Effect of Noise Adaptive Wavelet Filter on Diagnostic Performance in Stroke Perfusion

Background: Computed tomography perfusion (CTP) images include more noise than routine clinic computed tomography (CT) images. Singular value decomposition based deconvolution algorithms are widely used for obtaining several functional perfusion maps. Recently block circulant singular value decomposition algorithms become popular for its superior property of immunity to contrast bolus lag. It is well known from literature that these algorithms are very sensitive to noise. There are a lot of examples of noise reduction filters in the literature as well as commercial ones. Functional maps which help physicians in the diagnostic process can be obtained with better image quality by de-noising CTP images with adaptive noise reduction filters. Objective: In this study, the effect of a noise adaptive wavelet filtering method on diagnostic performance on CTP stroke patient images is investigated. Method: Images of acute stroke patients were de-noised by this method and their diagnostic value were evaluated by visual means, peak signal-to-noise ratio and time intensity profile metrics. An observer evaluation study was carried out in order to validate quantitative image quality metrics. The results are compared with Gaussian and a bilateral filter based filtering method called TIPS (Time Intensity Profile Similarity) on same images sets to benchmark proposed method. Results: The diagnostic value of the images obtained from noise adaptive wavelet filtering method were better than Gaussian filter method and were compatible with a wellknown time intensity profile similarity bilateral filter method. Diagnostic performance of the both observers were improved compared to both Gaussian and TIPS methods. Conclusion: The noise adaptive wavelet filter method succeeded to reduce noise while preserving details contained in the contrast bolus. Its final effect on the timeintensity profiles and generated perfusion maps are compatible with the literature and showed improvements on diagnostic performance on specificity and overall accuracy when compared to other methods.

CT angiogram findings in carotid-cavernous fistulas: stratification of imaging features to help radiologists avoid misdiagnosis

Background Carotid-cavernous fistulas (CCFs) are commonly misdiagnosed on computed tomography angiography (CTA). Purpose This study sought to identify the most sensitive and specific imaging features of CCFs on CTA. Material and Methods A retrospective review identified 18 consecutive patients suspected of having a CCF on CTA and subsequently underwent digital subtraction angiography (DSA). Two blinded reviewers assessed multiple findings on CTA: cavernous sinus asymmetry/enlargement; arterial-phase contrast in the cavernous sinus; proptosis; pre- or post-septal orbital edema; and dilated regional vasculature. Each was graded as positive, possible, and negative; “possible” was counted as positive. A third blinded reviewer served as a tiebreaker. Results Of 18 patients, nine were true-positive and nine were false-positive. Superior ophthalmic vein early enhancement and dilatation had 100.0% sensitivity (95% confidence interval [CI] 40.0–100.0) and 77.8% specificity (95% CI 44.4–100.0); arterial-phase contrast in the cavernous sinus had 88.9% sensitivity (95% CI 44.4–100.0) and 66.7% specificity (95% CI 18.5–90.1); peri-orbital edema had 88.9% sensitivity (95% CI 35.5–100.0) and 77.8% specificity (95% CI 22.2–100.0). The most specific markers of CCF were superior petrosal sinus and inferior ophthalmic vein dilatation/enhancement (100.0%, 95% CI 88.8–100.0 and 88.9%, 95% CI 44.4–100.0, respectively); the specificity of asymmetric cavernous enlargement was 44.4% (95% CI 11.1–77.7). Conclusions Among patients in whom a CCF is suspected on CTA, superior ophthalmic vein dilatation/enhancement and arterial-phase contrast within the cavernous sinus are the most sensitive findings. Asymmetric cavernous sinus enlargement has poor specificity and may result in false-positive diagnoses of CCFs. False positive cases were less likely to have an optimally timed contrast bolus.

Feasibility of a Single Contrast Bolus High-Pitch Pulmonary CT Angiography Protocol Followed by Low-Dose Retrospectively ECG-Gated Cardiac CT in Patients with Suspected Pulmonary Embolism

