Reduction of missed thoracic findings in emergency whole-body computed tomography using artificial intelligence assistance.

BACKGROUND: Radiology reporting of emergency whole-body computed tomography (CT) scans is time-critical and therefore involves a significant risk of pathology under-detection. We hypothesize a relevant number of initially missed secondary thoracic findings that would have been detected by an artificial intelligence (AI) software platform including several pathology-specific AI algorithms. METHODS: This retrospective proof-of-concept-study consecutively included 105 shock-room whole-body CT scans. Image data was analyzed by platform-bundled AI-algorithms, findings were reviewed by radiology experts and compared with the original radiologist's reports. We focused on secondary thoracic findings, such as cardiomegaly, coronary artery plaques, lung lesions, aortic aneurysms and vertebral fractures. RESULTS: We identified a relevant number of initially missed findings, with their quantification based on 105 analyzed CT scans as follows: up to 25 patients (23.8%) with cardiomegaly or borderline heart size, 17 patients (16.2%) with coronary plaques, 34 patients (32.4%) with aortic ectasia, 2 patients (1.9%) with lung lesions classified as "recommended to control" and 13 initially missed vertebral fractures (two with an acute traumatic origin). A high number of false positive or non-relevant AI-based findings remain problematic especially regarding lung lesions and vertebral fractures. CONCLUSIONS: We consider AI to be a promising approach to reduce the number of missed findings in clinical settings with a necessary time-critical radiological reporting. Nevertheless, algorithm improvement is necessary focusing on a reduction of "false positive" findings and on algorithm features assessing the finding relevance, e.g., fracture age or lung lesion malignancy.

Dynamic CTA in Native Kidneys Using a Multiphase CT Protocol-Potential of Significant Reduction of Contrast Medium.

RATIONALE AND OBJECTIVE: The objective of this study was to assess an optimized renal multiphase computed tomography angiography (MP-CTA) protocol regarding reduction of contrast volume. MATERIALS AND METHODS: Thirty patients underwent MP-CTA (12 phases, every 3.5 seconds, 80 kV/120 mAs) using 30 mL of contrast medium. The quality of MP-CTA was assessed quantitatively measuring vessel attenuation, image noise, and contrast-to-noise ratio. MP-CTA was evaluated qualitatively regarding depiction of vessels, cortex differentiation, and motion artifacts (grades 1-4, 1 = best). Mean effective radiation dose was registered. Results were compared to standard renal computed tomography angiography (CTA) (80 mL). Student t test was applied, if variables followed normal distribution. For other variables, nonparametric Mann-Whitney U test was used. RESULTS: All acquisitions were successfully performed, and no patient had to be excluded from the study. MP-CTA enabled high attenuation (aorta: 503 ± 91 HU, renal arteries: 450 ± 73 HU/456 ± 72 HU) at adequate image noise (13.7 ± 1.5) and good contrast-to-noise ratio (34.2 ± 10.2). Good attenuation of renal veins was observed (286 ± 43 HU/282 ± 42 HU). Arterial enhancement was significantly higher compared to renal CTA (aorta: 396 ± 90 HU, renal arteries: 331 ± 74 HU/333 ± 80 HU; P < .001). MP-CTA protocol enabled good image quality of renal arteries (1.5 ± 0.6) and veins (1.7 ± 0.6). Cortex differentiation and motion artifacts were ranked 1.8 ± 0.8 and 1.6 ± 0.8. The mean effective radiation dose was 9 mSv (MP-CTA). CONCLUSIONS: Compared to standard renal CTA, the renal MP-CTA enabled the significant reduction of contrast volume and simultaneously provided a significantly higher arterial attenuation.

Intracranial arterial calcifications as a prognostic factor for subsequent major adverse cardiovascular events (MACE).

BACKGROUND: Intracranial arterial calcifications (ICAC) are often detected on unenhanced CT of patients with an age > 60. However, association with the subsequent occurrence of major adverse cardiovascular events (MACE) has not yet been evaluated. PURPOSE: This study aimed at evaluating the association of ICAC with subsequent MACE and overall mortality. METHODS: In this retrospective, IRB approved study, we included 175 consecutive patients (89 males, mean age 78.3 ± 8.5 years) of age > 60 years who underwent an unenhanced CT of the head due to minor trauma or neurological disorders. Presence of ICAC was determined in seven intracranial arteries using a semi-quantitative scale, which resulted in the calcified plaque score (CPS). Clinical follow-up information was obtained by questionnaires and telephone interviews. MACE was defined as myocardial infarction or revascularization, stroke or death due to cardiovascular event. RESULTS: Mean follow-up time was 39.8 ± 7.8 months, resulting in 579.7 patient-years of follow-up. Overall, 36 MACE occurred during follow-up (annual event rate = 6.2%/year). Mean CPS was significantly higher in subjects with MACE during follow-up compared to subjects without MACE (p < 0.01). In 15 patients CPS was 0; in none of these patients MACE was registered. Kaplan-Meier-analysis revealed that patients with a low plaque burden (CPS < 5) had a significant longer MACE-free and overall survival than patients with a high plaque burden (CPS ≥ 5) (p < 0.01). CONCLUSION: Patients with ICAC have an increased risk for future cardio- or cerebrovascular events. Therefore, ICAC might be a prognostic factor to determine the risk for these events in older patients.

