BOA: A CT-Based Body and Organ Analysis for Radiologists at the Point of Care

PURPOSE

The study aimed to develop the open-source body and organ analysis (BOA), a comprehensive computed tomography (CT) image segmentation algorithm with a focus on workflow integration.

METHODS

The BOA combines 2 segmentation algorithms: body composition analysis (BCA) and TotalSegmentator. The BCA was trained with the nnU-Net framework using a dataset including 300 CT examinations. The CTs were manually annotated with 11 semantic body regions: subcutaneous tissue, muscle, bone, abdominal cavity, thoracic cavity, glands, mediastinum, pericardium, breast implant, brain, and spinal cord. The models were trained using 5-fold cross-validation, and at inference time, an ensemble was used. Afterward, the segmentation efficiency was evaluated on a separate test set comprising 60 CT scans. In a postprocessing step, a tissue segmentation (muscle, subcutaneous adipose tissue, visceral adipose tissue, intermuscular adipose tissue, epicardial adipose tissue, and paracardial adipose tissue) is created by subclassifying the body regions. The BOA combines this algorithm and the open-source segmentation software TotalSegmentator to have an all-in-one comprehensive selection of segmentations. In addition, it integrates into clinical workflows as a DICOM node-triggered service using the open-source Orthanc research PACS (Picture Archiving and Communication System) server to make the automated segmentation algorithms available to clinicians. The BCA model's performance was evaluated using the Sørensen-Dice score. Finally, the segmentations from the 3 different tools (BCA, TotalSegmentator, and BOA) were compared by assessing the overall percentage of the segmented human body on a separate cohort of 150 whole-body CT scans.

RESULTS

The results showed that the BCA outperformed the previous publication, achieving a higher Sørensen-Dice score for the previously existing classes, including subcutaneous tissue (0.971 vs 0.962), muscle (0.959 vs 0.933), abdominal cavity (0.983 vs 0.973), thoracic cavity (0.982 vs 0.965), bone (0.961 vs 0.942), and an overall good segmentation efficiency for newly introduced classes: brain (0.985), breast implant (0.943), glands (0.766), mediastinum (0.880), pericardium (0.964), and spinal cord (0.896). All in all, it achieved a 0.935 average Sørensen-Dice score, which is comparable to the one of the TotalSegmentator (0.94). The TotalSegmentator had a mean voxel body coverage of 31% ± 6%, whereas BCA had a coverage of 75% ± 6% and BOA achieved 93% ± 2%.

CONCLUSIONS

The open-source BOA merges different segmentation algorithms with a focus on workflow integration through DICOM node integration, offering a comprehensive body segmentation in CT images with a high coverage of the body volume.

Impact of Vaccination Status on Outcome of Patients With COVID-19 and Acute Ischemic Stroke Undergoing Mechanical Thrombectomy

BACKGROUND

Data on impact of COVID-19 vaccination and outcomes of patients with COVID-19 and acute ischemic stroke undergoing mechanical thrombectomy are scarce. Addressing this subject, we report our multicenter experience.

METHODS AND RESULTS

This was a retrospective analysis of patients with COVID-19 and known vaccination status treated with mechanical thrombectomy for acute ischemic stroke at 20 tertiary care centers between January 2020 and January 2023. Baseline demographics, angiographic outcome, and clinical outcome evaluated by the modified Rankin Scale score at discharge were noted. A multivariate analysis was conducted to test whether these variables were associated with an unfavorable outcome, defined as modified Rankin Scale score >3. A total of 137 patients with acute ischemic stroke (48 vaccinated and 89 unvaccinated) with acute or subsided COVID-19 infection who underwent mechanical thrombectomy attributable to vessel occlusion were included in the study. Angiographic outcomes between vaccinated and unvaccinated patients were similar (modified Thrombolysis in Cerebral Infarction ≥2b: 85.4% in vaccinated patients versus 86.5% in unvaccinated patients; P=0.859). The rate of functional independence (modified Rankin Scale score, ≤2) was 23.3% in the vaccinated group and 20.9% in the unvaccinated group (P=0.763). The mortality rate was 30% in both groups. In the multivariable analysis, vaccination status was not a significant predictor for an unfavorable outcome (P=0.957). However, acute COVID-19 infection remained significant (odds ratio, 1.197 [95% CI, 1.007-1.417]; P=0.041).

CONCLUSIONS

Our study demonstrated no impact of COVID-19 vaccination on angiographic or clinical outcome of COVID-19-positive patients with acute ischemic stroke undergoing mechanical thrombectomy, whereas worsening attributable to COVID-19 was confirmed.

Elexacaftor/tezacaftor/ivacaftor influences body composition in adults with cystic fibrosis: a fully automated CT-based analysis

A poor nutritional status is associated with worse pulmonary function and survival in people with cystic fibrosis (pwCF). CF transmembrane conductance regulator modulators can improve pulmonary function and body weight, but more data is needed to evaluate its effects on body composition. In this retrospective study, a pre-trained deep-learning network was used to perform a fully automated body composition analysis on chest CTs from 66 adult pwCF before and after receiving elexacaftor/tezacaftor/ivacaftor (ETI) therapy. Muscle and adipose tissues were quantified and divided by bone volume to obtain body size-adjusted ratios. After receiving ETI therapy, marked increases were observed in all adipose tissue ratios among pwCF, including the total adipose tissue ratio (+ 46.21%, p < 0.001). In contrast, only small, but statistically significant increases of the muscle ratio were measured in the overall study population (+ 1.63%, p = 0.008). Study participants who were initially categorized as underweight experienced more pronounced effects on total adipose tissue ratio (p = 0.002), while gains in muscle ratio were equally distributed across BMI categories (p = 0.832). Our findings suggest that ETI therapy primarily affects adipose tissues, not muscle tissue, in adults with CF. These effects are primarily observed among pwCF who were initially underweight. Our findings may have implications for the future nutritional management of pwCF.

FRED X flow diverter for the treatment of intracranial aneurysms: Two-center experience and mini-review of the literature

OBJECTIVE

The flow re-direction endoluminal device (FRED) is a safe and effective treatment option for intracranial aneurysms. The novel FRED X features an antithrombotic surface coating ("X Technology") on an otherwise unmodified stent design. This two-center study evaluates the clinical safety and efficacy of FRED X and compares it to the literature.

METHODS

Consecutive patients treated between 2020 and 2023 were retrospectively reviewed for aneurysm characteristics, procedural details and complications, and angiographic outcomes. A mini-review of the literature for FRED X clinical trials was performed and results were pooled using a random effects model.

RESULTS

Thirty-four patients (mean age 56 years) were treated for 34 aneurysms. The mean aneurysm size was 7.7 ± 5.0 mm, 7 (21%) were ruptured, 6 (18%) were recurrent after previous treatment, 11 (32.3%) were located in the posterior circulation, and 4 (12.5%) had non-saccular morphology. All procedures were technically successful and no balloon angioplasty was required. There was 1 (2.9%) symptomatic complication (a transient ischemic attack) and no procedural morbidity or mortality. Technical asymptomatic events included 1 procedural stent occlusion that was reopened with thrombectomy and 3 cases of vasospasm. Complete and adequate occlusion rates were 68% (19/28) and 89% (25/28) at a mean follow-up time of 6 months, respectively. The results of this study are comparable to previous FRED X studies.

CONCLUSIONS

The results demonstrate a high feasibility and procedural safety of the FRED X with adequate mid-term occlusion rates. Long-term and comparative studies are needed to evaluate the full potential of the FRED X.

Is Stent Retraction to ReLieve Arterial Cerebral VaSospasm Caused by SAH (Stent-ReLACSS) Using PRELAX the Long-awaited Solution for Treatment of Posthemorrhagic Cerebral Vasospasm? : Treatment of Posthemorrhagic Cerebral Vasospasm with PRESET and PRELAX: Technical Aspects, Efficacy, and Safety Margins in a Case Series

PURPOSE

Recent observational studies have indicated the efficacy of stent retriever devices for the treatment of posthemorrhagic cerebral vasospasm (CVS), both by deployment and on-site withdrawal into the microcatheter (stent angioplasty, SA) and deployment followed by retraction through the target vessel similar to thrombectomy (Stent Retraction to reLieve Arterial Cerebral vaSospasm caused by SAH, Stent-ReLACSS). This article reports the findings with each application of pRESET and pRELAX in the treatment of CVS.

METHODS

We retrospectively enrolled 25 patients with severe CVS following aneurysmal subarachnoid hemorrhage. For the SA group, a stent retriever or a pRELAX was temporarily deployed into a narrow vessel segment and retrieved into the microcatheter after 3 min. For the Stent-ReLACSS group, a pRELAX was temporarily deployed into a narrow vessel and pulled back unfolded into the internal carotid artery. If intra-arterial vasodilators were administered, they were given exclusively after mechanical vasospasmolysis to maximize the effectiveness of the stent treatment.

RESULTS

In this study fifteen patients and 49 vessels were treated with SA. All were technically successful without periprocedural complications; however, 8/15 patients (53.3%) required additional treatment of the CVS. A total of 10 patients and 23 vessel segments were treated with Stent-ReLACSS. All maneuvers were technically successful without periprocedural complications and all vessels showed significant angiographic improvement. No recurrent CVS requiring further endovascular treatment occurred in-hospital, and neither territorial ischemia in the treated vessels nor vascular injury were observed in follow-up angiography.

CONCLUSION

Based on the presented data it appears that Stent-ReLACSS with pRELAX does not pose any additional risks when used to treat CVS and might be superior to SA, especially concerning mid-term and long-term efficacy. The mechanism of action may be an effect on the endothelium rather than mechanical vasodilation. As many patients with CVS are diagnosed too late, prophylactic treatment of high-risk patients (e.g., poor grade, young, female) is potentially viable.

