Joint liver and hepatic lesion segmentation in MRI using a hybrid CNN with transformer layers.

Backgound and Objective: Deep learning-based segmentation of the liver and hepatic lesions therein steadily gains relevance in clinical practice due to the increasing incidence of liver cancer each year. Whereas various network variants with overall promising results in the field of medical image segmentation have been successfully developed over the last years, almost all of them struggle with the challenge of accurately segmenting hepatic lesions in magnetic resonance imaging (MRI). This led to the idea of combining elements of convolutional and transformer-based architectures to overcome the existing limitations. METHODS: This work presents a hybrid network called SWTR-Unet, consisting of a pretrained ResNet, transformer blocks as well as a common Unet-style decoder path. This network was primarily applied to single-modality non-contrast-enhanced liver MRI and additionally to the publicly available computed tomography (CT) data of the liver tumor segmentation (LiTS) challenge to verify the applicability on other modalities. For a broader evaluation, multiple state-of-the-art networks were implemented and applied, ensuring direct comparability. Furthermore, correlation analysis and an ablation study were carried out, to investigate various influencing factors on the segmentation accuracy of the presented method. RESULTS: With Dice similarity scores of averaged 98±2% for liver and 81±28% lesion segmentation on the MRI dataset and 97±2% and 79±25%, respectively on the CT dataset, the proposed SWTR-Unet proved to be a precise approach for liver and hepatic lesion segmentation with state-of-the-art results for MRI and competing accuracy in CT imaging. CONCLUSION: The achieved segmentation accuracy was found to be on par with manually performed expert segmentations as indicated by inter-observer variabilities for liver lesion segmentation. In conclusion, the presented method could save valuable time and resources in clinical practice.

Advanced Deep Learning Approach to Automatically Segment Malignant Tumors and Ablation Zone in the Liver With Contrast-Enhanced CT.

OBJECTIVE: Liver cancer is one of the most commonly diagnosed cancer, and energy-based tumor ablation is a widely accepted treatment. Automatic and robust segmentation of liver tumors and ablation zones would facilitate the evaluation of treatment success. The purpose of this study was to develop and evaluate an automatic deep learning based method for (1) segmentation of liver and liver tumors in both arterial and portal venous phase for pre-treatment CT, and (2) segmentation of liver and ablation zones in both arterial and portal venous phase for after ablation treatment. MATERIALS AND METHODS: 252 CT images from 63 patients undergoing liver tumor ablation at a large University Hospital were retrospectively included; each patient had pre-treatment and post-treatment multi-phase CT images. 3D voxel-wise manual segmentation of the liver, tumors and ablation region by the radiologist provided reference standard. Deep learning models for liver and lesion segmentation were initially trained on the public Liver Tumor Segmentation Challenge (LiTS) dataset to obtain base models. Then, transfer learning was applied to adapt the base models on the clinical training-set, to obtain tumor and ablation segmentation models both for arterial and portal venous phase images. For modeling, 2D residual-attention Unet (RA-Unet) was employed for liver segmentation and a multi-scale patch-based 3D RA-Unet for tumor and ablation segmentation. RESULTS: On the independent test-set, the proposed method achieved a dice similarity coefficient (DSC) of 0.96 and 0.95 for liver segmentation on arterial and portal venous phase, respectively. For liver tumors, the model on arterial phase achieved detection sensitivity of 71%, DSC of 0.64, and on portal venous phase sensitivity of 82%, DSC of 0.73. For liver tumors >0.5cm3 performance improved to sensitivity 79%, DSC 0.65 on arterial phase and, sensitivity 86%, DSC 0.72 on portal venous phase. For ablation zone, the model on arterial phase achieved detection sensitivity of 90%, DSC of 0.83, and on portal venous phase sensitivity of 90%, DSC of 0.89. CONCLUSION: The proposed deep learning approach can provide automated segmentation of liver tumors and ablation zones on multi-phase (arterial and portal venous) and multi-time-point (before and after treatment) CT enabling quantitative evaluation of treatment success.

Value of spectral detector computed tomography for the early assessment of technique efficacy after microwave ablation of hepatocellular carcinoma.

