Uncertainty-aware deep learning for trustworthy prediction of long-term outcome after endovascular thrombectomy.

Acute ischemic stroke (AIS) is a leading global cause of mortality and morbidity. Improving long-term outcome predictions after thrombectomy can enhance treatment quality by supporting clinical decision-making. With the advent of interpretable deep learning methods in recent years, it is now possible to develop trustworthy, high-performing prediction models. This study introduces an uncertainty-aware, graph deep learning model that predicts endovascular thrombectomy outcomes using clinical features and imaging biomarkers. The model targets long-term functional outcomes, defined by the three-month modified Rankin Score (mRS), and mortality rates. A sample of 220 AIS patients in the anterior circulation who underwent endovascular thrombectomy (EVT) was included, with 81 (37%) demonstrating good outcomes (mRS ≤ 2). The performance of the different algorithms evaluated was comparable, with the maximum validation under the curve (AUC) reaching 0.87 using graph convolutional networks (GCN) for mRS prediction and 0.86 using fully connected networks (FCN) for mortality prediction. Moderate performance was obtained at admission (AUC of 0.76 using GCN), which improved to 0.84 post-thrombectomy and to 0.89 a day after stroke. Reliable uncertainty prediction of the model could be demonstrated.

A novel fully automated method for measuring ASPECTS to improve stroke diagnosis: Comparison to traditional ASPECTS.

BACKGROUND AND PURPOSE: To compare the accuracy of subjective Alberta Stroke Program Early CT Score (sASPECTS) evaluation and that of an automated prototype software (aASPECTS) on nonenhanced CT (NECT) in patients with early anterior territory stroke and controls using side-to-side quantification of hypoattenuated brain areas. METHODS: We retrospectively analyzed the NECT scans of 42 consecutive patients with ischemic stroke before reperfusion and 42 controls using first sASPECTS and subsequently aASPECTS. We assessed the differences in Alberta Stroke Program Early CT Score (ASPECTS) and calculated the sensitivity and specificity of NECT with CT perfusion, whereas cerebral blood volume (CBV) served as the reference standard for brain infarction. RESULTS: The clot was located in the middle cerebral artery (MCA) in 47.6% of cases and the internal carotid artery (ICA) in 28.6% of cases. Ten cases presented combined ICA and MCA occlusions. The stroke was right sided in 52.4% of cases and left sided in 47.6%. Reader-based NECT analysis yielded a median sASPECTS of 10. The median CBV-based ASPECTS was 7. Compared to the area of decreased CBV, sASPECTS yielded a sensitivity of 12.5% and specificity of 86.8%. The software prototype (aASPECTS) yielded an overall sensitivity of 65.5% and a specificity of 92.2%. The interreader agreement for ASPECTS evaluation of admission NECT and follow-up CT was almost perfect (κ = .93). The interreader agreement of the CBV color map evaluation was substantial (κ = .77). CONCLUSIONS: aASPECTS of NECT can outperform sASPECTS for stroke detection.

Automated Intracranial Clot Detection: A Promising Tool for Vascular Occlusion Detection in Non-Enhanced CT.

(1) Background: to test the diagnostic performance of a fully convolutional neural network-based software prototype for clot detection in intracranial arteries using non-enhanced computed tomography (NECT) imaging data. (2) Methods: we retrospectively identified 85 patients with stroke imaging and one intracranial vessel occlusion. An automated clot detection prototype computed clot location, clot length, and clot volume in NECT scans. Clot detection rates were compared to the visual assessment of the hyperdense artery sign by two neuroradiologists. CT angiography (CTA) was used as the ground truth. Additionally, NIHSS, ASPECTS, type of therapy, and TOAST were registered to assess the relationship between clinical parameters, image results, and chosen therapy. (3) Results: the overall detection rate of the software was 66%, while the human readers had lower rates of 46% and 24%, respectively. Clot detection rates of the automated software were best in the proximal middle cerebral artery (MCA) and the intracranial carotid artery (ICA) with 88-92% followed by the more distal MCA and basilar artery with 67-69%. There was a high correlation between greater clot length and interventional thrombectomy and between smaller clot length and rather conservative treatment. (4) Conclusions: the automated clot detection prototype has the potential to detect intracranial arterial thromboembolism in NECT images, particularly in the ICA and MCA. Thus, it could support radiologists in emergency settings to speed up the diagnosis of acute ischemic stroke, especially in settings where CTA is not available.

Bifurcation matching for consistent cerebral vessel labeling in CTA of stroke patients.