Body composition impacts outcome of bronchoscopic lung volume reduction in patients with severe emphysema: a fully automated CT-based analysis

Chronic Obstructive Pulmonary Disease (COPD) is characterized by progressive and irreversible airflow limitation, with individual body composition influencing disease severity. Severe emphysema worsens symptoms through hyperinflation, which can be relieved by bronchoscopic lung volume reduction (BLVR). To investigate how body composition, assessed through CT scans, impacts outcomes in emphysema patients undergoing BLVR. Fully automated CT-based body composition analysis (BCA) was performed in patients with end-stage emphysema receiving BLVR with valves. Post-interventional muscle and adipose tissues were quantified, body size-adjusted, and compared to baseline parameters. Between January 2015 and December 2022, 300 patients with severe emphysema underwent endobronchial valve treatment. Significant improvements were seen in outcome parameters, which were defined as changes in pulmonary function, physical performance, and quality of life (QoL) post-treatment. Muscle volume remained stable (1.632 vs. 1.635 for muscle bone adjusted ratio (BAR) at baseline and after 6 months respectively), while bone adjusted adipose tissue volumes, especially total and pericardial adipose tissue, showed significant increase (2.86 vs. 3.00 and 0.16 vs. 0.17, respectively). Moderate to strong correlations between bone adjusted muscle volume and weaker correlations between adipose tissue volumes and outcome parameters (pulmonary function, QoL and physical performance) were observed. Particularly after 6-month, bone adjusted muscle volume changes positively corresponded to improved outcomes (ΔForced expiratory volume in 1 s [FEV1], r = 0.440; ΔInspiratory vital capacity [IVC], r = 0.397; Δ6Minute walking distance [6MWD], r = 0.509 and ΔCOPD assessment test [CAT], r = -0.324; all p < 0.001). Group stratification by bone adjusted muscle volume changes revealed that groups with substantial muscle gain experienced a greater clinical benefit in pulmonary function improvements, QoL and physical performance (ΔFEV1%, 5.5 vs. 39.5; ΔIVC%, 4.3 vs. 28.4; Δ6MWDm, 14 vs. 110; ΔCATpts, -2 vs. -3.5 for groups with ΔMuscle, BAR% < -10 vs. > 10, respectively). BCA results among patients divided by the minimal clinically important difference for forced expiratory volume of the first second (FEV1) showed significant differences in bone-adjusted muscle and intramuscular adipose tissue (IMAT) volumes and their respective changes after 6 months (ΔMuscle, BAR% -5 vs. 3.4 and ΔIMAT, BAR% -0.62 vs. 0.60 for groups with ΔFEV1 ≤ 100 mL vs > 100 mL). Altered body composition, especially increased muscle volume, is associated with functional improvements in BLVR-treated patients.

Triple platelet inhibition in intracranial thrombectomy with additional acute cervical stent angioplasty due to tandem lesion: a retrospective single-center analysis

BACKGROUND

Acute stroke treatment with intracranial thrombectomy and treatment of ipsilateral carotid artery stenosis/occlusion ("tandem lesion", TL) in one session is considered safe. However, the risk of stent restenosis after TL treatment is high, and antiplatelet therapy (APT) preventing restenosis must be well balanced to avoid intracranial hemorrhage. We investigated the safety and 90-day outcome of patients receiving TL treatment under triple-APT, focused on stent-patency and possible disadvantageous comorbidities.

METHODS

Patients receiving TL treatment in the setting of acute stroke between 2013 and 2022 were analyzed regarding peri-/postprocedural safety and stent patency after 90 days. All patients received intravenous eptifibatide and acetylsalicylic acid and one of the three drugs prasugrel, clopidogrel, or ticagrelor. Duplex imaging was performed 24 h after treatment, at discharge and 90 days, and digital subtraction angiography was performed if restenosis was suspected.

RESULTS

176 patients were included. Periprocedural complications occurred in 2.3% of the patients at no periprocedural death, and in-hospital death in 13.6%. Discharge mRS score was maintained or improved at the 90-day follow-up in 86%, 4.54% had an in-stent restenosis requiring treatment at 90 days. No recorded comorbidity considered disadvantageous for stent patency showed statistical significance, the duration of the endovascular procedure had no significant effect on outcome.

CONCLUSION

In our data, TL treatment with triple APT resulted in a low restenosis rate, low rates of sICH and a comparably high number of patients with favorable outcome. Aggressive APT in the initial phase may therefore have the potential to prevent recurrent stroke better than restrained platelet inhibition. Comorbidities did not influence stent patency.

Endovascular treatment of distal anterior cerebral artery aneurysms using flow modulation devices: mid- and long-term results from a two-center study

Purpose

Flow-diverter (FD) stents have become an established treatment for intracranial aneurysms in recent years, but their use for aneurysms in distal cerebral vessels with small carrier vessel diameters remains controversial. This study describes the method and mid- and long-term outcomes of FD treatment of distal anterior cerebral artery aneurysms (DACAAs) at two neurointerventional centers, to elucidate this topic and provide more in-depth data.

Methods

Data for all patients at two neurointerventional centers who were treated with FDs for DACAAs in the pericallosal and supracallosal segment of the anterior cerebral artery were retrospectively analyzed. Data on periprocedural complications, and short-, mid- and long-term follow-up findings were recorded.

Results

Forty-one patients were eligible for inclusion in the study. Three FD models were used, one of which had an anti-thrombotic coating. Two periprocedural complications (5%) occurred but did not cause a change in the mRS. In the long-term follow-up, at 29 months and beyond, 83% of assessable patients showed complete occlusion of the aneurysms without new neurological deficits.

Conclusion

FDs are a safe and effective treatment approach for DACAAs. This study indicated a low risk of complications, and high closure rates in short-, mid- and long-term follow-up.

MR Imaging Characteristics of Solitary Fibrous Tumors of the Orbit : Case Series of 18 Patients

PURPOSE

Solitary fibrous tumor (SFT) of the orbit is a rare tumor that was first described in 1994. We aimed to investigate its imaging characteristics that may facilitate the differential diagnosis between SFT and other types of orbital tumors.

MATERIAL AND METHODS

Magnetic resonance imaging (MRI) data of patients with immunohistochemically confirmed orbital SFT from 2002 to 2022 at a tertiary care center were retrospectively analyzed. Tumor location, size, morphological characteristics, and contrast enhancement features were evaluated.

RESULTS

Of the 18 eligible patients 10 were female (56%) with a mean age of 52 years. Most of the SFTs were oval-shaped (67%) with a sharp margin (83%). The most frequent locations were the laterocranial quadrant (44%), the extraconal space (67%) and the dorsal half of the orbit (67%). A flow void phenomenon was observed in nearly all cases (94%). On the T1-weighted imaging, tumor signal intensity (SI) was significantly lower than that of the retrobulbar fat and appeared predominantly equivalent (82%) to the temporomesial brain cortex, while on T2-weighted imaging its SI remained equivalent (50%) or slightly hyperintense to that of brain cortex. More than half of the lesions showed a homogeneous contrast enhancement pattern with a median SI increase of 2.2-fold compared to baseline precontrast imaging.

CONCLUSION

The SFT represents a rare orbital tumor with several characteristic imaging features. It was mostly oval-shaped with a sharp margin and frequently localized in the extraconal space and dorsal half of the orbit. Flow voids indicating hypervascularization were the most common findings.

Impact of CT and MRI in the diagnostic workup of malignant triton tumour-a monocentric analysis and review of the literature

OBJECTIVES

Malignant triton tumours (MTTs) are rare but aggressive subtypes of malignant peripheral nerve sheath tumours (MPNSTs) with a high recurrence rate and 5-year survival of 14%. Systematic imaging data on MTTs are scarce and mainly based on single case reports. Therefore, we aimed to identify typical CT and MRI features to improve early diagnosis rates of this uncommon entity.

METHODS

A systematic review on literature published until December 2022 on imaging characteristics of MTTs was performed. Based on that, we conducted a retrospective, monocentric analysis of patients with histopathologically proven MTTs from our department. Explorative data analysis was performed.

RESULTS

Initially, 29 studies on 34 patients (31.42 ± 22.6 years, 12 female) were evaluated: Literature described primary MTTs as huge, lobulated tumours (108 ± 99.3 mm) with central necrosis (56% [19/34]), low T1w (81% [17/21]), high T2w signal (90% [19/21]) and inhomogeneous enhancement on MRI (54% [7/13]). Analysis of 16 patients (48.9 ± 13.8 years; 9 female) from our institution revealed comparable results: primary MTTs showed large, lobulated masses (118 mm ± 64.9) with necrotic areas (92% [11/12]). MRI revealed low T1w (100% [7/7]), high T2w signal (100% [7/7]) and inhomogeneous enhancement (86% [6/7]). Local recurrences and soft-tissue metastases mimicked these features, while nonsoft-tissue metastases appeared unspecific.

CONCLUSIONS

MTTs show characteristic features on CT and MRI. However, these do not allow a reliable differentiation between MTTs and other MPNSTs based on imaging alone. Therefore, additional histopathological analysis is required.

ADVANCES IN KNOWLEDGE

This largest published systematic analysis on MTT imaging revealed typical but unspecific imaging features that do not allow a reliable, imaging-based differentiation between MTTs and other MPNSTs. Hence, additional histopathological analysis remains essential.

Radiomics Analysis of Multiparametric PET/MRI for N- and M-Staging in Patients with Primary Cervical Cancer

PURPOSE

 The aim of this study was to investigate the potential of multiparametric 18F-FDG PET/MR imaging as a platform for radiomics analysis and machine learning algorithms based on primary cervical cancers to predict N- and M-stage in patients.