Prediction of 90Y Radioembolization Outcome from Pretherapeutic Factors with Random Survival Forests.

Our objective was to predict the outcome of 90Y radioembolization in patients with intrahepatic tumors from pretherapeutic baseline parameters and to identify predictive variables using a machine-learning approach based on random survival forests. Methods: In this retrospective study, 366 patients with primary (n = 92) or secondary (n = 274) liver tumors who had received 90Y radioembolization were analyzed. A random survival forest was trained to predict individual risk from baseline values of cholinesterase, bilirubin, type of primary tumor, age at radioembolization, hepatic tumor burden, presence of extrahepatic disease, and sex. The predictive importance of each baseline parameter was determined using the minimal-depth concept, and the partial dependency of predicted risk on the continuous variables bilirubin level and cholinesterase level was determined. Results: Median overall survival was 11.4 mo (95% confidence interval, 9.7-14.2 mo), with 228 deaths occurring during the observation period. The random-survival-forest analysis identified baseline cholinesterase and bilirubin as the most important variables (forest-averaged lowest minimal depth, 1.2 and 1.5, respectively), followed by the type of primary tumor (1.7), age (2.4), tumor burden (2.8), and presence of extrahepatic disease (3.5). Sex had the highest forest-averaged minimal depth (5.5), indicating little predictive value. Baseline bilirubin levels above 1.5 mg/dL were associated with a steep increase in predicted mortality. Similarly, cholinesterase levels below 7.5 U predicted a strong increase in mortality. The trained random survival forest achieved a concordance index of 0.657, with an SE of 0.02, comparable to the concordance index of 0.652 and SE of 0.02 for a previously published Cox proportional hazards model. Conclusion: Random survival forests are a simple and straightforward machine-learning approach for prediction of overall survival. The predictive performance of the trained model was similar to a previously published Cox regression model. The model has revealed a strong predictive value for baseline cholinesterase and bilirubin levels with a highly nonlinear influence for each parameter.

Reproducibility and differentiation of cervical arteriopathies using in vivo high-resolution black-blood MRI at 3 T.

INTRODUCTION: The aims of the present study are to evaluate the potential of high-resolution black-blood MRI (hr-bb-cMRI) to differentiate common cervical arteriopathies and to evaluate interobserver reproducibility. METHODS: Forty-three consecutive patients with distinct cervical arteriopathies were examined with cervical hr-bb-cMRI at 3.0 Te with fat-saturated pre- and post-contrast T1w, T2w, and TOF images using dedicated carotid surface coils at our institution. Twenty-three patients had atherosclerotic disease, causing significant stenosis in 12 patients while 11 patients had moderate stenosis. Eight patients presented with cervical vasculitis, and five patients had arterial dissection. Furthermore, seven control subjects with no evidence of carotid disease were included. Two experienced readers blinded to all clinical information reviewed all MR images and classified both carotid and vertebral arteries as affected either by atherosclerosis, dissection, vasculitis, or no disease. Finally, a consensus reading was performed. RESULTS: On a per-vessel level, test performance parameters (sensitivity, specificity, positive predictive value, negative predictive value) were 95, 97.7, 92.9, and 98.5 % for atherosclerotic disease; 91, 100, 100, and 98.7 % for vasculitis; and 100, 100, 100, and 100 % for dissection, respectively. On a per-patient level, performance parameters were 95.7, 85.7, 97.2, and 85.7 % for the diagnosis of atherosclerosis and 100, 100, 100, and 100 % for the diagnosis of dissection and of vasculitis, respectively. Accuracy rates were all above 95 % for all entities. There was a high agreement between observers both in a per-vessel (κ = 0.83) and in a per-patient analysis (κ = 0.82). CONCLUSION: This study demonstrates that hr-bb-cMRI is able to non-invasively differentiate between the most common cervical arteriopathies with an excellent interreader reproducibility.