PURPOSE: Vessel labeling is a prerequisite for comparing cerebral vasculature across patients, e.g., for straightened vessel examination or for localization. Extracting vessels from computed tomography angiography scans may come with a trade-off in segmentation accuracy. Vessels might be neglected or artificially created, increasing the difficulty of labeling. Related work mainly focuses on magnetic resonance angiography without stroke and uses trainable approaches requiring costly labels. METHODS: We present a robust method to identify major arteries and bifurcations in cerebrovascular models generated from existing segmentations. To localize bifurcations of the Circle of Willis, candidate paths for the adjacent vessels of interest are identified using registered landmarks. From those paths, the optimal ones are extracted by recursively maximizing an objective function for all adjacent vessels starting from a bifurcation to avoid erroneous paths and compensate for stroke. RESULTS: In 100 CTA stroke data sets for evaluation, 6 bifurcation locations are placed correctly in 85% of cases; 92.5% when allowing a margin of 5 mm. On average, 14 vessels of interest are found in 90% of the cases and traced correctly end-to-end in 73.5%. The baseline achieves similar detection rates but only 35.5% of the arteries are traced in full. CONCLUSION: Formulating the vessel labeling process as a maximization task for bifurcation matching can vastly improve accurate vessel tracing. The proposed algorithm only uses simple features and does not require expensive training data.

Effects of different virtual monoenergetic CT image data on chest wall post-processing "unfolded ribs" and proposal of an algorithm improvement.

PURPOSE: To find out if the use of different virtual monoenergetic data sets enabled by DECT technology might have a negative impact on post-processing applications, specifically in case of the "unfolded ribs" algorithm. Metal or beam hardening artifacts are suspected to generate image artifacts and thus reduce diagnostic accuracy. This paper tries to find out how the generation of "unfolded rib" CT image reformates is influenced by different virtual monoenergetic CT images and looks for possible improvement of the post-processing tool. MATERIAL AND METHODS: Between March 2021 and April 2021, thin-slice dual-energy CT image data of the chest were used creating "unfolded rib" reformates. The same data sets were analyzed in three steps: first the gold standard with the original algorithm on mixed image data sets followed by the original algorithm on different keV levels (40-120 keV) and finally using a modified algorithm which in the first step used segmentation based on mixed image data sets, followed by segmentation based on different keV levels. Image quality (presence of artifacts), lesion and fracture detectability were assessed for all series. RESULTS: Both, the original and the modified algorithm resulted in more artifact-free image data sets compared to the gold standard. The modified algorithm resulted in significantly more artifact-free image data sets at the keV-edges (40-120 keV) compared the original algorithm. Especially "black artifacts" and pseudo-lesions, potentially inducing false positive findings, could be reduced in all keV level with the modified algorithm. Detection of focal sclerotic, lytic or mixed (k = 0.990-1.000) lesions was very good for all keV levels. The Fleiss-kappa test for detection of fresh and old rib fractures was ≥ 0.997. CONCLUSION: The use of different virtual monoenergetic keVs for the "unfolded rib" algorithm is generating different artifacts. Segmentation-based artifacts could be eliminated by the proposed new algorithm, showing the best results at 70-80 keV.

An algorithm for the labeling and interactive visualization of the cerebrovascular system of ischemic strokes.

During the diagnosis of ischemic strokes, the Circle of Willis and its surrounding vessels are the arteries of interest. Their visualization in case of an acute stroke is often enabled by Computed Tomography Angiography (CTA). Still, the identification and analysis of the cerebral arteries remain time consuming in such scans due to a large number of peripheral vessels which may disturb the visual impression. We propose VirtualDSA++, an algorithm designed to segment and label the cerebrovascular tree on CTA scans. Especially with stroke patients, labeling is a delicate procedure, as in the worst case whole hemispheres may not be present due to impeded perfusion. Hence, we extended the labeling mechanism for the cerebral arteries to identify occluded vessels. In the work at hand, we place the algorithm in a clinical context by evaluating the labeling and occlusion detection on stroke patients, where we have achieved labeling sensitivities comparable to other works between 92% and 95%. To the best of our knowledge, ours is the first work to address labeling and occlusion detection at once, whereby a sensitivity of 67% and a specificity of 81% were obtained for the latter. VirtualDSA++ also automatically segments and models the intracranial system leading to further processing possibilities. We present the generic concept of iterative systematic search for pathways on all nodes of said model, which enables new interactive features. Exemplary, we derive in detail, firstly, the interactive planning of vascular interventions like the mechanical thrombectomy and secondly, the interactive suppression of vessel structures that are not of interest in diagnosing strokes (like veins). We discuss both features as well as further possibilities emerging from the proposed concept.

Noninvasive Collateral Flow Velocity Imaging in Acute Ischemic Stroke: Intraindividual Comparison of 4D-CT Angiography with Digital Subtraction Angiography.