MATERIALS AND METHODS

 A total of 30 patients with histopathological confirmation of primary and untreated cervical cancer were prospectively enrolled for a multiparametric 18F-FDG PET/MR examination, comprising a dedicated protocol for imaging of the female pelvis. The primary tumor in the uterine cervix was manually segmented on post-contrast T1-weighted images. Quantitative features were extracted from the segmented tumors using the Radiomic Image Processing Toolbox for the R software environment for statistical computing and graphics. 45 different image features were calculated from non-enhanced as well as post-contrast T1-weighted TSE images, T2-weighted TSE images, the ADC map, the parametric Ktrans, Kep, Ve and iAUC maps and PET images, respectively. Statistical analysis and modeling was performed using Python 3.5 and the scikit-learn software machine learning library for the Python programming language.

RESULTS

 Prediction of M-stage was superior when compared to N-stage. Prediction of M-stage using SVM with SVM-RFE as feature selection obtained the highest performance providing sensitivity of 91 % and specificity of 92 %. Using receiver operating characteristic (ROC) analysis of the pooled predictions, the area under the curve (AUC) was 0.97. Prediction of N-stage using RBF-SVM with MIFS as feature selection reached sensitivity of 83 %, specificity of 67 % and an AUC of 0.82.

CONCLUSION

 M- and N-stage can be predicted based on isolated radiomics analyses of the primary tumor in cervical cancers, thus serving as a template for noninvasive tumor phenotyping and patient stratification using high-dimensional feature vectors extracted from multiparametric PET/MRI data.

KEY POINTS

  · Radiomics analysis based on multiparametric PET/MRI enables prediction of the metastatic status of cervical cancers. · Prediction of M-stage is superior to N-stage. · Multiparametric PET/MRI displays a valuable platform for radiomics analyses .

AI as a New Frontier in Contrast Media Research: Bridging the Gap Between Contrast Media Reduction, the Contrast-Free Question and New Application Discoveries

ABSTRACT

Artificial intelligence (AI) techniques are currently harnessed to revolutionize the domain of medical imaging. This review investigates 3 major AI-driven approaches for contrast agent management: new frontiers in contrast agent dose reduction, the contrast-free question, and new applications. By examining recent studies that use AI as a new frontier in contrast media research, we synthesize the current state of the field and provide a comprehensive understanding of the potential and limitations of AI in this context. In doing so, we show the dose limits of reducing the amount of contrast agents and demonstrate why it might not be possible to completely eliminate contrast agents in the future. In addition, we highlight potential new applications to further increase the radiologist's sensitivity at normal doses. At the same time, this review shows which network architectures provide promising approaches and reveals possible artifacts of a paired image-to-image conversion. Furthermore, current US Food and Drug Administration regulatory guidelines regarding AI/machine learning-enabled medical devices are highlighted.

The importance of educational tools and a new software solution for visualizing and quantifying report correction in radiology training

A novel software, DiffTool, was developed in-house to keep track of changes made by board-certified radiologists to preliminary reports created by residents and evaluate its impact on radiological hands-on training. Before (t0) and after (t2-4) the deployment of the software, 18 residents (median age: 29 years; 33% female) completed a standardized questionnaire on professional training. At t2-4 the participants were also requested to respond to three additional questions to evaluate the software. Responses were recorded via a six-point Likert scale ranging from 1 ("strongly agree") to 6 ("strongly disagree"). Prior to the release of the software, 39% (7/18) of the residents strongly agreed with the statement that they manually tracked changes made by board-certified radiologists to each of their radiological reports while 61% were less inclined to agree with that statement. At t2-4, 61% (11/18) stated that they used DiffTool to track differences. Furthermore, we observed an increase from 33% (6/18) to 44% (8/18) of residents who agreed to the statement "I profit from every corrected report". The DiffTool was well accepted among residents with a regular user base of 72% (13/18), while 78% (14/18) considered it a relevant improvement to their training. The results of this study demonstrate the importance of providing a time-efficient way to analyze changes made to preliminary reports as an additive for professional training.

Information extraction from weakly structured radiological reports with natural language queries

OBJECTIVES

Provide physicians and researchers an efficient way to extract information from weakly structured radiology reports with natural language processing (NLP) machine learning models.

METHODS

We evaluate seven different German bidirectional encoder representations from transformers (BERT) models on a dataset of 857,783 unlabeled radiology reports and an annotated reading comprehension dataset in the format of SQuAD 2.0 based on 1223 additional reports.

RESULTS

Continued pre-training of a BERT model on the radiology dataset and a medical online encyclopedia resulted in the most accurate model with an F1-score of 83.97% and an exact match score of 71.63% for answerable questions and 96.01% accuracy in detecting unanswerable questions. Fine-tuning a non-medical model without further pre-training led to the lowest-performing model. The final model proved stable against variation in the formulations of questions and in dealing with questions on topics excluded from the training set.

CONCLUSIONS

General domain BERT models further pre-trained on radiological data achieve high accuracy in answering questions on radiology reports. We propose to integrate our approach into the workflow of medical practitioners and researchers to extract information from radiology reports.

CLINICAL RELEVANCE STATEMENT

By reducing the need for manual searches of radiology reports, radiologists' resources are freed up, which indirectly benefits patients.

KEY POINTS

• BERT models pre-trained on general domain datasets and radiology reports achieve high accuracy (83.97% F1-score) on question-answering for radiology reports. • The best performing model achieves an F1-score of 83.97% for answerable questions and 96.01% accuracy for questions without an answer. • Additional radiology-specific pretraining of all investigated BERT models improves their performance.

Prognostic value of deep learning-derived body composition in advanced pancreatic cancer-a retrospective multicenter study

BACKGROUND

Despite the prognostic relevance of cachexia in pancreatic cancer, individual body composition has not been routinely integrated into treatment planning. In this multicenter study, we investigated the prognostic value of sarcopenia and myosteatosis automatically extracted from routine computed tomography (CT) scans of patients with advanced pancreatic ductal adenocarcinoma (PDAC).

PATIENTS AND METHODS

We retrospectively analyzed clinical imaging data of 601 patients from three German cancer centers. We applied a deep learning approach to assess sarcopenia by the abdominal muscle-to-bone ratio (MBR) and myosteatosis by the ratio of abdominal inter- and intramuscular fat to muscle volume. In the pooled cohort, univariable and multivariable analyses were carried out to analyze the association between body composition markers and overall survival (OS). We analyzed the relationship between body composition markers and laboratory values during the first year of therapy in a subgroup using linear regression analysis adjusted for age, sex, and American Joint Committee on Cancer (AJCC) stage.

RESULTS

Deep learning-derived MBR [hazard ratio (HR) 0.60, 95% confidence interval (CI) 0.47-0.77, P < 0.005] and myosteatosis (HR 3.73, 95% CI 1.66-8.39, P < 0.005) were significantly associated with OS in univariable analysis. In multivariable analysis, MBR (P = 0.019) and myosteatosis (P = 0.02) were associated with OS independent of age, sex, and AJCC stage. In a subgroup, MBR and myosteatosis were associated with albumin and C-reactive protein levels after initiation of therapy. Additionally, MBR was also associated with hemoglobin and total protein levels.

CONCLUSIONS

Our work demonstrates that deep learning can be applied across cancer centers to automatically assess sarcopenia and myosteatosis from routine CT scans. We highlight the prognostic role of our proposed markers and show a strong relationship with protein levels, inflammation, and anemia. In clinical practice, automated body composition analysis holds the potential to further personalize cancer treatment.

Fully automated MR-based virtual biopsy of primary CNS lymphomas

Background

Primary central nervous system lymphomas (PCNSL) pose a challenge as they may mimic gliomas on magnetic resonance imaging (MRI) imaging, compelling precise differentiation for appropriate treatment. This study focuses on developing an automated MRI-based workflow to distinguish between PCNSL and gliomas.

Methods

MRI examinations of 240 therapy-naive patients (141 males and 99 females, mean age: 55.16 years) with cerebral gliomas and PCNSLs (216 gliomas and 24 PCNSLs), each comprising a non-contrast T1-weighted, fluid-attenuated inversion recovery (FLAIR), and contrast-enhanced T1-weighted sequence were included in the study. HD-GLIO, a pre-trained segmentation network, was used to generate segmentations automatically. To validate the segmentation efficiency, 237 manual segmentations were prepared (213 gliomas and 24 PCNSLs). Subsequently, radiomics features were extracted following feature selection and training of an XGBoost algorithm for classification.

Results

The segmentation models for gliomas and PCNSLs achieved a mean Sørensen-Dice coefficient of 0.82 and 0.80 for whole tumors, respectively. Three classification models were developed in this study to differentiate gliomas from PCNSLs. The first model differentiated PCNSLs from gliomas, with an area under the curve (AUC) of 0.99 (F1-score: 0.75). The second model discriminated between high-grade gliomas and PCNSLs with an AUC of 0.91 (F1-score: 0.6), and the third model differentiated between low-grade gliomas and PCNSLs with an AUC of 0.95 (F1-score: 0.89).

Conclusions

This study serves as a pilot investigation presenting an automated virtual biopsy workflow that distinguishes PCNSLs from cerebral gliomas. Prior to clinical use, it is necessary to validate the results in a prospective multicenter setting with a larger number of PCNSL patients.

