Johannes Boos

Dose Monitoring in Radiology Departments: Status Quo and Future Perspectives

UNLABELLED The number of computed tomography examinations has continuously increased over the last decades and accounts for a major part of the collective radiation dose from medical investigations. For purposes of quality assurance in modern radiology a systematic monitoring and analysis of dose related data from radiological examinations is mandatory. Various ways of collecting dose data are available today, for example the Digital Imaging and Communication in Medicine - Structured Report (DICOM-SR), optical character recognition and DICOM-modality performed procedure steps (MPPS). The DICOM-SR is part of the DICOM-standard and provides the DICOM-Radiation Dose Structured Report, which is an easily applicable and comprehensive solution to collect radiation dose parameters. This standard simplifies the process of data collection and enables comprehensive dose monitoring. Various commercial dose monitoring software devices with varying characteristics are available today. In this article, we discuss legal obligations, various ways to monitor dose data, current dose monitoring software solutions and future perspectives in regard to the EU Council Directive 2013/59/EURATOM. KEY POINTS • Automated, systematic dose monitoring is an important element in quality assurance of radiology departments. • DICOM-RDSR-capable CT scanners facilitate the monitoring of dose data. • A variety of commercial and non-commercial dose monitoring software tools are available today. • Successful dose monitoring requires comprehensive infrastructure for monitoring, analysing and optimizing radiation exposure. Citation Format: • Boos J, Meineke A, Bethge OT et al. Dose Monitoring in Radiology Departments: Status Quo and Future Perspectives. Fortschr Röntgenstr 2016; 188: 443 - 450.

Ovarian Cancer: Prevalence in Incidental Simple Adnexal Cysts Initially Identified in CT Examinations of the Abdomen and Pelvis

Purpose To evaluate the rate of malignancy in incidentally detected simple adnexal cysts at computed tomography (CT) to determine if simple-appearing cysts require follow-up. Materials and Methods In this HIPAA-compliant, institutional review board-approved retrospective cohort study, an institutional database was searched for abdominal and pelvic CT studies performed between June 2003 and December 2010 in women reported to have adnexal cysts. Adnexal cyst characterization was determined by prospective report description as well as image review by a research fellow and by a fellowship-trained abdominal radiologist for examinations with disagreement between the original report and the research fellows assessment. Patients with known ovarian cysts or ovarian cancer at time of the index CT examination were excluded. Clinical outcome was assessed by using follow-up imaging studies, medical records, and the state cancer registry. Benign outcome was determined by benign findings at surgery, a decrease in size or resolution of a simple-appearing cyst at follow-up imaging, or stability of the cyst for at least 1 year. Descriptive statistics and 95% confidence intervals (CIs) were calculated. Results Among 42 111 women who underwent abdominal and pelvic CT examinations in the study period, 2763 (6.6%; 95% CI: 6.3%, 6.8%) (mean age, 48.1 years ± 18.1; range, 15-102 years) had a newly detected finding of ovarian cyst described in the body or impression section of the report. Median cyst size was 3.1 cm (range, 0.8-20.0 cm). Eighteen (0.7%; 95% CI: 0.4%, 1.0%) of 2763 patients were found to have ovarian cancer after an average follow-up of 5.1 years ± 3.8 (range, 0-12.8 years). None (95% CI: 0%, 0.4%) of 1031 women with simple-appearing cysts were given a diagnosis of ovarian cancer. This included none (95% CI: 0%, 0.4%) of 904 women with simple-appearing cysts with an adequate reference standard for benign outcome. Conclusion The prevalence of previously unknown adnexal cysts at CT was 6.6%, with an ovarian cancer rate of 0.7% (95% CI: 0.4%, 1.0%). All simple-appearing cysts were benign (95% CI: 99.6%, 100%).