PURPOSE: The collateral status can be defined not only by its morphological extent but also by the velocity of collateral filling characterized by the relative filling time delay (rFTD). The aim of our study was to compare different methods of noninvasive visualization of rFTD derived from 4D-CT angiography (4D-CTA) with digital substraction angiography (DSA) and to investigate the correlation between functional and morphological collateral status on timing-invariant CTA. MATERIALS AND METHODS: 50 consecutive patients with acute occlusion in the M1 segment who underwent DSA for subsequent mechanical recanalization after multimodal CT were retrospectively analyzed. 4D-CTA data were used to assess the relative filling time delay between the A1 segment of the affected hemisphere and the sylvian branches distal to the occluded M1 segment using source images (4D-CTA-SI) and color-coded flow velocity visualization with prototype software (fv-CTA) in comparison to DSA. The morphological extent of collaterals was assessed on the basis of the Collateral Score (CS) on temporal maximum intensity projections (tMIP) derived from CT perfusion data. RESULTS: There was very good correlation of rFTD between fv-CTA and DSA (n = 50, r = 0.9, p < 0.05). Differences of absolute rFTD values were not significant. 4D-CTA-SI and DSA also showed good correlation (n = 50, r = 0.6, p < 0.05), but mean values of rFTD were significantly different (p < 0.05). rFTD derived from fvCTA and CS derived from timing-invariant CTA showed a negative association (R = - 0.5; P = 0.000). In patients with a favorable radiological outcome defined by a TICI score of 2b or 3, there was a significant negative correlation of CS and mRS at 3 months (R = - 0.4, P = 0.006). CONCLUSION: Collateral status plays an important role in the outcome in stroke patients. rFTD derived from 4D-CTA is a suitable parameter for noninvasive imaging of collateral velocity, which correlates with the morphological extent of collaterals. Further studies are needed to define valid thresholds for rFTD and to evaluate the diagnostic and prognostic value. KEY POINTS: · Collateral supply in anterior circulation stroke can be defined by the velocity of collateral filling. · Relative filling time delay (rFTD) can serve for quantitative measurement of collateral flow and correlates with the morphological extent of collaterals. · 4D-CTA is a suitable noninvasive imaging technique. CITATION FORMAT: · Muehlen I, Kloska SP, Gölitz P et al. Noninvasive Collateral Flow Velocity Imaging in Acute Ischemic Stroke: Intraindividual Comparison of 4D-CT Angiography with Digital Subtraction Angiography. Fortschr Röntgenstr 2019; 191: 827 - 835.

Improved CT-detection of acute bowel ischemia using frequency selective non-linear image blending.

BACKGROUND: Computed tomography (CT) as a fast and reliable diagnostic technique is the imaging modality of choice for acute bowel ischemia. However, diagnostic is often difficult mainly due to low attenuation differences between ischemic and perfused segments. PURPOSE: To compare the diagnostic efficacy of a new post-processing tool based on frequency selective non-linear blending with that of conventional linear contrast-enhanced CT (CECT) image blending for the detection of bowel ischemia. MATERIAL AND METHODS: Twenty-seven consecutive patients (19 women; mean age = 73.7 years, age range = 50-94 years) with acute bowel ischemia were scanned using multidetector CT (120 kV; 100-200 mAs). Pre-contrast and portal venous scans (65-70 s delay) were acquired. All patients underwent surgery for acute bowel ischemia and intraoperative diagnosis as well as histologic evaluation of explanted bowel segments was considered "gold standard." First, two radiologists read the conventional CECT images in which linear blending was adapted for optimal contrast, and second (three weeks later) the frequency selective non-linear blending (F-NLB) image. Attenuation values were compared, both in the involved and non-involved bowel segments creating ratios between unenhanced and CECT. RESULTS: The mean attenuation difference between ischemic and non-ischemic wall in the portal venous scan was 69.54 HU (reader 2 = 69.01 HU) higher for F-NLB compared with conventional CECT. Also, the attenuation ratio between contrast-enhanced and pre-contrast CT data for the non-ischemic walls showed significantly higher values for the F-NLB image (CECT: reader 1 = 2.11 (reader 2 = 3.36), F-NLB: reader 1 = 4.46 (reader 2 = 4.98)]. Sensitivity in detecting ischemic areas increased significantly for both readers using F-NLB (CECT: reader 1/2 = 53%/65% versus F-NLB: reader 1/2 = 62%/75%). CONCLUSION: Frequency selective non-linear blending improves detection of bowel ischemia compared with conventional CECT by increasing attenuation differences between ischemic and perfused segments.

Differences in Texture Analysis Parameters Between Active Alveolitis and Lung Fibrosis in Chest CT of Patients with Systemic Sclerosis: A Feasibility Study.