Automated image quality assessment for selecting among multiple magnetic resonance image acquisitions in the German National Cohort study

In magnetic resonance imaging (MRI), the perception of substandard image quality may prompt repetition of the respective image acquisition protocol. Subsequently selecting the preferred high-quality image data from a series of acquisitions can be challenging. An automated workflow may facilitate and improve this selection. We therefore aimed to investigate the applicability of an automated image quality assessment for the prediction of the subjectively preferred image acquisition. Our analysis included data from 11,347 participants with whole-body MRI examinations performed as part of the ongoing prospective multi-center German National Cohort (NAKO) study. Trained radiologic technologists repeated any of the twelve examination protocols due to induced setup errors and/or subjectively unsatisfactory image quality and chose a preferred acquisition from the resultant series. Up to 11 quantitative image quality parameters were automatically derived from all acquisitions. Regularized regression and standard estimates of diagnostic accuracy were calculated. Controlling for setup variations in 2342 series of two or more acquisitions, technologists preferred the repetition over the initial acquisition in 1116 of 1396 series in which the initial setup was retained (79.9%, range across protocols: 73-100%). Image quality parameters then commonly showed statistically significant differences between chosen and discarded acquisitions. In regularized regression across all protocols, 'structured noise maximum' was the strongest predictor for the technologists' choice, followed by 'N/2 ghosting average'. Combinations of the automatically derived parameters provided an area under the ROC curve between 0.51 and 0.74 for the prediction of the technologists' choice. It is concluded that automated image quality assessment can, despite considerable performance differences between protocols and anatomical regions, contribute substantially to identifying the subjective preference in a series of MRI acquisitions and thus provide effective decision support to readers.

Frequency and Characteristics of Non-Neurological and Neurological Stroke Mimics in the Emergency Department

BACKGROUND

Stroke mimics are common in the emergency department (ED) and early detection is important to initiate appropriate treatment and withhold unnecessary procedures. We aimed to compare the frequency, clinical characteristics and predictors of non-neurological and neurological stroke mimics transferred to our ED for suspected stroke.

METHODS

This was a cross-sectional study of consecutive patients with suspected stroke transported to the ED of the University Hospital Essen between January 2017 and December 2021 by the city's Emergency Medical Service. We investigated patient characteristics, preclinical data, symptoms and final diagnoses in patients with non-neurological and neurological stroke mimics. Multinominal logistic regression analysis was performed to assess predictors of both etiologic groups.

RESULTS

Of 2167 patients with suspected stroke, 762 (35.2%) were diagnosed with a stroke mimic. Etiology was non-neurological in 369 (48.4%) and neurological in 393 (51.6%) cases. The most common diagnoses were seizures (23.2%) and infections (14.7%). Patients with non-neurological mimics were older (78.0 vs. 72.0 y, p < 0.001) and more likely to have chronic kidney disease (17.3% vs. 9.2%, p < 0.001) or heart failure (12.5% vs. 7.1%, p = 0.014). Prevalence of malignancy (8.7% vs. 13.7%, p = 0.031) and focal symptoms (38.8 vs. 57.3%, p < 0.001) was lower in this group. More than two-fifths required hospitalization (39.3 vs. 47.1%, p = 0.034). Adjusted multinominal logistic regression revealed chronic kidney and liver disease as independent positive predictors of stroke mimics regardless of etiology, while atrial fibrillation and hypertension were negative predictors in both groups. Prehospital vital signs were independently associated with non-neurological stroke mimics only, while age was exclusively associated with neurological mimics.

CONCLUSIONS

Up to half of stroke mimics in the neurological ED are of non-neurological origin. Preclinical identification is challenging and a high proportion requires hospitalization. Awareness of underlying etiologies and differences in clinical characteristics is important to provide optimal care.

Current use of percutaneous image-guided tumor ablation for the therapy of liver tumors: lessons learned from the registry of the German Society for Interventional Radiology and Minimally Invasive Therapy (DeGIR) 2018-2022

OBJECTIVES

Percutaneous image-guided tumor ablation of liver malignancies has become an indispensable therapeutic procedure. The aim of this evaluation of the prospectively managed multinational registry of the voluntary German Society for Interventional Radiology and Minimally Invasive Therapy (DeGIR) was to analyze its use, technical success, and complications in clinical practice.

MATERIALS AND METHODS

All liver tumor ablations from 2018 to 2022 were included. Technical success was defined as complete ablation of the tumor with an ablative margin.

RESULTS

A total of 7228 liver tumor ablations from 136 centers in Germany and Austria were analyzed. In total, 31.4% (2268/7228) of patients were female. Median age was 67 years (IQR 58-74 years). Microwave ablation (MWA) was performed in 65.1% (4703/7228), and radiofrequency ablation (RFA) in 32.7% (2361/7228). Of 5229 cases with reported tumor etiology, 60.3% (3152/5229) of ablations were performed for liver metastases and 37.3% (1950/5229) for hepatocellular carcinoma. The median lesion diameter was 19 mm (IQR 12-27 mm). In total, 91.8% (6636/7228) of ablations were technically successful. The rate of technically successful ablations was significantly higher in MWA (93.9%, 4417/4703) than in RFA (87.3%, 2061/2361) (p < 0.0001). The total complication rate was 3.0% (214/7228) and was significantly higher in MWA (4.0%, 189/4703) than in RFA (0.9%, 21/2361, p < 0.0001). Additional needle track ablation did not increase the rate of major complications significantly (24.8% (33/133) vs. 28.4% (23/81), p = 0.56)).

CONCLUSION

MWA is the most frequent ablation method. Percutaneous image-guided liver tumor ablations have a high technical success rate, which is higher for MWA than RFA. The complication rate is generally low but is higher for MWA than RFA.

CLINICAL RELEVANCE STATEMENT

Percutaneous image-guided liver ablation using microwave ablation and radiofrequency ablation are effective therapeutic procedures with low complication rates for the treatment of primary and secondary liver malignancies.

KEY POINTS

• Percutaneous image-guided liver tumor ablations have a high technical success rate, which is higher for microwave ablation than radiofrequency ablation. • Microwave ablation is the most frequent ablation method ahead of radiofrequency ablation. • The complication rate is generally low but is higher for microwave ablation than radiofrequency ablation.

Determining body height and weight from thoracic and abdominal CT localizers in pediatric and young adult patients using deep learning

In this retrospective study, we aimed to predict the body height and weight of pediatric patients using CT localizers, which are overview scans performed before the acquisition of the CT. We trained three commonly used networks (EfficientNetV2-S, ResNet-18, and ResNet-34) on a cohort of 1009 and 1111 CT localizers of pediatric patients with recorded body height and weight (between January 2013 and December 2019) and validated them in an additional cohort of 116 and 127 localizers (acquired in 2020). The best-performing model was then tested in an independent cohort of 203 and 225 CT localizers (acquired between January 2021 and March 2023). In addition, a cohort of 1401 and 1590 localizers from younger adults (acquired between January 2013 and December 2013) was added to the training set to determine if it could improve the overall accuracy. The EfficientNetV2-S using the additional adult cohort performed best with a mean absolute error of 5.58 ± 4.26 cm for height and 4.25 ± 4.28 kg for weight. The relative error was 4.12 ± 4.05% for height and 11.28 ± 12.05% for weight. Our study demonstrated that automated estimation of height and weight in pediatric patients from CT localizers can be performed.

Treatment of acute ischemic stroke in patients with active malignancy: insight from a comprehensive stroke center

Background

Despite the high incidence of acute ischemic stroke (AIS) in cancer patients, there is still no consensus about the safety of recanalization therapies in this cohort.

Objectives

In this observational study, our aim was to investigate the bleeding risk after acute recanalization therapy in AIS patients with active malignancy.

Methods and Study Design

We retrospectively analyzed observational data of 1016 AIS patients who received intravenous thrombolysis with rtPA (IVT) and/or endovascular therapy (EVT) between January 2017 and December 2020 with a focus on patients with active malignancy. The primary safety endpoint was the occurrence of stroke treatment-related major bleeding events, that is, symptomatic intracranial hemorrhage (SICH) and/or relevant systemic bleeding. The primary efficacy endpoint was neurological improvement during hospital stay (NI).

Results

None of the 79 AIS patients with active malignancy suffered from stroke treatment-related systemic bleeding. The increased rate (7.6% versus 4.7%) of SICH after therapy compared to the control group was explained by confounding factors. A total of nine patients with cerebral tumor manifestation received acute stroke therapy, two of them suffered from stroke treatment-related intracranial hemorrhage remote from the tumor, both asymptomatic. The group of patients with active malignancy and the control group showed comparable rates of NI.

Conclusion

Recanalization therapy in AIS patients with active malignancy was not associated with a higher risk for stroke treatment-related systemic or intracranial bleeding. IVT and/or EVT can be regarded as a safe therapy option for AIS patients with active malignancy.

Embolization of active arterial bleeding in COVID-19 patients: A multicenter study

PURPOSE

The purpose of this study was to assess the efficacy of transarterial embolization in COVID-19 patients with an arterial bleeding and to investigate differences between various patient groups concerning survival.

METHOD

We retrospectively reviewed COVID-19 patients undergoing transarterial embolization due to an arterial bleeding in a multicenter study from April 2020 to July 2022 and analyzed the technical success of embolization and survival rate. 30-day survival between various patient groups was analyzed. The Chi- square test and Fisher's exact test were used for testing association between the categorical variables.

RESULTS

53 COVID-19 patients (age: 57.3 ± 14.3 years, 37 male) received 66 angiographies due to an arterial bleeding. The initial embolization was technically successful in 98.1% (52/53). In 20.8% (11/53) of patients, additional embolization was necessary due to a new arterial bleeding. A majority of 58.5% (31/53) had a severe course of COVID-19 infection necessitating ECMO-therapy and 86.8% (46/53) of patients received anticoagulation. 30-day survival rate in patients with ECMO-therapy was significantly lower than without ECMO-therapy (45.2% vs. 86.4%, p = 0.004). Patients with anticoagulation did not have a lower 30-day survival rate than without anticoagulation (58.7% vs. 85.7%, p = 0.23). COVID-19 patients with ECMO-therapy developed more frequently a re-bleeding after embolization than non-ECMO-patients (32.3% vs. 4.5%, p = 0.02).