OBJECTIVES: To investigate whether virtual monoenergetic images (VMI) and iodine maps derived from spectral detector computed tomography (SDCT) improve early assessment of technique efficacy in patients who underwent microwave ablation (MWA) for hepatocellular carcinoma (HCC) in liver cirrhosis. METHODS: This retrospective study comprised 39 patients with 49 HCC lesions treated with MWA. Biphasic SDCT was performed 7.7±4.0 days after ablation. Conventional images (CI), VMI and IM were reconstructed. Signal- and contrast-to-noise ratio (SNR, CNR) in the ablation zone (AZ), hyperemic rim (HR) and liver parenchyma were calculated using regions-of-interest analysis and compared between CI and VMI between 40-100 keV. Iodine concentration and perfusion ratio of HR and residual tumor (RT) were measured. Two readers evaluated subjective contrast of AZ and HR, technique efficacy (complete vs. incomplete ablation) and diagnostic confidence at determining technique efficacy. RESULTS: Attenuation of liver parenchyma, HR and RT, SNR of liver parenchyma and HR, CNR of AZ and HR were significantly higher in low-keV VMI compared to CI (all p<0.05). Iodine concentration and perfusion ratio differed significantly between HR and RT (all p<0.05; e.g. iodine concentration, 1.6±0.5 vs. 2.7±1.3 mg/ml). VMI50keV improved subjective AZ-to-liver contrast, HR-to-liver contrast, visualization of AZ margin and vessels adjacent to AZ compared to CI (all p<0.05). Diagnostic accuracy for detection of incomplete ablation was slightly higher in VMI50keV compared to CI (0.92 vs. 0.89), while diagnostic confidence was significantly higher in VMI50keV (p<0.05). CONCLUSIONS: Spectral detector computed tomography derived low-keV virtual monoenergetic images and iodine maps provide superior early assessment of technique efficacy of MWA in HCC compared to CI.

Microwave ablation enhances tumor-specific immune response in patients with hepatocellular carcinoma.

Thermal ablative therapies are standard treatments for localized hepatocellular carcinoma (HCC). In addition to local tumor destruction, ablation leads to abscopal effects in distant lesions most likely mediated by an anti-tumor immune response. Although microwave ablation (MWA) is increasingly substituting other ablative techniques, its systemic immunostimulatory effects are poorly studied. We analyzed tumor-specific immune responses in peripheral blood of HCC patients after thermal ablation with regard to T cell responses and disease outcome. While comprehensive flow cytometric analyses in sequential samples of a prospective patient cohort (n = 23) demonstrated only moderate effects of MWA on circulating immune cell subsets, fluorospot analyses of specific T cell responses against seven tumor-associated antigens (TTAs) revealed de-novo or enhanced tumor-specific immune responses in 30% of patients. This anti-tumor immune response was related to tumor control as Interferon-y and Interleukin-5 T cell responses against TAAs were more frequent in patients with a long-time remission (> 1 year) after MWA (7/16) compared to patients suffering from an early relapse (0/13 patients) and presence of tumor-specific T cell response (IFN-y and/or IL-5) was associated to longer progression-free survival (27.5 vs. 10.0 months). Digital image analysis of immunohistochemically stained archival HCC samples (n = 18) of patients receiving combined MWA and resection revealed a superior disease-free survival of patients with high T cell abundance at the time of thermal ablation (37.4 vs. 13.1 months). Our data demonstrates remarkable immune-related effects of MWA in HCC patients and provides additional evidence for a combination of local ablation and immunotherapy in this challenging disease.

Structured Reporting of Solid and Cystic Pancreatic Lesions in CT and MRI: Consensus-Based Structured Report Templates of the German Society of Radiology (DRG). / Strukturierte Befundung von soliden und zystischen Pankreasläsionen im CT und MRT: Konsensus-basierte Befundvorlagen der Deutschen Röntgengesellschaft (DRG).

BACKGROUND: Radiological reports of pancreatic lesions are currently widely formulated as free texts. However, for optimal characterization, staging and operation planning, a wide range of information is required but is sometimes not captured comprehensively. Structured reporting offers the potential for improvement in terms of completeness, reproducibility and clarity of interdisciplinary communication. METHOD: Interdisciplinary consensus finding of structured report templates for solid and cystic pancreatic tumors in computed tomography (CT) and magnetic resonance imaging (MRI) with representatives of the German Society of Radiology (DRG), German Society for General and Visceral Surgery (DGAV), working group Oncological Imaging (ABO) of the German Cancer Society (DKG) and other radiologists, oncologists and surgeons. RESULTS: Among experts in the field of pancreatic imaging, oncology and pancreatic surgery, as well as in a public online survey, structured report templates were developed by consensus. These templates are available on the DRG homepage under www.befundung.drg.de and will be regularly revised to the current state of scientific knowledge by the participating specialist societies and responsible working groups. CONCLUSION: This article presents structured report templates for solid and cystic pancreatic tumors to improve clinical staging (cTNM, ycTNM) in everyday radiology. KEY POINTS: · Structured report templates offer the potential of optimized radiological reporting with regard to completeness, reproducibility and differential diagnosis.. · This article presents consensus-based, structured reports for solid and cystic pancreatic lesions in CT and MRI.. · These structured reports are available open source on the homepage of the German Society of Radiology (DRG) under www.befundung.drg.de.. CITATION FORMAT: · Persigehl T, Baumhauer M, Baeßler B et al. Structured Reporting of Solid and Cystic Pancreatic Lesions in CT and MRI: Consensus-Based Structured Report Templates of the German Society of Radiology (DRG). Fortschr Röntgenstr 2020; 192: 641 - 655.

