joint in short time, and so far - based on our experience - there has not been any complication. It can help to identify injuries which are difficult to diagnose with conventional MRI, and helps planning before surgery. Orv Hetil. 2020; 161(36): 1514-1521.

Safety and Effectiveness of UFE in Fibroids Larger than 10 cm.

INTRODUCTION: Early literature suggested that the size of the uterus, the size of the dominant fibroid, and the amount of applied embolization particles would be the risk factors for major postprocedural complications, but recent publications have confuted these early results. The purpose of our study was to evaluate whether the size of the dominant fibroid would influence the complication rate and effectiveness in a large single-center cohort. PATIENTS AND METHODS: From 28 April 2008 until 31 December 2012, 303 patients had uterine artery embolization (UAE). 262 patients had small [largest diameter <10 cm (Group 1)], 41 patients had large [largest diameter >10 cm (Group 2)] fibroid. UAE was performed from unilateral femoral access using 500-710 and 355-500 µm polyvinyl alcohol particles. Periprocedural and postprocedural complications and numerical analog quality-of-life scores (0-unbearable symptoms; 100-perfect quality of life) were listed and statistically analyzed. RESULTS: During the mean follow-up time [7.79 ± 5.16 (SD) month], data on 275 patients (275/303 = 90.8 %) were available. Quality-of-life score was 33.3 ± 23.5 and 33.5 ± 24.1 before, whereas 85.6 ± 16.0 and 81.5 ± 23.5 after UAE in Group 1 and Group 2, respectively, (Mann-Whitney U test one-sided, p = 0.365). There were 4 myoma expulsions, 1 acute myomectomy, and 2 acute hysterectomies reported from Group 1, meanwhile 1 myoma expulsion, 1 acute myomectomy, and 2 acute hysterectomies were documented from Group 2 (NS differences). CONCLUSION: There was no significant difference in the effectiveness and in the number of minor and major complications between fibroids with <10 cm largest diameter compared to those >10 cm.

Diffusion weighted magnetic resonance imaging demonstrates tumor response following palliative embolization of a recurrent shoulder plasmacytoma.

We report the palliative embolization and functional imaging follow-up of a recurrent shoulder plasmacytoma. The multiple myeloma patient complained of severe pain and discomfort, while he could not tolerate further chemotherapy. The left shoulder lesion had earlier received a high dose of irradiation. Thus, the well-vascularized lesion was embolized via feeding arteries branching off from the left subclavian artery in two sessions. The patient's symptoms rapidly improved post-embolization and the serum free light chain ratio stabilized at a lower level. The follow-up magnetic resonance image showed increased diffusivity in previously restricted tumor foci. This has negatively correlated with the decreased fludeoxyglucose uptake on PET, suggesting post-embolization necrosis.

FACE-R--a 3D database of 400 living individuals' full head CT- and face scans and preliminary GMM analysis for craniofacial reconstruction.

In the past, improvements in craniofacial reconstructions (CFR) methodology languished due to the lack of adequate 3D databases that were sufficiently large and appropriate for 3-dimensional shape statistics. In our study, we created the "FACE-R" database from CT records and 3D surface scans of 400 clinical patients from Hungary, providing a significantly larger sample that was available before. The uniqueness of our database is linking of two data types that makes possible to investigate the bone and skin surface of the same individual, in upright position, thus eliminating many of the gravitational effects on the face during CT scanning. We performed a preliminary geometric morphometric (GMM) study using 3D data that produces a general idea of skull and face shape correlations. The vertical position of the tip of the (soft) nose for a skull and landmarks such as rhinion need to be taken into account. Likewise, the anterior nasal spine appears to exert some influence in this regard.

Cerebral iodized lipid embolization via a pulmonary arteriovenous shunt: rare complication of transcatheter arterial embolization for hepatocellular carcinoma.

We report the first European case of cerebral iodized lipid embolism post transcatheter arterial embolization for hepatocellular carcinoma. Lipiodol emboli and corresponding multifocal brain ischemia were documented with computed tomography (CT) and magnetic resonance (MR) in the acutely symptomatic patient. Transcranial Doppler sonography with contrast indicated a right-to-left shunt, while on a follow-up CT scan lipiodol embolization was detected in both lungs. Dilated pulmonary vessels and thick vascular channels were seen in the vicinity of the right diaphragm suggestive of pulmonary arteriovenous shunt. The patient symptoms regressed with supportive care alone, but he died 5 months later due to hepatic failure unrelated to the procedure.