CONCLUSIONS

Transarterial embolization is a feasible, safe, and effective procedure in COVID-19 patients with arterial bleeding. ECMO-patients have a lower 30-day survival rate than non-ECMO-patients and have an increased risk for re-bleeding. Treatment with anticoagulation could not be identified as a risk factor for higher mortality.

Evaluation of [68Ga]-DOTATOC PET/MRI in Patients with Meningioma of the Subcranial and Intraorbital Space

Meningiomas are known to express somatostatin receptor (SSTR) type 2 to a high degree. Therefore, radiolabeled somatostatin analogs, such as DOTATOC, have been introduced for PET imaging of meningiomas. However, the benefit of hybrid SSTR PET/MRI is still debated. Here, we report our experience with [⁶⁸Ga]-DOTATOC PET/MRI. Methods: PET/MRI was performed in 60 patients with suspected or diagnosed meningiomas of the skull plane and eye socket. Acquired datasets were reported by 2 independent readers regarding local tumor extent and signal characteristics. Histopathologic results and follow-up imaging served as the reference standard. SUVs of target lesions were analyzed according to the corresponding maximal tracer uptake. The diagnostic accuracy of PET/MRI and conventional MRI was determined independently and compared with the reference standard. Results: In total, 60 target lesions were identified, with 54 considered to be meningiomas according to the reference standard. Sensitivity and specificity of PET/MRI versus MRI alone were 95% versus 96% and 75% versus 66%, respectively. The McNemar test was not able to distinguish any differences between PET/MRI and the reference standard or MRI and the reference standard. No differences were found between the 2 modalities with respect to local infiltration. Conclusion: SSTR PET/MRI and MRI yielded similar accuracy for the detection of meningiomas of the skull base and intraorbital space. Here, sequential low-dose SSTR PET/CT might be helpful for the planning of radioligand therapy or radiotherapy.

Contrast Agent Dose Reduction in MRI Utilizing a Generative Adversarial Network in an Exploratory Animal Study

OBJECTIVES

The aim of this study is to use virtual contrast enhancement to reduce the amount of hepatobiliary gadolinium-based contrast agent in magnetic resonance imaging with generative adversarial networks (GANs) in a large animal model.

METHODS

With 20 healthy Göttingen minipigs, a total of 120 magnetic resonance imaging examinations were performed on 6 different occasions, 50% with reduced (low-dose; 0.005 mmol/kg, gadoxetate) and 50% standard dose (normal-dose; 0.025 mmol/kg). These included arterial, portal venous, venous, and hepatobiliary contrast phases (20 minutes, 30 minutes). Because of incomplete examinations, one animal had to be excluded. Randomly, 3 of 19 animals were selected and withheld for validation (18 examinations). Subsequently, a GAN was trained for image-to-image conversion from low-dose to normal-dose (virtual normal-dose) with the remaining 16 animals (96 examinations). For validation, vascular and parenchymal contrast-to-noise ratio (CNR) was calculated using region of interest measurements of the abdominal aorta, inferior vena cava, portal vein, hepatic parenchyma, and autochthonous back muscles. In parallel, a visual Turing test was performed by presenting the normal-dose and virtual normal-dose data to 3 consultant radiologists, blinded for the type of examination. They had to decide whether they would consider both data sets as consistent in findings and which images were from the normal-dose study.

RESULTS

The pooled dynamic phase vascular and parenchymal CNR increased significantly from low-dose to virtual normal-dose (pooled vascular: P < 0.0001, pooled parenchymal: P = 0.0002) and was found to be not significantly different between virtual normal-dose and normal-dose examinations (vascular CNR [mean ± SD]: low-dose 17.6 ± 6.0, virtual normal-dose 41.8 ± 9.7, and normal-dose 48.4 ± 12.2; parenchymal CNR [mean ± SD]: low-dose 20.2 ± 5.9, virtual normal-dose 28.3 ± 6.9, and normal-dose 29.5 ± 7.2). The pooled parenchymal CNR of the hepatobiliary contrast phases revealed a significant increase from the low-dose (22.8 ± 6.2) to the virtual normal-dose (33.2 ± 6.1; P < 0.0001) and normal-dose sequence (37.0 ± 9.1; P < 0.0001). In addition, there was no significant difference between the virtual normal-dose and normal-dose sequence. In the visual Turing test, on the median, the consultant radiologist reported that the sequences of the normal-dose and virtual normal-dose are consistent in findings in 100% of the examinations. Moreover, the consultants were able to identify the normal-dose series as such in a median 54.5% of the cases.

CONCLUSIONS

In this feasibility study in healthy Göttingen minipigs, it could be shown that GAN-based virtual contrast enhancement can be used to recreate the image impression of normal-dose imaging in terms of CNR and subjective image similarity in both dynamic and hepatobiliary contrast phases from low-dose data with an 80% reduction in gadolinium-based contrast agent dose. Before clinical implementation, further studies with pathologies are needed to validate whether pathologies are correctly represented by the network.

Individualized scan protocols for CT angiography: an animal study for contrast media or radiation dose optimization

BACKGROUND

We investigated about optimization of contrast media (CM) dose or radiation dose in thoracoabdominal computed tomography angiography (CTA) by automated tube voltage selection (ATVS) system configuration and CM protocol adaption.

METHODS

In six minipigs, CTA-optimized protocols were evaluated regarding objective (contrast-to-noise ratio, CNR) and subjective (6 criteria assessed by Likert scale) image quality. Scan parameters were automatically adapted by the ATVS system operating at 90-kV semi-mode and configured for standard, CM saving, or radiation dose saving (image task, quality settings). Injection protocols (dose, flow rate) were adapted manually. This approach was tested for normal and simulated obese conditions.

RESULTS

Radiation exposure (volume-weighted CT dose index) for normal (obese) conditions was 2.4 ± 0.7 (5.0 ± 0.7) mGy (standard), 4.3 ± 1.1 (9.0 ± 1.3) mGy (CM reduced), and 1.7 ± 0.5 (3.5 ± 0.5) mGy (radiation reduced). The respective CM doses for normal (obese) settings were 210 (240) mgI/kg, 155 (177) mgI/kg, and 252 (288) mgI/kg. No significant differences in CNR (normal; obese) were observed between standard (17.8 ± 3.0; 19.2 ± 4.0), CM-reduced (18.2 ± 3.3; 20.5 ± 4.9), and radiation-saving CTAs (16.0 ± 3.4; 18.4 ± 4.1). Subjective analysis showed similar values for optimized and standard CTAs. Only the parameter diagnostic acceptability was significantly lower for radiation-saving CTA compared to the standard CTA.

CONCLUSIONS

The CM dose (-26%) or radiation dose (-30%) for thoracoabdominal CTA can be reduced while maintaining objective and subjective image quality, demonstrating the feasibility of the personalization of CTA scan protocols.

KEY POINTS

• Computed tomography angiography protocols could be adapted to individual patient requirements using an automated tube voltage selection system combined with adjusted contrast media injection. • Using an adapted automated tube voltage selection system, a contrast media dose reduction (-26%) or radiation dose reduction (-30%) could be possible.

Germany's journey toward 14 Tesla human magnetic resonance

Multiple sites within Germany operate human MRI systems with magnetic fields either at 7 Tesla or 9.4 Tesla. In 2013, these sites formed a network to facilitate and harmonize the research being conducted at the different sites and make this technology available to a larger community of researchers and clinicians not only within Germany, but also worldwide. The German Ultrahigh Field Imaging (GUFI) network has defined a strategic goal to establish a 14 Tesla whole-body human MRI system as a national research resource in Germany as the next progression in magnetic field strength. This paper summarizes the history of this initiative, the current status, the motivation for pursuing MR imaging and spectroscopy at such a high magnetic field strength, and the technical and funding challenges involved. It focuses on the scientific and science policy process from the perspective in Germany, and is not intended to be a comprehensive systematic review of the benefits and technical challenges of higher field strengths.

Circadian rhythm and aneurysmal subarachnoid hemorrhage: Is there an alarm clock for the rupture timing?

PURPOSE AND BACKGROUND

Data on the temporal distribution of the bleeding time of intracranial aneurysms is limited to few small studies. With this study, we aimed to analyze time patterns of the occurrence of aneurysmal subarachnoid hemorrhage (SAH), particularly focusing on the impact of patients' socio-demographic and clinical characteristics on the ictus timing.

METHODS

This study is based on the institutional SAH cohort with 782 consecutive cases treated between January 2003 and June 2016. Data was collected on the ictus time, patients' socio-demographic and clinical characteristics, as well as the initial severity and outcome. Univariate and multivariate analyses were performed on the bleeding timeline.

RESULTS

There were two peaks in the circadian rhythm of SAH, one in the morning (7 to 9 a.m.) and the other in the evening (7 to 9 p.m.). The strongest alterations in the bleeding time patterns were observed for weekdays, patients' age, sex, and ethnicity. Individuals with chronic alcohol and painkiller consumption showed a higher bleeding peak between 1 and 3 p.m. Finally, the bleeding time showed no impact on the severity, clinically relevant complications and the outcome of SAH patients.

CONCLUSIONS

This study is one of the very few detailed analyses of the impact of specific socio-demographic, ethnic, behavioral and clinical characteristics on the rupture timing of aneurysms. Our results point to the possible relevance of the circadian rhythm for the rupture event, and therefore might be useful in the elaboration of preventive measures against aneurysm rupture.