Respiratory motion artefacts in Gd-EOB-DTPA (Primovist/Eovist) and Gd-DOTA (Dotarem)-enhanced dynamic phase liver MRI after intensified and standard pre-scan patient preparation: A bi-institutional analysis.

OBJECTIVE: The objective of this study is to evaluate if intensified pre-scan patient preparation (IPPP) that comprises custom-made educational material on dynamic phase imaging and supervised pre-imaging breath-hold training in addition to standard informative conversation with verbal explanation of breath-hold commands (standard pre-scan patient preparation-SPPP) might reduce the incidence of gadoxetate disodium (Gd-EOB-DTPA)-related transient severe respiratory motion (TSM) and severity of respiratory motion (RM) during dynamic phase liver MRI. MATERIAL AND METHODS: In this bi-institutional study 100 and 110 patients who received Gd-EOB-DTPA for dynamic phase liver MRI were allocated to either IPPP or SPPP at site A and B. The control group comprised 202 patients who received gadoterate meglumine (Gd-DOTA) of which each 101 patients were allocated to IPPP or SPPP at site B. RM artefacts were scored retrospectively in dynamic phase images (1: none- 5: extensive) by five and two blinded readers at site A and B, respectively, and in the hepatobiliary phase of the Gd-EOB-DTPA-enhanced scans by two blinded readers at either site. RESULTS: The incidence of TSM was 15% at site A and 22.7% at site B (p = 0.157). IPPP did not reduce the incidence of TSM in comparison to SPPP: 16.7% vs. 21.6% (p = 0.366). This finding was consistent at site A: 12% vs. 18% (p = 0.401) and site B: 20.6% vs. 25% (p = 0.590). The TSM incidence in patients with IPPP and SPPP did not differ significantly between both sites (p = 0.227; p = 0.390). IPPP did not significantly mitigate RM in comparison to SPPP in any of the Gd-EOB-DTPA-enhanced dynamic phases and the hepatobiliary phase in patients without TSM (all p≥0.072). In the Gd-DOTA control group on the other hand, IPPP significantly mitigated RM in all dynamic phases in comparison to SPPP (all p≤0.031). CONCLUSIONS: We conclude that Gd-EOB-DTPA-related TSM cannot be mitigated by education and training and that Gd-EOB-DTPA-related breath-hold difficulty does not only affect the subgroup of patients with TSM or exclusively the arterial phase as previously proposed.

Iodine overlays to improve differentiation between peritoneal carcinomatosis and benign peritoneal lesions.

OBJECTIVES: Peritoneal carcinomatosis (PC) is a prognostically relevant metastatic disease which may be difficult to depict in postoperative patients, particularly in early stages. This study aimed to determine whether PC could be diagnosed more accurately when using a combination of spectral detector CT (SDCT)-derived conventional images (CI) and iodine overlay images (IO) compared with CI only. METHODS: Thirty patients with PC and 30 patients with benign peritoneal alterations (BPA) who underwent portal-venous abdominal SDCT were included. Four radiologists determined the presence/absence of PC for each patient and assessed lesion conspicuity, diagnostic certainty, and image quality using 5-point Likert scales. Subjective assessment was conducted in two sessions comprising solely CI and CI/IO between which a latency of 6 weeks was set. Iodine uptake and HU attenuation were determined ROI-based to analyze quantitative differentiation of PC/BPA. RESULTS: Specificity for PC was significantly higher when using CI/IO compared with using CI only (0.86 vs. 0.78, p ≤ 0.05), while sensitivity was comparable (0.79 vs. 0.81, p = 1). In postoperative patients, the increase in specificity was the highest (0.93 vs. 0.80, p ≤ 0.05). Lesion conspicuity was rated higher in CI/IO (4 (3-5)) compared with that in CI only (3 (3-4); p ≤ 0.05). Diagnostic certainty was comparable (both 4 (3-5); p = 0.5). CI/IO received the highest rating for overall image quality and assessability (CI/IO 5 (4-5) vs. CI 4 (4-4) vs. IO 4 (3-4); p ≤ 0.05). Area under the receiver operating characteristics curve (AUC) for quantitative differentiation between PC and BPA was higher for iodine (AUCIodine = 0.95, AUCHU = 0.90). CONCLUSIONS: Compared with CI, combination of CI/IO improves specificity in the assessment of peritoneal carcinomatosis at comparable sensitivity, particularly in postoperative patients. KEY POINTS: • Combination of iodine overlays and conventional images improves specificity when assessing patients with peritoneal carcinomatosis at comparable sensitivity. • Particularly in postsurgical patients, iodine overlays could help to avoid false-positive diagnosis of peritoneal disease. • Iodine overlays alone provided inferior image quality and assessability than conventional images, while the combination of both received the highest ratings. Iodine overlays should therefore be used in addition to and not as a substitute for conventional images.