CT fluoroscopy-guided closed-tip catheter placement before regional hyperthermia treatment of soft tissue sarcomas: 5-Year experience in 35 consecutive patients.

PURPOSE: This study was designed to assess technical success and complications in patients with high-risk soft tissue sarcomas undergoing CT fluoroscopy-guided closed-tip catheter placement before treatment with combined chemotherapy and regional hyperthermia. MATERIALS AND METHODS: This retrospective study comprised all patients referred for insertion of closed-tip catheters for the introduction of thermometry probes before regional hyperthermia treatment at a single university centre from 2010 to 2015. Catheter placements were performed under local anaesthesia and intermittent CT fluoroscopy guidance. Technical success, complication rate, duration of catheter insertion and dose-length product (DLP) were analysed. Technical success was defined as intratumoural catheter placement suitable for subsequent thermometry. RESULTS: A total of 35 procedures were performed on 35 patients (22 men, 13 women). In 34 out of 35 interventions catheters were inserted successfully; in one patient catheter placement was not feasible. No intra-interventional complications occurred. In six patients post-interventional complications were observed - two major (one abscess formation and one severe catheter dislocation) and four minor complications. Technical failure was observed in 11.4% of patients, especially catheter kinking. A total of 55 catheters were placed, with a mean number of 1.7 ± 0.7 per patient. Mean total DLP was 723.2 ± 355.9 mGy*cm. CONCLUSION: CT fluoroscopy-guided closed-tip catheter placement into high-risk soft tissue sarcomas was characterised by high technical success and relatively low complication rate. While major complications were rarely observed, catheter-kinking preventing successful thermometry represented the most frequent technical failure.

Diffusion-weighted MRI Before and After Robotic Radiosurgery (Cyberknife®) in Primary and Secondary Liver Malignancies: A Pilot Study.

The purpose of this study was to evaluate the role of diffusion-weighted MR imaging (DW-MRI) in the assessment of treatment response of primary or secondary liver malignancies after stereotactic radiosurgery (SRS) using robotic radiosurgery. All patients who underwent SRS therapy for hepatic malignancies who had pre- and post-interventional MR examinations including DW-MRI at our hospital between 02/2010 and 02/2012 were included. A retrospective analysis of the institutional imaging database identified 13 patients (4 men, 9 women, mean age: 66 years) with a total of 14 primary or secondary liver malignancies. Criteria of tumor response to treatment were a decrease in size and/or loss of contrast enhancement as assessed with respect to RECIST criteria. Mean apparent diffusion coefficient (ADC) values for normal liver parenchyma and hepatic masses in each MR examination were calculated and compared using two-tailed, paired t-test with a significance level of 0.05. Mean ADC values of liver malignancies were 1.10 ± 0.30 x 10(-3) mm(2)/s, 1.48 ± 0.35 x 10(-3) mm(2)/s and 1.56 ± 0.40 x 10(-3) mm(2)/s on pre-interventional, the first post-interventional, and the second post-interventional DW-MRI. There was a significant increase of ADC values between the pre-interventional examination and the first and second post-interventional follow-up exams (p < 0.01 and p = 0.01, respectively). However, there was no statistical difference between the ADC values of the first and second post-interventional MRI. ADC values of normal liver parenchyma did not show significant changes in ADC values during pre- and post-interventional MRI. ADC measurements may allow the evaluation of response to treatment of hepatic malignancies by SRS. ADC values of liver malignancies increase significantly after SRS treatment.

Parallel-transmit-accelerated spatially-selective excitation MRI for reduced-FOV diffusion-weighted-imaging of the pancreas.

OBJECTIVES: To find out whether the use of accelerated 2D-selective parallel-transmit excitation MRI for diffusion-weighted EPI (pTX-EPI) offers advantages over conventional single-shot EPI (c-EPI) with respect to different aspects of image quality in the MRI of the pancreas. MATERIALS AND METHODS: The MRI examinations of 33 consecutive patients were evaluated in this prospective and IRB-approved study. PTX-EPI was performed with a reduced (zoomed) FOV of 230 × 118 mm(2). The 2D-RF pulse of pTX-EPI was accelerated, i.e. shortened by a factor of 1.7 (pTX-acceleration factor). C-EPI used a full-FOV of 380 × 285 mm(2). In a qualitative analysis, two experienced readers evaluated 3 different aspects of image quality on 3- to 5-point Likert scales. Additionally, apparent diffusion coefficients (ADCs) were determined in both c-EPI and pTX-EPI in normal-appearing pancreatic tissue using regions of interests (ROIs). Mean ADC values and standard deviations were compared between the two techniques. RESULTS: The reduced-FOV pTX-EPI was superior to c-EPI with respect to overall image quality (p<0.0001) and identifiability of the pancreatic ducts (p<0.01). Artifacts were significantly less severe in pTX-EPI (p<0.01). The mean ADC values of c-EPI (1.29 ± 0.19 × 10(-3)mm(2)/s) and pTX-EPI (1.27 ± 0.17 × 10(-3)mm(2)/s) did not differ significantly between the two techniques (p=0.44). The variation within the ROIs as measured by the standard deviation was significantly lower in pTX-EPI (0.095 × 10(-3)mm(2)/s) than in c-EPI (0.135 × 10(-3)mm(2)/s), p<0.05. CONCLUSIONS: PTX-accelerated EPI with spatially-selective excitation and reduced FOV leads to substantial improvements in DWI of the pancreas with respect to different aspects of image quality without significantly influencing the ADC values.