AI co-pilot: content-based image retrieval for the reading of rare diseases in chest CT

The aim of the study was to evaluate the impact of the newly developed Similar patient search (SPS) Web Service, which supports reading complex lung diseases in computed tomography (CT), on the diagnostic accuracy of residents. SPS is an image-based search engine for pre-diagnosed cases along with related clinical reference content ( https://eref.thieme.de ). The reference database was constructed using 13,658 annotated regions of interest (ROIs) from 621 patients, comprising 69 lung diseases. For validation, 50 CT scans were evaluated by five radiology residents without SPS, and three months later with SPS. The residents could give a maximum of three diagnoses per case. A maximum of 3 points was achieved if the correct diagnosis without any additional diagnoses was provided. The residents achieved an average score of 17.6 ± 5.0 points without SPS. By using SPS, the residents increased their score by 81.8% to 32.0 ± 9.5 points. The improvement of the score per case was highly significant (p = 0.0001). The residents required an average of 205.9 ± 350.6 s per case (21.9% increase) when SPS was used. However, in the second half of the cases, after the residents became more familiar with SPS, this increase dropped to 7%. Residents' average score in reading complex chest CT scans improved by 81.8% when the AI-driven SPS with integrated clinical reference content was used. The increase in time per case due to the use of the SPS was minimal.

Radiation Exposure in Computed Tomography

BACKGROUND

Computed tomography (CT) studies are requested by specialists from most medical disciplines and play a vital role in the diagnosis and treatment of patients. It follows that physicians of all specialties should possess basic knowledge of computed tomography, its proper use, and the radiation exposure associated with it.

METHODS

This review is based on publications retrieved by a selective search of the literature.

RESULTS

Approximately 12 million CT studies are carried out in Germany each year, and the trend is rising. Approximately 9% of all diagnostic studies involving ionizing radiation are CT studies. On average, more than 60% of the collective effective dose due to medical radiation exposure is attributable to CT. There are two types of radiation effects caused by ionizing radiation: stochastic and deterministic. The additional, individual relative lifetime cancer mortality risk due to ionizing radiation with whole-body exposure at a low single dose is estimated at 5% per sievert. Radiation exposure from CT studies of the head and trunk, e.g. of a patient with polytrauma, corresponds to an additional lifetime cancer mortality risk of approximately 0.1% at an effective dose of approximately 20 millisievert.

CONCLUSION

The radiation exposure due to CT, and the risks to which patients are subjected by it, have become more important with greater use of CT. Technical advances, targeted dose monitoring, and analyses of dose data can help identify areas where improvement is necessary, in furtherance of the overriding goal of lowering patients' radiation exposure while preserving adequate image quality.

Results of interventional treatment of peripheral slow-flow malformations

BACKGROUND

In recent years sclerotherapy has increasingly become the treatment of choice for peripheral slow-flow malformations. However, the long-term effectiveness of sclerotherapy is still a matter of debate, especially when it comes to new sclerosing agents like polidocanol. This study aims at gathering further information concerning its long-term effectiveness and safety.

RESULTS

Most patients reported a reduction of symptoms which include pain (57,7%), swelling (65,4%) and functional impairment (60%). Cosmetic complaints were less likely to be reduced by sclerotherapy (44,4%). In most cases a relief of symptoms was stable for many years, especially after several consecutive treatment sessions. Complication rates were comparably low, with only 2 patients requiring additional treatment at hospital and no lasting damages. (…) (7) Most patients (70,9%) were at least partially satisfied with the treatment. Satisfaction was closely linked to a partial or complete relief of symptoms (p = 0.001).

CONCLUSION

Sclerotherapy is a promising way of treating slow-flow-malformations. Polidocanol has proved to be a save sclerosing agent. The reduction of major symptoms was substantial in most cases and lasted for many years.

Management of carotid cavernous fistulas: A single center experience

Purpose

Multimodal endovascular therapy (EVT) of carotid cavernous fistula (CCF) with different approaches and a variety of available embolization material enable high occlusion rates with good clinical and functional outcome but until now there is still little evidence available. This retrospective single-center study aims to evaluate EVT of CCF with different neuroendovascular techniques regarding occlusion rates, complications and outcomes.

Materials and methods

From 2001 to 2021 59 patients with CCF were treated at our tertiary university hospital. Patient records and all imaging data including angiograms were reviewed for demographic and epidemiological data, symptoms, fistula type, number of EVTs, complications of EVT, type of embolic materials, occlusion rates and recurrences.

Results

Etiology of the CCF were spontaneous (41/59, 69.5%) post-traumatic (13/59, 22%) and ruptured cavernous aneurysms (5/59, 8.5%). Endovascular therapy was completed in one session in 74.6% (44/59). Transvenous access was most frequent (55.9% 33/59) followed by transarterial catheterization in 33.9% (20/59) and a combination of both (6/59, 10.2%). Exclusively coils were used in 45.8% (27/59), a combination of ethylene vinyl alcohol (EVOH) copolymer (Onyx) and coils in 42.4% (25/59). Complete obliteration was achieved in 96.6% of patients (57/59) with an intraprocedural-related complication rate of 5.1% (3/59) and no mortality.

Conclusion

Endovascular therapy of CCF has been shown to be safe and effective with high cure rates and low rates of intraprocedural complications and morbidity even in complex scenarios.

Deep learning-based assessment of body composition and liver tumour burden for survival modelling in advanced colorectal cancer

BACKGROUND

Personalized therapy planning remains a significant challenge in advanced colorectal cancer care, despite extensive research on prognostic and predictive markers. A strong correlation of sarcopenia or overall body composition and survival has been described. Here, we explore whether automated assessment of body composition and liver metastases from standard of care CT images can add to clinical parameters in personalized survival risk prognostication.

METHODS

We retrospectively analysed clinical imaging data from 85 patients (50.6% female, mean age 58.9 SD 12.2 years) with colorectal cancer and synchronous liver metastases. Pretrained deep learning models were used to assess body composition and liver metastasis geometry from abdominal CT images before the initiation of systemic treatment. Abdominal muscle-to-bone ratio (MBR) was calculated by dividing abdominal muscle volume by abdominal bone volume. MBR was compared with body mass index (BMI), abdominal muscle volume, and abdominal muscle volume divided by height squared. Differences in overall survival based on body composition and liver metastasis parameters were compared using Kaplan-Meier survival curves. Results were correlated with clinical and biomarker data to develop a machine learning model for survival risk prognostication.

RESULTS

The MBR, unlike abdominal muscle volume or BMI, was significantly associated with overall survival (HR 0.39, 95% CI: 0.19-0.80, P = 0.009). The MBR (P = 0.022), liver metastasis surface area (P = 0.01) and primary tumour sidedness (P = 0.007) were independently associated with overall survival in multivariate analysis. Body composition parameters did not correlate with KRAS mutational status or primary tumour sidedness. A prediction model based on MBR, liver metastasis surface area and primary tumour sidedness achieved a concordance index of 0.69.

CONCLUSIONS

Automated segmentation enables to extract prognostic parameters from routine imaging data for personalized survival modelling in advanced colorectal cancer patients.

Detection of Clinically Significant Prostate Cancer Using Targeted Biopsy with Four Cores Versus Target Saturation Biopsy with Nine Cores in Transperineal Prostate Fusion Biopsy: A Prospective Randomized Trial

BACKGROUND

Multiparametric magnetic resonance imaging (mpMRI) and targeted biopsy (TB) facilitate accurate detection of clinically significant prostate cancer (csPC). However, it remains unclear how targeted cores should be applied for accurate diagnosis of csPC.

OBJECTIVE

To assess csPC detection rates for two target-directed MRI/transrectal ultrasonography (TRUS) fusion biopsy approaches, conventional TB and target saturation biopsy (TS).

DESIGN, SETTING, AND PARTICIPANTS

This was a prospective single-center study of outcomes for transperineal MRI/TRUS fusion biopsies for 170 men. Half of the men (n = 85) were randomized to conventional TB with four cores per lesion and half (n = 85) to TS with nine cores. Biopsies were performed by three experienced board-certified urologists.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

PC and csPC (International Society of Urological Pathology grade group ≥2) detection rates for systematic biopsy (SB), TB, and TS were analyzed using McNemar's test for intrapatient comparisons and Fisher's exact test for TS versus TB. A combination of targeted biopsy (TS or TB) and SB served as the reference.

RESULTS AND LIMITATIONS

According to the reference, csPC was diagnosed for 57 men in the TS group and 36 men in the TB group. Of these, TS detected 57/57 csPC cases and TB detected 33/36 csPC cases (p = 0.058). Detection of Gleason grade group 1 disease was 10/12 cases with TS and 8/17 cases with TB (p = 0.055). In addition, TS detected 97% of 63 csPC lesions, compared to 86% with TB (p = 0.1). Limitations include the single-center design, the limited generalizability owing to the transperineal biopsy route, the lack of central review of pathology and radical prostatectomy correlation, and uneven distributions of csPC prevalence, Prostate Imaging-Reporting and Data System (PI-RADS) 5 lesions, men with two or more PI-RADS ≥3 lesions, and prostate-specific antigen density between the groups, which may have affected the results.

CONCLUSIONS

In our study, rates of csPC detection did not significantly differ between TS and TB.

PATIENT SUMMARY

In this study, we investigated two targeted approaches for taking prostate biopsy samples after observation of suspicious lesions on prostate scans. We found that the rates of detection of prostate cancer did not significantly differ between the two approaches.