Value of spectral detector computed tomography for assessment of pancreatic lesions.

PURPOSE: Dual energy CT (DECT) can contribute to the diagnosis of benign and malignant pancreatic lesions. This study examined whether a novel, detector-based spectral CT scanner (SDCT) may improve subjective assessment of different types of pancreatic lesions and if various quantitative maps may improve lesion contrast and differentiation. MATERIALS AND METHODS: 61 consecutive patients who underwent clinical, contrast-agent enhanced, abdominal SDCT scans and showed pancreatic lesions of different origins were included. Subjective image analysis was performed by two readers who assessed image quality, lesion conspicuity and diagnostic confidence on 5-point Likert scales for conventional polyenergetic reconstructions (polyE), virtual monoenergetic images (monoE), virtual non-contrast images, iodine density, iodine overlay, and Z effective (Zeff) maps. Two readers acquired quantitative values from these maps ROI-based from which contrast-to-noise and lesion-to-parenchyma ratios were calculated. RESULTS: MonoE images at low keV levels yielded highest Likert scores regarding lesion conspicuity and reader confidence; iodine overlays facilitated lesion delineation. Inter-observer agreement ranged between substantial and excellent (kappa values 0.73-0.81). Contrast-to-noise-ratios for low keV monoE images were significantly higher, compared to polyE images (e.g. monoE 40 keV p < 0.0001). Marked overlap between PDAC and miscellaneous non-PDAC lesions was present in various spectral reconstructions. CONCLUSIONS: In line with previous studies, monoE images at low keV levels and iodine overlay maps facilitated subjective lesion delineation which was substantiated by the quantitative analysis. Hence, spectral detector CT improves pancreatic lesion conspicuity, while its value for lesion differentiation needs to be further evaluated in larger study cohorts.

Precision and reliability of liver iodine quantification from spectral detector CT: evidence from phantom and patient data.

OBJECTIVE: To comprehensively assess precision, reproducibility, and repeatability of iodine maps from spectral detector CT (SDCT) in a phantom and in patients with repetitive examination of the abdomen. METHODS: Seventy-seven patients who underwent examination two (n = 52) or three (n = 25) times according to clinical indications were included in this IRB-approved, retrospective study. The anthropomorphic liver phantom and all patients were scanned with a standardized protocol (SSDE in patients 15.8 mGy). In patients, i.v. contrast was administered and portal venous images were acquired using bolus-tracking technique. The phantom was scanned three times at three time points; in one acquisition, image reconstruction was repeated three times. Region of interest (ROI) were placed automatically (phantom) or manually (patients) in the liver parenchyma (mimic) and the portal vein; attenuation in conventional images (CI [HU]) and iodine map concentrations (IM [mg/ml]) were recorded. The coefficient of variation (CV [%]) was used to compare between repetitive acquisitions. If present, additional ROI were placed in cysts (n = 29) and hemangioma (n = 29). RESULTS: Differences throughout all phantom examinations were < 2%. In patients, differences between two examinations were higher (CV for CI/IM: portal vein, 2.5%/3.2%; liver parenchyma, -0.5%/-3.0% for CI/IM). In 80% of patients, these differences were within a ± 20% limit. Differences in benign liver lesions were even higher (68% and 38%, for CI and IM, respectively). CONCLUSIONS: Iodine maps from SDCT allow for reliable quantification of iodine content in phantoms; while in patients, rather large differences between repetitive examinations are likely due to differences in biological distribution. This underlines the need for careful clinical interpretation and further protocol optimization. KEY POINTS: • Spectral detector computed tomography allows for reliable quantification of iodine in phantoms. • In patients, the offset between repetitive examinations varies by 20%, likely due to differences in biological distribution. • Clinically, iodine maps should be interpreted with caution and should take the intra-individual variability of iodine distribution over time into account.

Radioablation by Image-Guided (HDR) Brachytherapy and Transarterial Chemoembolization in Hepatocellular Carcinoma: A Randomized Phase II Trial.