RATIONALE AND OBJECTIVES: This study aimed to determine the diagnostic aid of computed tomography (CT) features for the differentiation of active alveolitis and fibrosis using a CT texture analysis (CTTA) prototype and CT densitometry in patients with systemic sclerosis (SSc) using ancillary high-resolution computed tomography (HRCT) features and their longitudinal course as standard of reference. MATERIALS AND METHODS: We retrospectively analyzed thin-slice noncontrast chest CT image data of 43 patients with SSc (18 men, mean age 51.55 ± 15.52 years; range 23-71 years). All of them had repeated noncontrast enhanced HRCT of the lung. Classification into active alveolitis or fibrosis was done on HRCT based on classical HRCT findings (active alveolitis [19; 44.2%] and fibrosis [24; 55.8%]) and their course at midterm. Results were compared to pulmonary functional tests and were followed up by CT. Ground glass opacity was considered suggestive of alveolitis, whereas coarse reticulation with parenchymal distortion, traction bronchiectasis, and honeycombing were assigned to fibrosis. RESULTS: Statistically significant differences in CTTA were found for first-order textural features (mean intensity, average, deviation, skewness) and second-order statistics (entropy of co-occurrence matrix, mean number of nonuniformity (NGLDM), entropy of NGLDM, entropy of heterogeneity, intensity, and average). Cut-off value for the prediction of fibrosis at baseline was significant for entropy of intensity (P value < .001) and for mean deviation (P value < .001), and for prediction of alveolitis was significant for uniformity of intensity (P value < .001) and for NGLDM (P value < .001). At pulmonary functional tests, forced expiratory volume in 1 second and single-breath diffusion capacity for carbon monoxide were significantly lower in fibrosis than in alveolitis 2.03 ± 0.78 vs. 2.61 ± 0.83, P < .016 and 4.51 ± 1.61 vs. 6.04 ± 1.75, P < .009, respectively. Differences in CT densitometry between alveolitis and fibrosis were not significant. CONCLUSIONS: CTTA parameters are significantly different in active alveolitis vs. fibrosis in patients with SSc and may be helpful for differentiation of these two entities.

Improved MDCT monitoring of pelvic myeloma bone disease through the use of a novel longitudinal bone subtraction post-processing algorithm.

PURPOSE: To evaluate the diagnostic performance of a novel CT post-processing software that generates subtraction maps of baseline and follow-up CT examinations in the course of myeloma bone lesions. MATERIALS AND METHODS: This study included 61 consecutive myeloma patients who underwent repeated whole-body reduced-dose MDCT at our institution between November 2013 and June 2015. CT subtraction maps classified a progressive disease (PD) vs. stable disease (SD)/remission. Bone subtraction maps (BSMs) only and in combination with 1-mm (BSM+) source images were compared with 5-mm axial/MPR scans. RESULTS: Haematological response categories at follow-up were: complete remission (n = 9), very good partial remission (n = 2), partial remission (n = 17) and SDh (n = 19) vs. PDh (n = 14). Five-millimetre CT scan yielded PD (n = 14) and SD/remission (n = 47) whereas bone subtraction + 1-mm axial scans (BSM+) reading resulted in PD (n = 18) and SD/remission (n = 43). Sensitivity/ specificity/accuracy for 5-mm/1-mm/BSM(alone)/BSM + in "lesion-by-lesion" reading was 89.4 %/98.9 %/98.3 %/ 99.5 %; 69.1 %/96.9 %/72 %/92.1 % and 83.8 %/98.4 %/92.1 %/98.3 %, respectively. The use of BSM+ resulted in a change of response classification in 9.8 % patients (n = 6) from SD to PD. CONCLUSION: BSM reading is more accurate for monitoring myeloma compared to axial scans whereas BSM+ yields similar results with 1-mm reading (gold standard) but by significantly reduced reading time. KEY POINTS: • CT evaluation of myeloma bone disease using a longitudinal bone subtraction post-processing algorithm. • Bone subtraction post-processing algorithm is more accurate for assessment of therapy. • Bone subtraction allowed improved and more efficient detection of myeloma bone lesions. • Post-processing tool demonstrating a change in response classification in 9.8 % patients (all showing PD). • Reading time could be substantially shortened as compared to regular CT assessment.

Longitudinal Computed Tomography Monitoring of Pelvic Bones in Patients With Breast Cancer Using Automated Bone Subtraction Software.