Patterns of cardiac involvement in different muscular dystrophies assessed by magnetic resonance imaging

BACKGROUND

In muscular dystrophies, it is not only skeletal muscles that can be affected, but also the myocardium. This cardiac involvement can represent a major cause of morbidity and mortality.

PURPOSE

To investigate cardiac involvement in Duchenne (DMD), Becker (BMD), and limb girdle muscular dystrophy (LGMD) patients, and carriers of DMD/BMD by cardiac magnetic resonance (CMR) imaging and to search for differences in the pattern of cardiac involvement.

MATERIAL AND METHODS

All patients with genetically or histologically proven DMD, BMD, and LGMD, or confirmed carriers of DMD/BMD who had undergone CMR at our clinic between January 2008 and November 2018 were retrospectively included and re-evaluated for regional and global left ventricular function, increased trabecularization, and late enhancement.

RESULTS

A total of 26 DMD, 10 BMD, 11 LGMD, and seven DMD/BMD carriers were included. Only one carrier of DMD presented with normal CMR results; all other participants showed cardiac abnormalities. Regional wall motion abnormalities (RWMA; prevalence in LGMD patients: 55%) and late enhancement (prevalence in LGMD patients: 82%) were frequent. RWMA were accentuated basal inferolateral in DMD/BMD carriers, while in LGMD they were accentuated apical. In all groups late enhancement was located mainly subepicardial/midmyocardial with a basal inferolateral accentuation. Apart from the different RWMA distribution, no further group-specific differences were found.

CONCLUSION

We found a high rate of cardiac involvement not only in DMD/BMD, but also in LGMD and DMD/BMD carriers with a different RWMA accentuation (apical in LGMD and basal inferolateral in DMD/BMD) as a single group-specific difference.

Small intracranial aneurysms of the anterior circulation: A negligible risk?

BACKGROUND AND PURPOSE

According to the International Study of Unruptured Intracranial Aneurysms, small (<7 mm) unruptured intracranial aneurysms (IAs) of the anterior circulation (aC) carry a neglectable 5-year rupture risk. In contrast, some studies report frequencies of >20% of all ruptured IAs being small IAs of the aC. This contradiction was addressed in this study by analyzing the rates and risk factors for rupture of small IAs within the aC.

METHODS

Of the institutional observational cohort, 1676 small IAs of the aC were included. Different demographic, clinical, laboratory, and radiographic characteristics were collected. A rupture risk score was established using all independent prognostic factors. The score performance was checked using receiver operating characteristic curve analysis.

RESULTS

Of all registered small IAs of the aC, 20.1% were ruptured. The developed small IAs of the aC (SIAAC) score (range = -4 to +13 points) contained five major risk factors: IA location and size, arterial hypertension, alcohol abuse, and chronic renal failure. In addition, three putative protective factors were also included in the score: hypothyroidism, dyslipidemia, and peripheral arterial disease. Increasing rates of ruptured IA with increasing SIAAC scores were observed, from 0% (≤-1 points) through >50% (≥8 points) and up to 100% in patients scoring ≥12 points. The SIAAC score achieved excellent discrimination (area under the curve_SIAAC  = 0.803) and performed better than the PHASES (Population,Hypertension, Age, Size of the aneurysm, Earlier SAH from another aneurysm, Site of aneurysm) score.

CONCLUSIONS

Small IAs of the aC carry a considerable rupture risk. After external validation, the proposed rupture risk score might provide a basis for better decision-making regarding the treatment of small unruptured IAs of the aC.

Predictors of Symptomatic Intracranial Hemorrhage after Endovascular Thrombectomy in Acute Ischemic Stroke Patients with Anterior Large Vessel Occlusion-Procedure Time and Reperfusion Quality Determine

PURPOSE

We aimed to evaluate predictors of symptomatic intracranial hemorrhage (sICH) in acute ischemic stroke (AIS) patients following thrombectomy due to anterior large vessel occlusion (LVO).

METHODS

Data on stroke patients from January 2018 to December 2020 in a tertiary care centre were retrospectively analysed. sICH was defined as intracranial hemorrhage associated with a deterioration of at least four points in the National Institutes of Health Stroke Scale (NIHSS) score or hemorrhage leading to death. A smoothed ridge regression model was run to analyse the impact of 15 variables on their association with sICH.

RESULTS

Of the 174 patients (median age 77, 41.4% male), sICH was present in 18 patients. Short procedure time from groin puncture to reperfusion (per 10 min OR 1.24; 95% CI 1.071-1.435; p = 0.004) and complete reperfusion (TICI 3) (OR 0.035; 95% CI 0.003-0.378; p = 0.005) were significantly associated with a lower risk of sICH. On the contrary, successful reperfusion (TICI 3 and TICI 2b) was not associated with a lower risk of sICH (OR 0.508; 95% CI 0.131-1.975, p = 0.325). Neither the total time from symptom onset to reperfusion nor the intravenous thrombolysis was a predictor of sICH (per 10 min OR 1.0; 95% CI 0.998-1.001, p = 0.745) (OR 1.305; 95% CI 0.338-5.041, p = 0.697).

CONCLUSION

Our findings addressed the paramount importance of short procedure time and complete reperfusion to minimize sICH risk. The total ischemic time from onset to reperfusion was not a predictor of sICH.

CT and DSA for evaluation of spontaneous intracerebral lobar bleedings

Purpose

This study retrospectively examined the extent to which computed tomography angiography (CTA) and digital subtraction angiography (DSA) can help identify the cause of lobar intracerebral bleeding.

Materials and methods

In the period from 2002 to 2020, data from patients who were >18 years at a university and an academic teaching hospital with lobar intracerebral bleeding were evaluated retrospectively. The CTA DSA data were reviewed separately by two neuroradiologists, and differences in opinion were resolved by consensus after discussion. A positive finding was defined as an underlying vascular etiology of lobar bleeding.

Results

The data of 412 patients were retrospectively investigated. DSA detected a macrovascular cause of bleeding in 125/412 patients (33%). In total, sixty patients had AVMs (15%), 30 patients with aneurysms (7%), 12 patients with vasculitis (3%), and 23 patients with dural fistulas (6%). The sensitivity, specificity, positive and negative predictive values, and accuracy of CTA compared with DSA were 93, 97, 100, and 97%. There were false-negative CTA readings for two AVMs and one dural fistula.

Conclusion

The DSA is still the gold standard diagnostic modality for detecting macrovascular causes of ICH; however, most patients with lobar ICH can be investigated first with CTA, and the cause of bleeding can be found. Our results showed higher sensitivity and specificity than those of other CTA studies.

Platelet depletion does not alter long-term functional outcome after cerebral ischaemia in mice

Platelets are key mediators of thrombus formation and inflammation during the acute phase of ischaemic stroke. Particularly, the platelet glycoprotein (GP) receptors GPIbα and GPVI have been shown to mediate platelet adhesion and activation in the ischaemic brain. GPIbα and GPVI blockade could reduce infarct volumes and improve functional outcome in mouse models of acute ischaemic stroke, without concomitantly increasing intracerebral haemorrhage. However, the functional role of platelets during long-term stroke recovery has not been elucidated so far.

Thus, we here examined the impact of platelet depletion on post-stroke recovery after transient middle cerebral artery occlusion (tMCAO) in adult male mice. Platelet depleting antibodies or isotype control were applied from day 3–28 after tMCAO in mice matched for infarct size. Long-term functional recovery was assessed over the course of 28 days by behavioural testing encompassing motor and sensorimotorical functions, as well as anxiety-like or spontaneous behaviour. Whole brain flow cytometry and light sheet fluorescent microscopy were used to identify resident and infiltrated immune cell types, and to determine the effects of platelet depletion on the cerebral vascular architecture, respectively.

We found that delayed platelet depletion does not improve long-term functional outcome in the tMCAO stroke model. Immune cell abundance, the extent of thrombosis and the organisation of the cerebral vasculature were also comparable between platelet-depleted and control mice. Our study demonstrates that, despite their critical role in the acute stroke setting, platelets appear to contribute only marginally to tissue reorganisation and functional recovery at later stroke stages.

•

Stable and safe global platelet depletion can be achieved for a prolonged period.

•

Platelets only play a minor role in neurological recovery during the chronic phase.

•

Platelet depletion after infarct maturation does not alter inflammatory response.

•

Cerebral architecture after stroke is not influenced by delayed platelet depletion.

Contrast Media Reduction in Computed Tomography With Deep Learning Using a Generative Adversarial Network in an Experimental Animal Study

OBJECTIVE

This feasibility study aimed to use optimized virtual contrast enhancement through generative adversarial networks (GAN) to reduce the dose of iodine-based contrast medium (CM) during abdominal computed tomography (CT) in a large animal model.

METHODS

Multiphasic abdominal low-kilovolt CTs (90 kV) with low (low CM, 105 mgl/kg) and normal contrast media doses (normal CM, 350 mgl/kg) were performed with 20 healthy Göttingen minipigs on 3 separate occasions for a total of 120 examinations. These included an early arterial, late arterial, portal venous, and venous contrast phase. One animal had to be excluded because of incomplete examinations. Three of the 19 animals were randomly selected and withheld for validation (18 studies). Subsequently, the GAN was trained for image-to-image conversion from low CM to normal CM (virtual CM) with the remaining 16 animals (96 examinations). For validation, region of interest measurements were performed in the abdominal aorta, inferior vena cava, portal vein, liver parenchyma, and autochthonous back muscles, and the contrast-to-noise ratio (CNR) was calculated. In addition, the normal CM and virtual CM data were presented in a visual Turing test to 3 radiology consultants. On the one hand, they had to decide which images were derived from the normal CM examination. On the other hand, they had to evaluate whether both images are pathological consistent.