BACKGROUND AND AIMS: The aim of this single-center, open-label phase II study was to assess the efficacy of image-guided high-dose-rate (HDR) brachytherapy (iBT) compared with conventional transarterial embolization (cTACE) in unresectable hepatocellular carcinoma. METHODS: Seventy-seven patients were treated after randomization to iBT or cTACE, as single or repeated interventions. Crossover was allowed if clinically indicated. The primary endpoint was time to untreatable progression (TTUP). Eligibility criteria included a Child-Pugh score of ≤ 8 points, absence of portal vein thrombosis (PVT) at the affected liver lobe, and ≤ 4 lesions. Survival was analyzed by using the Cox proportional hazard model with stratification for Barcelona Clinic Liver Cancer (BCLC) stages. RESULTS: Twenty patients were classified as BCLC-A (iBT/cTACE 8/12), 35 as BCLC-B (16/19), and 22 as BCLC-C (13/9). The 1-, 2-, and 3-year TTUP probabilities for iBT compared with cTACE were 67.5% versus 55.2%, 56.0% versus 27.4%, and 29.5% versus 11.0%, respectively, with an adjusted hazard ratio (HR) of 0.49 (95% confidence interval 0.27-0.89; p = 0.019). The 1-, 2-, and 3-year TTPs for iBT versus cTACE were 56.0% versus 28.2%, 23.9% versus 6.3%, and 15.9% versus 6.3%, respectively, with an adjusted HR of 0.49 (0.29-0.85; p = 0.011). The 1-, 2-, and 3-year OS rates were 78.4% versus 67.7%, 62.0% versus 47.3%, and 36.7% versus 27.0%, respectively, with an adjusted HR of 0.62 (0.33-1.16; p = 0.136). CONCLUSIONS: This explorative phase II trial showed a superior outcome of iBT compared with cTACE in hepatocellular carcinoma and supports proceeding to a phase III trial.

Thoracic-abdominal imaging with a novel dual-layer spectral detector CT: intra-individual comparison of image quality and radiation dose with 128-row single-energy acquisition.

BACKGROUND: A novel, multi-energy, dual-layer spectral detector computed tomography (SDCT) is commercially available now with the vendor's claim that it yields the same or better quality of polychromatic, conventional CT images like modern single-energy CT scanners without any radiation dose penalty. PURPOSE: To intra-individually compare the quality of conventional polychromatic CT images acquired with a dual-layer spectral detector (SDCT) and the latest generation 128-row single-energy-detector (CT128) from the same manufacturer. MATERIAL AND METHODS: Fifty patients underwent portal-venous phase, thoracic-abdominal CT scans with the SDCT and prior CT128 imaging. The SDCT scanning protocol was adapted to yield a similar estimated dose length product (DLP) as the CT128. Patient dose optimization by automatic tube current modulation and CT image reconstruction with a state-of-the-art iterative algorithm were identical on both scanners. CT image contrast-to-noise ratio (CNR) was compared between the SDCT and CT128 in different anatomic structures. Image quality and noise were assessed independently by two readers with 5-point-Likert-scales. Volume CT dose index (CTDIvol), and DLP were recorded and normalized to 68 cm acquisition length (DLP68). RESULTS: The SDCT yielded higher mean CNR values of 30.0% ± 2.0% (26.4-32.5%) in all anatomic structures ( P < 0.001) and excellent scores for qualitative parameters surpassing the CT128 (all P < 0.0001) with substantial inter-rater agreement (κ ≥ 0.801). Despite adapted scan protocols the SDCT yielded lower values for CTDIvol (-10.1 ± 12.8%), DLP (-13.1 ± 13.9%), and DLP68 (-15.3 ± 16.9%) than the CT128 (all P < 0.0001). CONCLUSION: The SDCT scanner yielded better CT image quality compared to the CT128 and lower radiation dose parameters.

Case Report: A Cryptosporidium infection in a patient with relapsed T-lymphoblastic lymphoma undergoing allogeneic stem cell transplantation.

Cryptosporidium infection is a rare cause of enterocolitis. In immunocompromised patients, cryptosporidiosis may lead to debilitating and life-threatening diarrhea and malabsorption, occasionally with multi-organ involvement. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) requires long-term immunosuppressive therapy, while cellular immunity is usually compromised due to intensive conditioning chemotherapy. Diarrhea in patients who underwent allo-HSCT may be a sign of an infection, but can also be the result of intestinal graft-versus-host disease (GvHD). Here, we describe the case of a patient who developed severe diarrhea following allo-HSCT for relapsed T-lymphoblastic lymphoma. Initially, GvHD was suspected and treatment was initiated accordingly. However, a colon biopsy showed signs of cryptosporide oocysts alongside only low-grade GvHD. Following molecular confirmation of the diagnosis of cryptosporidiosis, an intensive treatment regimen was started. Despite the severe clinical course, the patient recovered and was discharged with only residual symptoms.