Split-Bolus Injection Producing Simultaneous Late Arterial and Portal Venous Phases in CT Enterography: Preliminary Results

OBJECTIVE The purpose of this article is to evaluate the image quality and added value of split-bolus contrast agent injection combining late arterial and portal venous phases compared with single-bolus contrast agent injection late arterial phase CT enterography. MATERIALS AND METHODS Consecutive patients who underwent CT enterography before and after implementation of a single-bolus CT enterography protocol were included. Attenuation and contrast-to-noise ratio (CNR) were assessed by ROI measurements of the bowel wall and arterial and venous structures. Subjective enhancement of the bowel wall (1, arterial; 2, mucosal; 3, transmural; 4, transmural with mucosal hyperenhancement) and bowel abnormalities were assessed by two independent readers. MR enterography examinations, endoscopy reports, and surgery reports within 30 days after CT enterography were used to produce a composite outcome. RESULTS Sixty-six patients were included in our study: 33 (mean [± SD] age, 46.0 ± 19.8 years) who underwent split-bolus CT enterography and 33 (mean age, 49.9 ± 19.0 years) who underwent single-bolus CT enterography. Bowel wall attenuation and CNR were higher for split-bolus CT enterography than for single-bolus CT enterography at 120 kVp (enhancement, 98.7 ± 23.1 HU vs 85.1 ± 23.3 HU; CNR, 6.4 ± 2.5 vs 4.4 ± 2.3; p < 0.01). Subjective ratings of bowel wall enhancement were higher with the split-bolus CT enterography than the single-bolus CT enterography (2.6 ± 0.8 vs 2.3 ± 0.6; p < 0.001). Split-bolus CT enterography led to a higher detection rate of mucosal hyperenhancement than did single-bolus CT enterography in patients with active inflammatory bowel disease (100.0% [7/7; 95% CI, 59.0-100.0%] vs 33.3% [2/6; 95% CI, 4.3-77.7%]; p = 0.02), whereas both protocols had a specificity of 100.0% (9/9). CONCLUSION Split-bolus CT enterography led to improved CNR (47%) compared with single-bolus CT enterography and significantly increased the detection rate of mucosal hyperenhancement in patients with active inflammatory bowel disease.

Heparin-bonded stent graft treatment for major visceral arterial injury after upper abdominal surgery

PurposeTo analyse technical success, complications, and short- and intermediate-term outcomes after heparin-bonded stent graft implantation for the treatment of major abdominal vessel injury after upper abdominal surgery.MethodsThis retrospective, IRB-approved analysis included 29 consecutive patients (female: n = 6, male: n = 23, mean age 65.9 ± 11.2 years). All patients underwent angiography and attempted heparin-bonded stent-graft implantation because of a major visceral arterial injury after upper abdominal surgery. Electronic clinical records, angiographic reports and imaging datasets were reviewed to assess technical success and complications. Telephone interviews were performed to obtain follow-up information and to estimate short- (> 30 days) and intermediate-term (> 90 days) outcomes.ResultsSuccessful stent graft placement was achieved in 82.8% (24/29). Peri-interventional complications were observed in 20.7% (6/29) and delayed, angiography-associated complications were observed in 34.5% (10/29) of the patients. Symptomatic re-bleeding occurred in 24.1% (7/29). Short-term survival (> 30 days) was 72.4% (21/29). Intermediate survival (> 90 days) was 37.9% (11/29).ConclusionTreatment of major vascular injuries with heparin-bonded stent grafts is feasible with a high technical success rate. However, survival depends on the underlying surgical condition, making interdisciplinary patient management mandatory.Key Points• Stent graft implantation is challenging, but has a high technical success rate.• Complications are frequent but surgical conversion is rarely necessary.• Survival depends on the underlying surgical condition causing the vascular injury.• Interdisciplinary management is crucial for the survival of these patients.

Performance and clinical impact of machine learning based lung nodule detection using vessel suppression in melanoma patients

PURPOSE To evaluate performance and the clinical impact of a novel machine learning based vessel-suppressing computer-aided detection (CAD) software in chest computed tomography (CT) of patients with malignant melanoma. MATERIALS AND METHODS We retrospectively included consecutive malignant melanoma patients with a chest CT between 01/2015 and 01/2016. Machine learning based CAD software was used to reconstruct additional vessel-suppressed axial images. Three radiologists independently reviewed a maximum of 15 lung nodules per patient. Vessel-suppressed reconstructions were reviewed independently and results were compared. Follow-up CT examinations and clinical follow-up were used to assess the outcome. Impact of additional nodules on clinical management was assessed. RESULTS In 46 patients, vessel-suppressed axial images led to the detection of additional nodules in 25/46 (54.3%) patients. CT or clinical follow up was available in 25/25 (100%) patients with additionally detected nodules. 2/25 (8%) of these patients developed new pulmonary metastases. None of the additionally detected nodules were found to be metastases. None of the lung nodules detected by the radiologists was missed by the CAD software. The mean diameter of the 92 additional nodules was 1.5 ± 0.8 mm. The additional nodules did not affect therapeutic management. However, in 14/46 (30.4%) of patients the additional nodules might have had an impact on the radiological follow-up recommendations. CONCLUSION Machine learning based vessel suppression led to the detection of significantly more lung nodules in melanoma patients. Radiological follow-up recommendations were altered in 30% of the patients. However, all lung nodules turned out to be non-malignant on follow-up.