OBJECTIVE: The aim of this study was to optimize computed tomography (CT) surveillance of skeletal metastases in patients with breast cancer through the use of osseous subtraction maps between baseline and follow-up examinations created by a novel software algorithm. The new postprocessing algorithm segments the original bone followed by image intensity-based rigid alignment creating gray-shaded maps that highlight focal or diffuse loss or increase in bone attenuation. MATERIALS AND METHODS: Institutional review board was obtained for this retrospective data evaluation. A total of 33 consecutive patients (31 female; 2 male; mean age, 59.13 ± 12.68 years; range, 32-81 years) with breast cancer were included, who underwent 143 standardized baseline and follow-up CT examinations between February 2014 and June 2016. We classified bone metastases into lytic, sclerotic, and mixed osseous lesions. Any new osteolysis inside a known sclerotic lesion and enlargement of pre-existing sclerotic lesions were considered to represent progressive disease (PD), whereas no change was classified as stable disease (SD). Results were compared additionally with the course of the disease considering the entire skeleton and other involved organs. Software-created automated bone subtraction maps were compared with conventional CT interpretations of axial 5-mm and coronal 1-mm reformatted images. Region of interest measurements were used to quantify new lesions. Results were validated by clinical and CT follow-up. Reading time was evaluated. RESULTS: Skeletal metastases were present in 17/33 (51%) patients (9 sclerotic, 2 lytic, 6 mixed) at baseline. The use of bone subtraction maps resulted in an overall change of response classification into PD in 9/33 (8.1%) patients. Compared with conventional CT evaluation, the bone subtraction maps disclosed 123 new or enlarging sclerotic and 32 new lytic metastases in 23/33 (30.9%) examinations. Mean attenuation of new bone lesions (sclerotic or lytic) significantly increased or decreased (P < 0.01) in all patients. Bone attenuation in pelvic areas without evident metastatic disease significantly increased in patients with PD (P = 0.019), whereas there was no change in SD (P = 0.076). Lesion-based sensitivity, specificity, accuracy, positive predictive values, and negative predictive values were 98.7%, 79.5%, 94.5%, 95.1%, and 94.5%, respectively. Interobserver agreement was good (κ = 0.80; P = 0.077). Reading time was significantly faster for the bone subtraction maps versus 5-mm axial images (P < 0.001). CONCLUSIONS: Longitudinal bone subtraction maps increase the accuracy and efficiency of CT diagnosis of skeletal metastases in patients with breast cancer.

Improved CT Detection of Acute Herpes Simplex Virus Type 1 Encephalitis Based on a Frequency-Selective Nonlinear Blending: Comparison With MRI.

OBJECTIVE: The purpose of this study is to compare the diagnostic efficacy of a new CT postprocessing tool based on frequency-selective nonlinear blending (best-contrast CT) with that of standard linear blending of unenhanced head CT in patients with herpes simplex virus type 1 and herpes simplex virus encephalitis (HSE), using FLAIR MRI sequences as the standard of reference. MATERIALS AND METHODS: Fifteen consecutive patients (six women and nine men; mean [± SD] age, 60 ± 19 years) with proven HSE (positive polymerase chain reaction results from CSF analysis and the presence of neurologic deficits) were retrospectively enrolled. All patients had undergone head CT and MRI (mean time interval, 2 ± 2 days). After standardized unenhanced head CT scans were read, presets of the best-contrast algorithm were determined (center, 30 HU; delta, 5 HU; slope, 5 nondimensional), and resulting images were analyzed. Contrast enhancement was objectively measured by ROI analysis, comparing contrast-to-noise ratios (CNRs) of unenhanced CT and best-contrast CT. FLAIR and DWI MRI sequences were analyzed, and FLAIR was considered as the standard of reference. For assessment of disease extent, a previously reported 50-point score (HSE score) was used. RESULTS: CNR values for unenhanced head CT (CNR, 5.42 ± 2.77) could be statistically significantly increased using best-contrast CT (CNR, 9.62 ± 4.28) (p = 0.003). FLAIR sequences yielded a median HSE score of 9.0 (range, 6-17) and DWI sequences yielded HSE scores of 6.0 (range, 5-17). By comparison, unenhanced head CT resulted in a median HSE score of 3.5 (range, 1-6). The median best-contrast CT HSE score was 7.5 (range, 6-10). Agreement between FLAIR and unenhanced CT was 54.44%, that between DWI and best-contrast CT was 95.36%, and that between FLAIR and best-contrast CT was 85.21%. The most frequently overseen findings were located at the level of the upper part of the mesencephalon and at the subthalamic or insular level. CONCLUSION: Frequency-selective nonlinear blending significantly increases contrast and detects brain parenchymal involvement in HSE more sensitively compared with unenhanced CT. The sensitivity of best-contrast CT seems to be equal to that of DWI and almost as good as that of FLAIR.

Image Quality of a Novel Frequency Selective Nonlinear Blending Algorithm: An Ex Vivo Phantom Study in Comparison to Single-Energy Acquisitions and Dual-Energy Acquisitions With Monoenergetic Reconstructions.