Intra-individual comparison between abdominal virtual mono-energetic spectral and conventional images using a novel spectral detector CT.

OBJECTIVES: To quantitatively and qualitatively assess abdominal arterial and venous phase contrast-enhanced spectral detector computed tomography (SDCT) virtual mono-energetic (MonoE) datasets in comparison to conventional CT reconstructions provided by the same system. MATERIALS AND METHODS: Conventional and MonoE images at 40-120 kilo-electron volt (keV) levels with a 10 keV increment as well as 160 and 200 keV were reconstructed in abdominal SDCT datasets of 55 patients. Attenuation, image noise, and contrast- / signal-to-noise ratios (CNR, SNR) of vessels and solid organs were compared between MonoE and conventional reconstructions. Two readers assessed contrast conditions, detail visualization, overall image quality and subjective image noise with both, fixed and adjustable window settings. RESULTS: Attenuation, CNR and SNR of vessels and solid organs showed a stepwise increase from high to low keV reconstructions in both contrast phases while image noise stayed stable at low keV MonoE reconstruction levels. Highest levels were found at 40 keV MonoE reconstruction (p<0.001), respectively. Solid abdominal organs showed a stepwise decrease from low to high energy levels in regard to attenuation, CNR and SNR with significantly higher values at 40 and 50 keV, compared to conventional images. The 70 keV MonoE was comparable to conventional poly-energetic reconstruction (p≥0.99). Subjective analysis displayed best image quality for the 70 keV MonoE reconstruction level in both phases at fixed standard window presets and at 40 keV if window settings could be adjusted. CONCLUSION: SDCT derived low keV MonoE showed markedly increased CNR and SNR values due to constantly low image noise values over the whole energy spectrum from 40 to 200 keV.

Joint deformable liver registration and bias field correction for MR-guided HDR brachytherapy.

PURPOSE: In interstitial high-dose rate brachytherapy, liver cancer is treated by internal radiation, requiring percutaneous placement of applicators within or close to the tumor. To maximize utility, the optimal applicator configuration is pre-planned on magnetic resonance images. The pre-planned configuration is then implemented via a magnetic resonance-guided intervention. Mapping the pre-planning information onto interventional data would reduce the radiologist's cognitive load during the intervention and could possibly minimize discrepancies between optimally pre-planned and actually placed applicators. METHODS: We propose a fast and robust two-step registration framework suitable for interventional settings: first, we utilize a multi-resolution rigid registration to correct for differences in patient positioning (rotation and translation). Second, we employ a novel iterative approach alternating between bias field correction and Markov random field deformable registration in a multi-resolution framework to compensate for non-rigid movements of the liver, the tumors and the organs at risk. In contrast to existing pre-correction methods, our multi-resolution scheme can recover bias field artifacts of different extents at marginal computational costs. RESULTS: We compared our approach to deformable registration via B-splines, demons and the SyN method on 22 registration tasks from eleven patients. Results showed that our approach is more accurate than the contenders for liver as well as for tumor tissues. We yield average liver volume overlaps of 94.0 ± 2.7% and average surface-to-surface distances of 2.02 ± 0.87 mm and 3.55 ± 2.19 mm for liver and tumor tissue, respectively. The reported distances are close to (or even below) the slice spacing (2.5 - 3.0 mm) of our data. Our approach is also the fastest, taking 35.8 ± 12.8 s per task. CONCLUSION: The presented approach is sufficiently accurate to map information available from brachytherapy pre-planning onto interventional data. It is also reasonably fast, providing a starting point for computer-aidance during intervention.

Image quality evaluation of dual-layer spectral detector CT of the chest and comparison with conventional CT imaging.