What Is the Optimal Abdominal Aortic Aneurysm Sac Measurement on CT Images during Follow-up after Endovascular Repair?

Purpose To develop a computed tomographic (CT) angiographic postprocessing protocol with two- and three-dimensional measurements for follow-up of patients who underwent endovascular aortic repair. Materials and Methods This HIPAA-compliant institutional review board-approved retrospective study included 159 patients (129 men, 30 women; mean age ± standard deviation, 74.9 years ± 8.2) who underwent 824 CT examinations (median of five examinations per patient; range, two to 14) with unenhanced and arterial -phase imaging performed between September 2004 and March 2015. The largest diameter on the axial plane; coronal, sagittal, and maximal diameter perpendicular to the reconstructed centerline; volume of the abdominal aortic aneurysm sac; and volume from the lowest renal artery to the aortic bifurcation and to the common iliac artery bifurcation were measured. Endoleaks on contrast material-enhanced images were considered the reference standard, and the predictive value of diameter and volume changes was analyzed. Intraclass correlation was used to compare diameters and volumes. Results All diameters and volumes showed excellent correlation (intraclass coefficient, 0.95 and 0.94, respectively). Average interobserver difference for diameters and volumes was 2%-3% and 4%-12%, respectively. Endoleaks were observed in 80 (50%) of 159 patients (59 [74%] at initial and 21 [26%] at later CT angiography). New endo-leaks were associated with increased aneurysm size measured as the largest diameter on the axial plane (P = .04) and perpendicular to the centerline (P = .01), and volume was measured from the lowest renal artery to the aortic bifurcation (P = .03) and to the common iliac artery bifurcation (P = .01). With a 5% size threshold, sensitivity and specificity for detection of endoleaks was optimal for centerline diameter (64.3% and 81.7%, respectively) and volume from the lowest renal artery to the common iliac artery bifurcation (57.1% and 63.5%). Conclusion The maximal diameter and volume of an abdominal aortic aneurysm sac can be used for temporal monitoring after endovascular aortic repair, with excellent correlation and interobserver agreement. An increase in the centerline diameter and volume from the lowest renal artery to the iliac bifurcation were the most sensitive criteria for detecting endoleaks.

CT Intensity Distribution Curve (Histogram) Analysis of Patients Undergoing Antiangiogenic Therapy for Metastatic Renal Cell Carcinoma

OBJECTIVE The purpose of this study was to evaluate the median versus mean attenuation obtained from histograms in the response assessment of renal cell carcinoma (RCC) target lesions in patients treated with vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitors. MATERIALS AND METHODS Imaging findings of 19 patients with RCC and treated with a VEGFR tyrosine kinase inhibitor that included a baseline study and at least two follow-up studies were retrospectively analyzed. Histograms of target lesions were created from ROI measurements, and mean and median lesion attenuation were obtained. Shift of mean and median was used to classify response of lesions between baseline CT and the first follow-up examination according to the Choi criteria; morphology, attenuation, size, and structure (MASS) criteria; and the Response Evaluation Criteria in Solid Tumors (RECIST). RECIST classification of the last follow-up CT examination was considered the reference standard. RESULTS Among the 19 patients (six women, 13 men; mean age, 60 ± 8 [SD] years; body mass index, 24.5 ± 4.6 kg/m2), three (16%) had evidence of disease progression; nine (47%), partial response; and seven (37%), stable disease. Temporal change in the shape of the histograms from or to gaussian occurred in 11 of the 19 (58%) lesions, which accounted for a significant difference between mean (64.3 ± 38.5 HU) and median (71.0 ± 36.5 HU) lesion attenuation (p < 0.001). According to the shift of median for outcome analysis, the number of properly classified lesions increased from 12 of 19 (63%) to 13 of 19 (68%) for the Choi criteria and from 14 of 19 (74%) to 15 of 19 (79%) for the MASS criteria. CONCLUSION Mean and median attenuation of RCC target lesions differed significantly. Use of shift of median rather than shift of mean increased the number of accurate classifications of lesions at the first follow-up CT examination.