Introduction To prospectively evaluate the feasibility of single contrast bolus high-pitch CT pulmonary angiography (CTPA) subsequently followed by low-dose retrospectively ECG-gated cardiac CT (4D-cCT) in patients with suspected pulmonary embolism (PE) to accurately evaluate right ventricular (RV) function. Materials and Methods 62 patients (33 female, age 65.1 ± 17.5 years) underwent high-pitch CTPA examination with 80cc of iodinated contrast material. 5 s after the end of the high-pitch CTPA study, a low-dose retrospectively ECG-gated cardiac CT examination was automatically started. The volume CT dose index (CTDI vol) and dose length product (DLP) were recorded in all patients and the effective dose was calculated. For the assessment of image quality, attenuation was measured as Hounsfield units (HUs) within various regions of interest (ROIs). These ROIs were used to calculate the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). Subjective image quality was assessed using a five-point Likert scale. On 4D-cCT, the ejection fraction of both ventricles (RVEF, LVEF) as well as the ratio of RVEF and LVEF (RVEF/LVEF) was assessed. The statistical difference of all parameters between the PE and non-PE group was calculated. Results The mean effective radiation dose was 4.22 ± 2.05 mSv. Attenuation measurements on CTPA showed the highest attenuation values in the main pulmonary artery (442.01 ± 187.64). On 4D-cCT attenuation values were highest in the descending aorta (560.59 ± 208.81). The CNR and SNR values on CTPA were highest within the main pulmonary artery (CNR = 12.43 ± 4.57; SNR = 15.14 ± 4.90). On 4D-cCT images, the highest SNR and CNR could be measured in the descending aorta (CNR = 10.26 ± 5.57; SNR = 10.86 ± 5.17). The mean LVEF was 60.73 %± 14.65 %, and the mean RVEF was 44.90 %± 9.54 %. The mean RVEF/LVEF was 0.79 ± 0.29. There was no significant difference between the PE and non-PE group for either of the parameters. Conclusion The investigated combined CTPA and 4D-cCT protocol is feasible using a single contrast bolus and allows the evaluation of RV function in patients with suspected PE. Further studies have to evaluate the additional value of this protocol regarding risk stratification in patients with PE. Key Points: Citation Format

Triple-split-bolus versus single-bolus CT in abdominal trauma patients: a comparative study

Background Split-bolus computed tomography (CT) is a recent development in trauma imaging. Instead of multiple scans in different contrast phases after a single contrast bolus, split-bolus protocols consist of one single scan of the thorax and abdomen after two or three contrast injections at different points of time. Purpose To evaluate and compare image quality and injury findings of a new triple-split-bolus CT (TS-CT) protocol of thorax and abdomen with those of a portal venous phase CT (PV-CT) in the same patient group. Material and Methods Trauma patients in 2009–2012 who underwent both the TS-CT initially and a PV-CT during the next six weeks were included. The TS-CT examination was performed as one CT run after application of three contrast boluses (total 175 mL) to enhance renal pelvis and urinary tract, the abdominal organs, and the large arterial vessels. The PV-CT had a fixed delay of 85 s. We measured attenuation in Hounsfield units (HU), evaluated possible organ injury and assessed image quality on a 5-point scale. Results Thirty-five patients were included. Attenuation measurements of major abdominal vessels, organs, and renal pelvis were significantly higher with the TS-CT protocol. Performance in organ injury diagnosis and image quality was equal in both protocols. Conclusion The overall performance of the TS-CT protocol is similar to the standard PV-CT. Excellent visualization of the arterial tree and the collecting system may eliminate the need for separate scans.

Abstract 60: Multiphase versus Single Phase CT Angiography for Detection of Distal Intracranial Occlusion

Introduction: Detecting distal intracranial arterial occlusion is of increasing importance as current studies begin to evaluate the potential benefit of tPA in minor acute stroke. Multiphase CTA (mCTA) is a technique used routinely at our centre in the setting of suspected acute ischemic stroke (AIS). Hypothesis: We hypothesized that mCTA is diagnostically superior to CTA for detection of distal intracranial arterial occlusion in suspected AIS. Methods: Population comprised of patients with suspected AIS who had mCTA, followed by MR brain with DWI within 24 hours. mCTA involves 3 phases of enhanced CT from skull base to vertex following administration of a single iodinated contrast bolus. Prospective interpretation of unenhanced CT followed by single phase CTA was performed by a radiology resident, a stroke fellow, or a neuroradiologist. Unenhanced CT followed by mCTA was interpreted > 1 week after single phase CTA interpretation using the same dataset. MR exams were separately adjudicated to establish each case as positive or negative according to pre-specified criteria that suggests a high likelihood for the presence of an arterial occlusion on CTA (DWI lesion > 2mm diameter, not multiple small infarcts consistent with microembolic shower, not exclusively white matter lesion, not small deep gray matter lesion consistent with lacunar type infarct. Results and Conclusion: Of 322 patients, single phase CTA detected 69 occlusions (sensitivity 52.5%, specificity 98.1%) and mCTA detected 113 (sens = 81.8%, spec = 99.5%). mCTA detected a greater number of proximal occlusions (76 vs. 66), and a greater number of distal occlusions (37 vs. 3) than CTA. mCTA demonstrated a greater rate of detection for distal occlusions (relative rate = 9.67, 95% CI 3.76 - 24.84). CTA/MR agreement was moderate (kappa = 0.59) while mCTA/MR agreement was substantial (kappa = 0.81). Mean interpretation time for CTA was 239.8 seconds (95% CI 221.3 - 258.4) and 195.0 seconds (95% CI 176.9 - 213.1) for mCTA. In conclusion, our data suggest mCTA is superior to single phase for detection of distal intracranial occlusion.