RESULTS

Average vascular CNR (low CM 6.9 ± 7.0 vs virtual CM 28.7 ± 23.8, P < 0.0001) and parenchymal (low CM 1.5 ± 0.7 vs virtual CM 3.8 ± 2.0, P < 0.0001) CNR increased significantly by GAN-based contrast enhancement in all contrast phases and was not significantly different from normal CM examinations (vascular: virtual CM 28.7 ± 23.8 vs normal CM 34.2 ± 28.8; parenchymal: virtual CM 3.8 ± 2.0 vs normal CM 3.7 ± 2.6). During the visual Turing testing, the radiology consultants reported that images from normal CM and virtual CM were pathologically consistent in median in 96.5% of the examinations. Furthermore, it was possible for the examiners to identify the normal CM data as such in median in 91% of the cases.

CONCLUSIONS

In this feasibility study, it could be demonstrated in an experimental setting with healthy Göttingen minipigs that the amount of CM for abdominal CT can be reduced by approximately 70% by GAN-based contrast enhancement with satisfactory image quality.

Utility of Integrated PET/MRI for the Primary Diagnostic Work-Up of Patients with Ewing Sarcoma: Preliminary Results

Background: This study was conducted to evaluate the clinical applicability of integrated PET/MRI for staging and monitoring the effectiveness of neoadjuvant chemotherapy in Ewing sarcoma patients. Methods: A total of 11 juvenile patients with confirmed Ewing sarcoma, scheduled for induction polychemotherapy, were prospectively enrolled for a PET/MR examination before, during and after the end of treatment. Two experienced physicians analysed the imaging datasets. They were asked to perform a whole-body staging in all three examinations and to define treatment response according to the RECIST1.1 and PERCIST criteria for each patient. Results: In eight patients lymph node and/or distant metastases were detected at initial diagnosis. According to the reference standard, three patients achieved complete response, six patients partial response, and one patient showed stable disease while another patient showed progressive disease. RECIST1.1 categorized the response to treatment in 5/11 patients correctly and showed a tendency to underestimate the response to treatment in the remaining six patients. PERCIST defined response to treatment in 9/11 patients correctly and misclassified two patients with a PR as CR. Conclusion: PET/MRI may serve as a valuable imaging tool for primary staging and response assessment of juvenile patients with Ewing sarcoma to induction chemotherapy, accompanied by a reasonable radiation dose for the patient.

Mechanical thrombectomy for acute ischemic stroke in COVID-19 patients: multicenter experience in 111 cases

BACKGROUND

Data on the frequency and outcome of mechanical thrombectomy (MT) for large vessel occlusion (LVO) in patients with COVID-19 is limited. Addressing this subject, we report our multicenter experience.

METHODS

A retrospective cohort study was performed of consecutive acute stroke patients with COVID-19 infection treated with MT at 26 tertiary care centers between January 2020 and November 2021. Baseline demographics, angiographic outcome and clinical outcome evaluated by the modified Rankin Scale (mRS) at discharge and 90 days were noted.

RESULTS

We identified 111 out of 11 365 (1%) patients with acute or subsided COVID-19 infection who underwent MT due to LVO. Cardioembolic events were the most common etiology for LVO (38.7%). Median baseline National Institutes of Health Stroke Scale score and Alberta Stroke Program Early CT Score were 16 (IQR 11.5-20) and 9 (IQR 7-10), respectively. Successful reperfusion (mTICI ≥2b) was achieved in 97/111 (87.4%) patients and 46/111 (41.4%) patients were reperfused completely. The procedure-related complication rate was 12.6% (14/111). Functional independence was achieved in 20/108 (18.5%) patients at discharge and 14/66 (21.2%) at 90 days follow-up. The in-hospital mortality rate was 30.6% (33/108). In the subgroup analysis, patients with severe acute COVID-19 infection requiring intubation had a mortality rate twice as high as patients with mild or moderate acute COVID-19 infection. Acute respiratory failure requiring ventilation and time interval from symptom onset to groin puncture were independent predictors for an unfavorable outcome in a logistic regression analysis.

CONCLUSION

Our study showed a poor clinical outcome and high mortality, especially in patients with severe acute COVID-19 infection undergoing MT due to LVO.

High volume retrograde portography for better discrimination of the portal vein during TIPS procedure

Background: Imaging of the portal vein prior to puncture for TIPS is essential.

Purpose: With this study, we examined a modified retrograde portography with regard to the reliable representation of the portal vein.

Material and Methods: Prospective evaluation of 65 TIPS interventions with regard to the delimitation of the portal vein and the exact parameters of retrograde portography such as catheter diameter and contrast medium volume per injection.

Results: Retrograde portographies with a large-lumen catheter (10 F) and a large contrast medium volume (40 mL) were performed in 35/63 patients with significantly better delineation of the portal vein than when using 5 F catheters with 10 mL contrast medium.

Conclusion: The so-called high volume retrograde portography leads to better delimitation of the portal vein during TIPS application.

Radiation exposure, organ and effective dose of CT-guided liver biopsy as a function of lesion depth and size

Computed tomography (CT)-guided percutaneous biopsies play an important role in the diagnostic workup of liver lesions. Because radiation dose accumulates rapidly due to repeated image acquisition in a relatively small scan area, analysing radiation exposure is critical for improving radiation protection of CT-guided interventions. The aim of this study was to assess the radiation dose of CT-guided liver biopsies and the influence of lesion parameters, and to establish a local diagnostic reference level (DRL). In this observational retrospective cohort study, dose data of 60 CT-guided liver biopsies between September 2016 and July 2017 were analysed. Radiation exposure was reported for volume-weighted CT dose index (CTDI_vol), size-specific dose estimate (SSDE), dose-length product (DLP) and effective dose (ED). Radiation dose of CT-guided liver biopsy was (median (interquartile range)): CTDI_vol9.91 mGy (8.33-11.45 mGy), SSDE 10.42 mGy (9.39-11.70 mGy), DLP 542 mGy cm (410-733 mGy cm), ED 8.52 mSv (7.17-13.25 mSv). Radiation exposure was significantly higher in biopsies of deep liver lesions compared to superficial lesions (DLP 679 ± 285 mGy cm vs. 497 ± 167 mGy cm,p= 0.0046). No significant dose differences were observed for differences in lesion or needle size. With helical CT spirals additional to the biopsy-guiding axial CT scans, radiation exposure was significantly increased: 797 ± 287 mGy cm vs. 495 ± 162 mGy cm,p< 0.0001. The local DRL is CTDI_vol9.91 mGy, DLP 542 mGy cm. Radiation dose is significantly increased in biopsies of deeper liver lesions compared with superficial lesions. Interventions with additional biopsy-guiding CT spirals lead to higher radiation doses. This study provides a detailed analysis of local radiation doses for CT-guided liver biopsies and provides a benchmark for optimising radiation protection in interventional radiology.

Effectiveness of durvalumab consolidation in stage III non-small-cell lung cancer: focus on treatment selection and prognostic factors

The pivotal PACIFIC trial defined durvalumab consolidation as the new standard of care in patients with stage III non-small-cell lung cancer treated with definitive radiochemotherapy. The authors characterized the durvalumab effect after induction chemotherapy according to the ESPATUE trial and definitive radiochemotherapy. All consecutive patients with stage III non-small-cell lung cancer receiving definitive radiochemotherapy between January 2017 and February 2020 were included. Primary end points were progression-free survival and overall survival. Altogether, 160 patients (75 PD-L1-positive, 62 PD-L1-negative, 23 unknown) received definitive radiochemotherapy, 146 (91%) of whom received prior induction chemotherapy. Durvalumab consolidation showed high effectiveness overall and in the good-risk group according to the PACIFIC trial (log-rank test: p < 0.005). Hazard ratios for progression-free survival and overall survival were at the lower limits of those in the PACIFIC trial. These results were robust to adjustment for potential confounders by propensity score weighting. Eastern Cooperative Oncology Group (ECOG) performance status was the most important pretreatment prognostic factor.

Deep Learning-driven classification of external DICOM studies for PACS archiving

OBJECTIVES

Over the course of their treatment, patients often switch hospitals, requiring staff at the new hospital to import external imaging studies to their local database. In this study, the authors present MOdality Mapping and Orchestration (MOMO), a Deep Learning-based approach to automate this mapping process by combining metadata analysis and a neural network ensemble.

METHODS

A set of 11,934 imaging series with existing anatomical labels was retrieved from the PACS database of the local hospital to train an ensemble of neural networks (DenseNet-161 and ResNet-152), which process radiological images and predict the type of study they belong to. We developed an algorithm that automatically extracts relevant metadata from imaging studies, regardless of their structure, and combines it with the neural network ensemble, forming a powerful classifier. A set of 843 anonymized external studies from 321 hospitals was hand-labeled to assess performance. We tested several variations of this algorithm.

RESULTS

MOMO achieves 92.71% accuracy and 2.63% minor errors (at 99.29% predictive power) on the external study classification task, outperforming both a commercial product (82.86% accuracy, 1.36% minor errors, 96.20% predictive power) and a pure neural network ensemble (72.69% accuracy, 10.3% minor errors, 99.05% predictive power) performing the same task. We find that the highest performance is achieved by an algorithm that combines all information into one vote-based classifier.

CONCLUSION

Deep Learning combined with metadata matching is a promising and flexible approach for the automated classification of external DICOM studies for PACS archiving.

KEY POINTS

• The algorithm can successfully identify 76 medical study types across seven modalities (CT, X-ray angiography, radiographs, MRI, PET (+CT/MRI), ultrasound, and mammograms). • The algorithm outperforms a commercial product performing the same task by a significant margin (> 9% accuracy gain). • The performance of the algorithm increases through the application of Deep Learning techniques.