OBJECTIVE: Aim of this ex vivo phantom study was to evaluate the contrast enhancement applying a new frequency split nonlinear blending algorithm (best contrast [BC]) and to compare it with standard 120-kV single-energy computed tomography (SECT) images, as well as with low-kiloelectron volt monoenergetic extrapolations (Mono+40-100keV) from dual-energy CT (DECT) and with low-kilovolt (70-100 kV) SECT acquisitions. MATERIALS AND METHODS: A dilution series of iodinated contrast material-filled syringes was centered in an attenuation phantom and was scanned with SECT70-120kV and DECT80-100/Sn150. Monoenergetic images (40-100 keV) were reconstructed, and a new manual frequency split nonlinear blending algorithm (BC) was applied to SECT70kV and SECT120kV images. Manual BC settings were set to simulate a reading situation with fixed overall best values (FVBC120kV) as well as to achieve best possible values for each syringe (BVBC120kV) for maximum contrast enhancement. Contrast-to-noise ratios (CNRs) were used as an objective region of interest-based image analysis parameter. Two radiologists evaluated the detectability of hyperdense and hypodense syringes (Likert). Results were compared between SECT70-100kV, Mono+40-100keV, and DECT80-100/Sn150kV, as well as FVBC120kV, BVBC120kV, and BC70kV. RESULTS: Highest CNR without BC was detected at SECT70kV (5.04 ± 0.12) and Mono+40keV (4.40 ± 0.11). FVBC and BVBC images allow a significant increase of CNR compared with SECT120kV (CNRBVBC, 5.21 ± 0.15; CNRFVBC, 5.12 ± 0.16; CNRSECT120kV, 2.5 ± 0.08; all P ≤ 0.01). There was no significant difference in CNR between BVBC and FVBC. Averaged CNR in BVBC and FVBC was significantly higher compared with Mono+40-100keV (all P ≤ 0.01). Compared with SECT70kV, averaged CNR in BVBC and FVBC show no significant differences. BVBC70kV (7.67 ± 0.17) significantly increases CNR in SECT70kV up to 213%.Subjective image analysis showed an interobserver agreement of 0.63 to 0.83 (κ), confirming the superiority of BC in detecting hyperdense and hypodense syringes, when compared with SECT120kV. Compared with SECT120kV, BVBC70kV was scored highest, followed by SECT70kV. BVBC showed higher scores when comparing to Mono+40keV, however almost identical to those of SECT70kV. Scores of FVBC were slightly lower than SECT70kV, but in the range of Mono+40keV. CONCLUSIONS: The new frequency split nonlinear blending algorithm with fixed settings offers a superior differentiation of contrast levels from low- to high-contrast settings. Using the optimal settings, this algorithm shows an equivalent contrast enhancement when compared with SECT70kV. Because of the non-DECT-based algorithm of BC, the new method of contrast enhancement seems to be particularly valuable for implementation in CT systems not equipped for dual-energy or spectral CT imaging.

Enhanced gray-white matter differentiation on non-enhanced CT using a frequency selective non-linear blending.

INTRODUCTION: The aim if this study is to find out if contrast between gray (GM) and white matter (WM) on non-enhanced brain CT (NECT) can be enhanced by using a frequency selective non-linear blending. METHODS: Thirty consecutive patients (40 % female; mean age 67.73 ± 12.71 years), who underwent NECT of the brain, were retrospectively included in this study. Brain scan readings were performed by two radiologists independently, for NECT and subsequently the images were read using a new frequency selective non-linear blending algorithm (best contrast, BC). Optimal settings of BC for enhanced delineation of anatomical structures were set at an averaged center of 30 HU, averaged delta of 5 HU, and a slope of 5. For contrast-to-noise ratio calculation (CNR), gray and white matter attenuation values were measured for both NECT and BC in different anatomical structures. RESULTS: CNR increase in the gray matter was 5.91 ± 2.45 for the cortical gray matter and 4.41 ± 1.82 for the basal ganglia. The contrast ratio between cortical gray and white matter was 1.87 and 1.7 (basal ganglia/WM) for BC quantification vs. 1.43 (cortex/WM) and 1.33 (basal ganglia/WM) for standard NECT (both p < 0.0001). Improved CNR did not depend on the anatomical structures measured. CONCLUSION: Frequency selective non-linear blending allows better discrimination between WM and GM and therefore may enhance diagnostic accuracy of NECT.

Accuracy of Non-Enhanced CT in Detecting Early Ischemic Edema Using Frequency Selective Non-Linear Blending.

PURPOSE: Ischemic brain edema is subtle and hard to detect by computed tomography within the first hours of stroke onset. We hypothesize that non-enhanced CT (NECT) post-processing with frequency-selective non-linear blending ("best contrast"/BC) increases its accuracy in detecting edema and irreversible tissue damage (infarction). METHODS: We retrospectively analyzed the NECT scans of 76 consecutive patients with ischemic stroke (exclusively middle cerebral artery territory-MCA) before and after post-processing with BC both at baseline before reperfusion therapy and at follow-up (5.73±12.74 days after stroke onset) using the Alberta Stroke Program Early CT Score (ASPECTS). We assessed the differences in ASPECTS between unprocessed and post-processed images and calculated sensitivity, specificity, and predictive values of baseline NECT using follow-up CT serving as reference standard for brain infarction. RESULTS: NECT detected brain tissue hypoattenuation in 35 of 76 patients (46.1%). This number increased to 71 patients (93.4%) after post-processing with BC. Follow-up NECT confirmed brain infarctions in 65 patients (85.5%; p = 0.012). Post-processing increased the sensitivity of NECT for brain infarction from 35/65 (54%) to 65/65 (100%), decreased its specificity from 11/11 (100%) to 7/11 (64%), its positive predictive value (PPV) from 35/35 (100%) to 65/69 (94%) and increased its accuracy 46/76 (61%) to 72/76 (95%). CONCLUSIONS: This post-hoc analysis suggests that post-processing of NECT with BC may increase its sensitivity for ischemic brain damage significantly.