OBJECTIVES: To evaluate image quality parameters of virtual mono-energetic (MonoE) and conventional (CR) imaging derived from a dual-layer spectral detector CT (DLCT) in oncological follow-up venous phase imaging of the chest and comparison with conventional multi-detector CT (CRMDCT) imaging. MATERIALS AND METHODS: A total of 55 patients who had oncologic staging with conventional CT and DLCT of the chest in venous phase were included in this study. Established image quality parameters were derived from all datasets in defined thoracic landmarks. Attenuation, image noise, and signal-/contrast- to noise ratios (SNR, CNR) were compared between CRDLCT and MonoE as well as CRMDCT imaging. Two readers performed subjective image analysis. RESULTS: CRMDCT showed significant lower attenuation values compared to CRDLCT and MonoE at 40-70keV (p≤0.05). Moreover, MonoE at 40-70keV revealed significantly higher attenuations values compared to CRDLCT (p<0.001). Noise was statistically lower in CRMDCT compared with CRDLCT and MonoE at 40keV (11.4±2.3 HU vs. 12.0±3.1 HU vs. 11.7±5.2 HU; p<0.001). In contrast, all MonoE levels showed significantly lower noise levels compared to CRDLCT (p<0.001). SNR was not significantly different between CRMDCT and CRDLCT (13.5±3.7 vs. 14.4±5.3; p>0.99). SNR values were significantly increased for MonoE at 40-80keV compared to CRMDCT and CRDLCT (p<0.001). CRDLCT and MonoE (40-70keV) from DLCT revealed significantly higher CNR values than CRMDCT (p<0.001). In subjective analysis, MonoE at 40keV surpassed all other image reconstructions except for noise in MonoE at 70 keV. CONCLUSION: In dual-layer spectral detector CT, MonoE at low keV showed superior image quality compared to conventional images derived from the same system and may therefore be added to clinical routine imaging protocols. Whether MonoE reconstructions yield additional diagnostic information is still unknown.

[Angioneurotic Edema as a Differential Diagnosis of Recurrent Abdominal Pain]. / Angioneurotisches Ödem als Differenzialdiagnose rezidivierender Bauchschmerzen.

History and admission findings A 46-year-old woman was admitted because of recurrent postprandial vomiting including weight loss of 30 kg within the last 5 - 6 months. The physical investigation was indicative of preileus. Investigations CT scan revealed an unclear thickening of the jejunum. Treatment and course Explorative laparotomy was performed because of the considerable degree of suffering and a tendency of worsening. Intraoperatively, the preoperatively observed thickening was confirmed at 3 subsequent segments of the jejunum (3 cm long each) prompting to segmental resection of the jejunum. The postoperative course was uneventful. Histopathology revealed submucous, partially transmural, segmental edema compatible with an angioneurotic edema. Amyloidosis, sprue and malignancy could be excluded. A C1 esterase inhibitor test war normal. A further search for possible causes revealed ACE inhibitor medication for hypertension. Conclusion The differential diagnostic assessment of isolated small intestine diseases must take into account rare systemic disease. The knowledge of the history of medicine can be useful to detect the - not mandatory - connection to angioedema.

Hybrid Approach for Biliary Interventions Employing MRI-Guided Bile Duct Puncture with Near-Real-Time Imaging.

OBJECTIVE: To assess the feasibility of a hybrid approach employing MRI-guided bile duct (BD) puncture for subsequent fluoroscopy-guided biliary interventions in patients with non-dilated (≤3 mm) or dilated BD (≥3 mm) but unfavorable conditions for ultrasonography (US)-guided BD puncture. METHODS: A total of 23 hybrid interventions were performed in 21 patients. Visualization of BD and puncture needles (PN) in the interventional MR images was rated on a 5-point Likert scale by two radiologists. Technical success, planning time, BD puncture time and positioning adjustments of the PN as well as technical success of the biliary intervention and complication rate were recorded. RESULTS: Visualization even of third-order non-dilated BD and PN was rated excellent by both radiologists with good to excellent interrater agreement. MRI-guided BD puncture was successful in all cases. Planning and BD puncture times were 1:36 ± 2.13 (0:16-11:07) min. and 3:58 ± 2:35 (1:11-9:32) min. Positioning adjustments of the PN was necessary in two patients. Repeated capsular puncture was not necessary in any case. All biliary interventions were completed successfully without major complications. CONCLUSION: A hybrid approach which employs MRI-guided BD puncture for subsequent fluoroscopy-guided biliary intervention is feasible in clinical routine and yields high technical success in patients with non-dilated BD and/or unfavorable conditions for US-guided puncture. Excellent visualization of BD and PN in near-real-time interventional MRI allows successful cannulation of the BD.

Improvement of image quality and dose management in CT fluoroscopy by iterative 3D image reconstruction.