Technical and clinical outcome of percutaneous CT fluoroscopy-guided screw placement in unstable injuries of the posterior pelvic ring.

PURPOSE: To evaluate technical success, complications, and effective dose in patients undergoing CT fluoroscopy-guided iliosacral screw placement for the fixation of unstable posterior pelvic ring injuries. MATERIALS AND METHODS: Our retrospective analysis includes all consecutive patients with vertical sacral fractures and/or injury of the iliosacral joint treated with CT fluoroscopy-guided screw placement in our department from 11/2005 to 03/2013. Interventions were carried out under general anesthesia and CT fluoroscopy (10-20 mAs; 120 kV; 16- or 128-row scanner, Siemens Healthcare, Erlangen, Germany). Technical outcome, major and minor complications, and effective patient dose were analyzed. RESULTS: We treated 99 consecutive patients (mean age 53.1 ± 21.7 years, 50 male, 49 female) with posterior pelvic ring instability with CT fluoroscopy-guided screw placement. Intervention was technically successful in all patients (n = 99). No major and one minor local complication occurred (1 %, secondary screw dislocation). General complications included three cases of death (3 %) due to pulmonary embolism (n = 1), hemorrhagic shock (n = 1), or cardiac event (n = 1) during a follow-up period of 30 days. General complications were not related to the intervention. Mean effective patient radiation dose per intervention was 12.28 mSv ± 7.25 mSv. Mean procedural time was 72.1 ± 37.4 min. CONCLUSIONS: CT fluoroscopy-guided screw placement for the treatment of posterior pelvic ring instabilities can be performed with high technical success and a low complication rate. This method provides excellent intrainterventional visualization of iliac and sacral bones, as well as the sacral neuroforamina for precise screw placement by applying an acceptable effective patient dose.

Twelve-month results of a randomized trial comparing mono with dual antiplatelet therapy in endovascularly treated patients with peripheral artery disease.

PURPOSE: To present the 12-month results of a trial investigating the effects of dual antiplatelet therapy on target lesion revascularization (TLR) after balloon angioplasty ± stenting in the femoropopliteal segment. METHODS: A prospective, randomized, single-center, double-blinded and placebo-controlled clinical trial randomly assigned 40 patients to receive pre- and postinterventional therapy with aspirin and clopidogrel. Another 40 patients received the same doses of aspirin and placebo instead of clopidogrel. Clopidogrel and placebo were stopped after 6 months, and patients remained on aspirin only. At 12 months after the intervention, 36 clopidogrel patients and 37 placebo patients were reevaluated. RESULTS: At 6 months, clopidogrel patients had significantly lower rates of TLR compared to placebo patients [2 (5%) vs. 8 (20%), p=0.04]. After stopping clopidogrel/placebo after 6 months, there was no significant difference in TLR at 12 months after treatment [9 (25%) clopidogrel vs. 12 (32.4%) placebo, p=0.35]. Mortality was 0 vs. 1 in the placebo group at 6 months (p=0.32) and 0 vs. 3 at 12 months (p=0.08). CONCLUSION: In contrast to the first report of a reduction in the TLR at 6 months, this advantage of dual antiplatelet therapy does not persist after stopping clopidogrel. Prolonged dual therapy (>6 months) should be considered in patients who are at high risk for restenosis.

Impact of cardiovascular risk factors on vessel wall inflammation and calcified plaque burden differs across vascular beds: a PET-CT study.