Segmental bronchi collapsibility: computed tomography-based quantification in patients with chronic obstructive pulmonary disease and correlation with emphysema phenotype, corresponding lung volume changes and clinical parameters.

BACKGROUND: Global pulmonary function tests lack region specific differentiation that might influence therapy in severe chronic obstructive pulmonary disease (COPD) patients. Therefore, the aim of this work was to assess the degree of expiratory 3rd generation bronchial lumen collapsibility in patients with severe COPD using chest-computed tomography (CT), to evaluate emphysema-phenotype, lobar volumes and correlate results with pulmonary function tests. METHODS: Thin-slice chest-CTs acquired at end-inspiration & end-expiration in 42 COPD GOLD IV patients (19 females, median-age: 65.9 y) from November 2011 to July 2014 were re-evaluated. The cross-sectional area of all segmental bronchi was measured 5 mm below the bronchial origin in both examinations. Lung lobes were semi-automatically segmented, volumes calculated at end-inspiratory and end-expiratory phase and visually defined emphysema-phenotypes defined. Results of CT densitometry were compared with lung functional tests including forced expiratory volume at 1 s (FEV1), total lung capacity (TLC), vital capacity (VC), residual volume (RV), diffusion capacity parameters and the maximal expiratory flow rates (MEFs). RESULTS: Mean expiratory bronchial collapse was 31%, stronger in lobes with homogenous (38.5%) vs. heterogeneous emphysema-phenotype (27.8%, P=0.014). The mean lobar expiratory volume reduction was comparable in both emphysema-phenotypes (volume reduction 18.6%±8.3% in homogenous vs. 17.6%±16.5% in heterogeneous phenotype). The degree of bronchial lumen collapsibility, did not correlate with expiratory volume reduction. MEF25 correlated weakly with 3rd generation airway collapsibility (r=0.339, P=0.03). All patients showed a concentric expiratory reduction of bronchial cross-sectional area. CONCLUSIONS: Changes in collapsibility of 3rd generation bronchi in COPD grade IV patients is significantly lower than that in the trachea and the main bronchi. Collapsibility did not correlate with the reduction in lung volume but was significantly higher in lobes with homogeneous vs. heterogeneous emphysema phenotype. Changes in the 3rd generation bronchial calibres between inspiration and expiration are not predictive for the degree of small airway collapsibility and related airflow limitation.

Improved Follow-Up and Response Monitoring of Thoracic Cage Involvement in Multiple Myeloma Using a Novel CT Postprocessing Software: The Lessons We Learned.

OBJECTIVE: The purpose of this study is to evaluate the benefit of using novel CT postprocessing software that generates unfolded rib images for more-accurate evaluation of multiple myeloma (MM) at follow-up, response monitoring, and visualization of treatment-related bone changes. MATERIALS AND METHODS: Between January 2012 and February 2015, 40 consecutive patients with MM underwent repeated whole-body reduced-dose CT at our institution. The results were retrospectively evaluated and compared with established hematologic markers. Unfolded rib reformatted images were compared with 5- and 1-mm-thick slices with regard to bone changes, bone marrow attenuation, and bone sclerosis. RESULTS: Hematologic response categories at follow-up were complete response (CR; n = 2), very good partial response (VGPR; n = 1), partial response (PR; n = 9), stable disease (n = 9), and progressive disease (PD; n = 19). The number of lesions increased in 11 patients (all with PD), decreased in two patients (both with CR), and stayed unchanged in 27 patients. The size of the lesions increased in 14 patients (all with PD), decreased in five patients (two with CR, two with PR, and one with stable disease), and remained unchanged in 21 patients. There was a mean (± SD) difference of 27.99 ± 19.71 HU in bone marrow attenuation for patients with PD (p < 0.0001) and -31.24 ± 13.57 HU in the responders group (p = 0.002), whereas patients with stable disease showed stable bone marrow attenuation at follow-up (mean, -3.37 ± 10.55 HU). Increased bone sclerosis was detected in 12 patients (all of whom were receiving therapy). The sensitivity and specificity of unfolded rib images, 5-mm slices, and 1-mm slices were, respectively, 78.9% and 100%, 52.6% and 100%, and 63.2% and 100% for accurate bone response assessment; 100% and 95.2%, 94.74% and 42.9%, and 89.47% and 47.62% for bone marrow attenuation; and 100% and 100%, 58.3% and 100%, and 91.67% and 100% for sclerosis. CONCLUSION: For therapy response assessment, unfolded rib reading is more accurate than transverse CT slices.

Enhanced reading time efficiency by use of automatically unfolded CT rib reformations in acute trauma.