OBJECTIVES: The objective of this study was to assess the influence of an iterative CT reconstruction algorithm (IA), newly available for CT-fluoroscopy (CTF), on image noise, readers' confidence and effective dose compared to filtered back projection (FBP). METHODS: Data from 165 patients (FBP/IA = 82/74) with CTF in the thorax, abdomen and pelvis were included. Noise was analysed in a large-diameter vessel. The impact of reconstruction and variables (e.g. X-ray tube current I) influencing noise and effective dose were analysed by ANOVA and a pairwise t-test with Bonferroni-Holm correction. Noise and readers' confidence were evaluated by three readers. RESULTS: Noise was significantly influenced by reconstruction, I, body region and circumference (all p ≤ 0.0002). IA reduced the noise significantly compared to FBP (p = 0.02). The effect varied for body regions and circumferences (p ≤ 0.001). The effective dose was influenced by the reconstruction, body region, interventional procedure and I (all p ≤ 0.02). The inter-rater reliability for noise and readers' confidence was good (W ≥ 0.75, p < 0.0001). Noise and readers' confidence were significantly better in AIDR-3D compared to FBP (p ≤ 0.03). Generally, IA yielded a significant reduction of the median effective dose. CONCLUSION: The CTF reconstruction by IA showed a significant reduction in noise and effective dose while readers' confidence increased. KEY POINTS: • CTF is performed for image guidance in interventional radiology. • Patient exposure was estimated from DLP documented by the CT. • Iterative CT reconstruction is appropriate to reduce image noise in CTF. • Using iterative CT reconstruction, the effective dose was significantly reduced in abdominal interventions.

Cardiac Magnetic Resonance Imaging Using an Open 1.0T MR Platform: A Comparative Study with a 1.5T Tunnel System.

BACKGROUND: Cardiac magnetic resonance imaging (cMRI) has become the non-invasive reference standard for the evaluation of cardiac function and viability. The introduction of open, high-field, 1.0T (HFO) MR scanners offers advantages for examinations of obese, claustrophobic and paediatric patients.The aim of our study was to compare standard cMRI sequences from an HFO scanner and those from a cylindrical, 1.5T MR system. MATERIAL/METHOD: Fifteen volunteers underwent cMRI both in an open HFO and in a cylindrical MR system. The protocol consisted of cine and unenhanced tissue sequences. The signal-to-noise ratio (SNR) for each sequence and blood-myocardium contrast for the cine sequences were assessed. Image quality and artefacts were rated. The location and number of non-diagnostic segments was determined. Volunteers' tolerance to examinations in both scanners was investigated. RESULTS: SNR was significantly lower in the HFO scanner (all p<0.001). However, the contrast of the cine sequence was significantly higher in the HFO platform compared to the 1.5T MR scanner (0.685±0.41 vs. 0.611±0.54; p<0.001). Image quality was comparable for all sequences (all p>0.05). Overall, only few non-diagnostic myocardial segments were recorded: 6/960 (0.6%) by the HFO and 17/960 (1.8%) segments by the cylindrical system. The volunteers expressed a preference for the open MR system (p<0.01). CONCLUSIONS: Standard cardiac MRI sequences in an HFO platform offer a high image quality that is comparable to the quality of images acquired in a cylindrical 1.5T MR scanner. An open scanner design may potentially improve tolerance of cardiac MRI and therefore allow to examine an even broader patient spectrum.

Validation of VX2 as a Hepatocellular Carcinoma Model: Comparison of the Molecular Reaction of VX2 and HepG2 Tumor Cells to Sorafenib In Vitro.

As there is currently no superior hepatocellular carcinoma (HCC) model with percutaneous vascular access for transarterial treatments available, the VX2 rabbit model is frequently used for in vivo investigations on liver carcinoma. However, the VX2 cell line was derived from a virus-induced skin papilloma that can form carcinosarcoma in liver of rabbits and the transferability of obtained results to HCC treatment remains open. Here we compared the most frequently investigated human HCC model cell line, HepG2, with VX2 cells in vitro in terms of sensitivity towards the broad specificity kinase inhibitor sorafenib and responsiveness to the addition of platelet-derived growth factor AB (PDGF-AB), vascular endothelial growth factor (VEGF) and hepatic growth factor (HGF), as well as insulin and interleukin-1ß (IL1ß). Phosphorylation of protein kinase B (AKT) the mitogen-activated protein kinases (MAPKs) p38 and p42/44 (extracellular signal-regulated kinase, ERK1/2) and inhibitor of kappa light chain gene enhancer alpha (IĸBα) was determined by western blotting as these events are associated with early signaling cascades. Additionally, the inhibition of phosphorylation under sorafenib treatment was investigated. Sorafenib was equally toxic to both cell lines, but only in HepG2 was activation of caspase 3/7 activity, as a sign of apoptosis, observed. VX2 cells exhibited generally more intense phosphorylation signals in response to the growth factors and also serum. In contrast to VX2, HepG2 cells showed no response to PDGF-AB or VEGF as determined by kinase phosphorylation. In both cell lines, sorafenib inhibited growth factor-induced phosphorylation of ERK and p38-MAPK. AKT phosphorylation was only inhibited in VX2 cells and IĸBα phosphorylation was not influenced by this kinase inhibitor in either cell type. Taken together, the two cellular models for HCC share several features related to sorafenib application, but differed in their responsiveness towards growth factors. Therefore, results obtained with the VX2 model cannot be extended to human HCC without appropriate caution.