To evaluate the effect of age, gender and cardiovascular risk factors on vessel wall inflammation and the calcified plaque burden in different vascular beds as assessed by PET/CT. 315 patients (mean age: 57.8 years, 123 male and 192 female) who underwent whole body 18F-FDG PET/CT examinations were included in the study. Blood pool-corrected standardised uptake value (TBR) and the calcified plaque score (CPS, grade 0-4) were determined in the thoracic and abdominal aorta, both common carotid and both iliac arteries. The following cardiovascular risk factors were documented: Age ≥65 years (n = 114), male gender (n = 123), diabetes (n = 15), hyperlipidemia (n = 62), hypertension (n = 76), body mass index (BMI) ≥ 30 (n = 38), current smoker (n = 32). Effects of risk factors on TBR and CPS in different arterial beds were assessed using multivariate regression analysis. In the thoracic aorta TBR was independently associated with age ≥65 years and male gender, CPS was independently associated with age ≥65 years, male gender, hypertension and diabetes. In the abdominal aorta, TBR was independently associated with age ≥65 years and male gender, CPS with age ≥65 years, diabetes and smoking. Independent associations in the carotid arteries were found for age ≥65 years, male gender and BMI ≥ 30 in TBR and for age ≥65 and diabetes in CPS. In the iliac arteries, TBR was independently associated with age ≥65 and CPS with age ≥65, male gender, hypertension, diabetes and smoking. Findings of this PET/CT study demonstrate that the impact of cardiovascular risk factors on vessel wall inflammation and calcified plaque burden differs across vascular territories. Overall, CPS was more closely associated with cardiovascular risk factors compared to TBR.

Whole-brain CT perfusion: reliability and reproducibility of volumetric perfusion deficit assessment in patients with acute ischemic stroke.

INTRODUCTION: The aim of this study was to examine reliability and reproducibility of volumetric perfusion deficit assessment in patients with acute ischemic stroke who underwent recently introduced whole-brain CT perfusion (WB-CTP). METHODS: Twenty-five consecutive patients underwent 128-row WB-CTP with extended scan coverage of 100 mm in the z-axis using adaptive spiral scanning technique. Volumetric analysis of cerebral blood volume (CBV), cerebral blood flow (CBF), mean transit time (MTT), time to peak (TTP), and time to drain (TTD) was performed twice by two blinded and experienced readers using OsiriX V.4.0 imaging software. Interreader agreement and intrareader agreement were assessed by intraclass correlation coefficients (ICCs) and Bland-Altman Analysis. RESULTS: Interreader agreement was highest for TTD (ICC 0.982), followed by MTT (0.976), CBF (0.955), CBV (0.933), and TTP (0.865). Intrareader agreement was also highest for TTD (ICC 0.993), followed by MTT (0.988), CBF (0.981), CBV (9.953), and TTP (0.927). The perfusion deficits showed the highest absolute volumes in the time-related parametric maps TTD (mean volume 121.4 ml), TTP (120.0 ml), and MTT (112.6 ml) and did not differ significantly within this group (each with p > 0.05). In comparison to time-related maps, the mean CBF perfusion deficit volume was significantly smaller (92.1 ml, each with p < 0.05). The mean CBV lesion size was 23.4 ml. CONCLUSIONS: Volumetric assessment in WB-CTP is reliable and reproducible. It might serve for a more accurate assessment of stroke outcome prognosis and definition of flow-volume mismatch. Time to drain showed the highest agreement and therefore might be an interesting parameter to define tissue at risk.

Management of peripheral arterial interventions with mono or dual antiplatelet therapy--the MIRROR study: a randomised and double-blinded clinical trial.

OBJECTIVES: To investigate the influence of dual antiplatelet therapy vs. aspirin alone on local platelet activation and clinical endpoints in patients with PAD treated with endovascular therapy. METHODS: Patients received either 500 mg aspirin and 300 mg clopidogrel before intervention followed by a daily dose of 100 mg aspirin and 75 mg clopidogrel for 6 months, or the same doses of aspirin plus placebo instead of clopidogrel. Primary endpoints were local concentrations of platelet activation markers ß-thromboglobulin and CD40L, and the rate of patient's resistant to clopidogrel. Secondary endpoints included the clinical development 6 months after the intervention. RESULTS: Eighty patients, 40 in each group, were enrolled. The median peri-interventional concentration of ß-TG was 224.5 vs. 365.5 (P = 0.03) in the clopidogrel and placebo group. The concentration of CD40L was 127 and 206.5 (P = 0.05). Thirty per cent of patients who had received clopidogrel were resistant. Two clopidogrel and eight placebo patients required TLR (P = 0.04). The clopidogrel patients who needed revascularisation were both resistant to clopidogrel. Minor bleeding complications occurred in one clopidogrel and two placebo patients. CONCLUSION: Dual antiplatetet therapy reduces peri-interventional platelet activation and improves functional outcome without higher bleeding complications. An individual tailored dual antiplatelet therapy seems desirable for endovascularly treated patients with PAD.