Rationale of this study was to evaluate whether unfolded rib images enhance time efficiency in detection of rib fractures and time efficiency in patients with acute thoracic trauma. 51 subsequent patients with thoracic trauma underwent 64-slice computed tomography. 1mm thick axial slices were reformatted using a commercially available post-processing software application generating rotatable unfolded rib images. Diagnostic accuracy was evaluated by 3 readers and compared to multiplanar reformations of the original CT images. Reformation and evaluation times were recorded. 116 rib fractures were detected. The multiplanar reformation analysis yielded a sensitivity of 87.9%/93.9%/79.7% with a specificity of 97%/97%/82.2%, whilst the unfolded rib image analysis yielded a sensitivity of 94.8%/94.8%/92.2% and a specificity of 85.2/87.8%/82.4 (p=0.06/0.8/0.04) with high inter-observer agreement (k=0.79-0.85). The mean reading time for the multiplanar reformations was significantly longer (reader 1: 103.7 ± 27.1s/reader 2: 81.8 ± 40.6s/reader 3: 154.3 ± 39.2s) than the evaluation of the unfolded rib images (19.4 ± 4.9s/26.9 ± 15.0s/49.9 ± 18.7s; p<0.01). Concluding, the unfolded rib display reduces reading time for detection of rib fractures in acute thoracic trauma patients significantly and does not compromise the diagnostic accuracy significantly in experienced radiologists. However, unexperienced readers may profit from use of this display.

Improvement of diagnostic confidence for detection of multiple myeloma involvement of the ribs by a new CT software generating rib unfolded images: Comparison with 5- and 1-mm axial images.

OBJECTIVE: To investigate the performance of a new CT software generating rib unfolded images for improved detection of rib osteolyses in patients with multiple myeloma. MATERIALS AND METHODS: One hundred sixteen patients who underwent whole-body reduced-dose multidetector computed tomography (WBRD-MDCT) for multiple myeloma diagnosis and during follow-up were retrospectively evaluated. Nonenhanced CT scans with 5- and 1-mm slice thickness were interpreted by two readers with focus on detection of rib involvement (location, number, fracture). Image analysis of "unfolded," 1-mm-based CT rib images was subsequently undertaken. We classified the number of lytic bone lesions into 0, 1, 2, <5, <10 and ≥10. For all three data sets the reading time was registered. RESULTS: An approximated sum of 6,727 myeloma-related rib lesions was found. On a patient-based analysis, CT (5 mm), CT (1 mm) and CT (1 mm "unfolded rib") yielded a sensitivity, specificity and accuracy of 79.7/94.7/87.1, 88.1/93/90.5 and 98.3/96.5/97.4, respectively. In a lesion-based analysis, the sensitivity, specificity and accuracy of the three evaluations were 69.7/87.2/70.5, 79.8/55.9/78 and 96.5/89.7/96.1. Mean reading time for 5 mm/1 mm axial images and unfolded images was 178.7/215.1/90.8 s, respectively. CONCLUSION: The generation of "unfolded rib" images improves detection of rib involvement in patients with multiple myeloma and significantly reduces reading time.

Color-coded cerebral computed tomographic angiography: implementation of a convolution-based algorithm and first clinical evaluation in patients with acute ischemic stroke.

OBJECTIVE: The objective of this study was to develop a new method of displaying dynamic cerebral computed tomographic (CT) angiography (CTA) data sets in which the time delay to maximum enhancement (Tdelay) is displayed in a range of colors (color-coded CT angiography [cCTA]). MATERIALS AND METHODS: This institutional review board-approved study included multiparametric CT data sets from 16 patients with different types of supra-aortic large vessel occlusions. Color-coded CT angiography was reconstructed from CT perfusion raw data sets. All voxel enhancement curves were fitted to f(t) = α · AIFmtt(t - Δt), with AIFmtt(t), indicating enhancement of AIF dilated by convolution with boxcar function (with mean transit time [mtt]); α, scaling factor; and [INCREMENT]t, transition along the time. The time delay to maximum enhancement was defined as Tdelay = Δt +0.5 · mtt. Values of Tdelay were color-coded and superimposed on temporal maximum intensity projections CTA resulting in colored angiographic composite images. For a pilot clinical evaluation, diagnostic confidence in determining the pathology, quality of the visualization of leptomeningeal collaterals, and additional diagnostic information were assessed. RESULTS: The reconstruction of cCTA was technically feasible in all 16 patients. Both diagnostic confidence (P < 0.01) and the quality of the visualization of collaterals (P < 0.0001) were significantly higher when using the combination of single-phase CTA and cCTA compared with single-phase CTA alone. Additional diagnostic information was obtained with cCTA regarding occlusion type (reader 1: 5 cases and reader 2: 4 cases), differentiation between arteries and veins (11/13), differentiation between antegrade and retrograde filling (12/13), as well as leptomeningeal collateralization (13/14). CONCLUSIONS: Color-coded CT angiography is a technically feasible technique that provides additional information on cerebral hemodynamics in ischemic stroke patients.