Explore the vast range of titles available.

AI provides tremendous opportunities for improving patient care, but at present there is little evidence of real-world uptake. An important barrier is the lack of well-designed, vendor-neutral and future-proof infrastructures for deployment. Because current AI algorithms are very narrow in scope, it is expected that a typical hospital will deploy many algorithms concurrently. Managing stand-alone point solutions for all of these algorithms will be unmanageable. A solution to this problem is a dedicated platform for deployment of AI. Here we describe a blueprint for such a platform and the high-level design and implementation considerations of such a system that can be used clinically as well as for research and development. Close collaboration between radiologists, data scientists, software developers and experts in hospital IT as well as involvement of patients is crucial in order to successfully bring AI to the clinic.

Objectives We aimed to evaluate the real-world variation in CT perfusion (CTP) imaging protocols among stroke centers and to explore the potential for standardizing vendor software to harmonize CTP images. Methods Stroke centers participating in a nationwide multicenter healthcare evaluation were requested to share their CTP scan and processing protocol. The impact of these protocols on CTP imaging was assessed by analyzing data from an anthropomorphic phantom with center-specific vendor software with default settings from one of three vendors (A–C): IntelliSpace Portal, syngoVIA, and Vitrea. Additionally, standardized infarct maps were obtained using a logistic model. Results Eighteen scan protocols were studied, all varying in acquisition settings. Of these protocols, seven, eight, and three were analyzed with center-specific vendor software A, B, and C respectively. The perfusion maps were visually dissimilar between the vendor software but were relatively unaffected by the acquisition settings. The median error [interquartile range] of the infarct core volumes (mL) estimated by the vendor software was − 2.5 [6.5] (A)/ − 18.2 [1.2] (B)/ − 8.0 [1.4] (C) when compared to the ground truth of the phantom (where a positive error indicates overestimation). Taken together, the median error [interquartile range] of the infarct core volumes (mL) was − 8.2 [14.6] before standardization and − 3.1 [2.5] after standardization. Conclusions CTP imaging protocols varied substantially across different stroke centers, with the perfusion software being the primary source of differences in CTP images. Standardizing the estimation of ischemic regions harmonized these CTP images to a degree. Clinical relevance statement The center that a stroke patient is admitted to can influence the patient’s diagnosis extensively. Standardizing vendor software for CT perfusion imaging can improve the consistency and accuracy of results, enabling a more reliable diagnosis and treatment decision. Key Points • CT perfusion imaging is widely used for stroke evaluation, but variation in the acquisition and processing protocols between centers could cause varying patient diagnoses. • Variation in CT perfusion imaging mainly arises from differences in vendor software rather than acquisition settings, but these differences can be reconciled by standardizing the estimation of ischemic regions. • Standardizing the estimation of ischemic regions can improve CT perfusion imaging for stroke evaluation by facilitating reliable evaluations independent of the admission center.

In this study, we investigate whether a Convolutional Neural Network (CNN) can generate informative parametric maps from the pre-processed CT perfusion data in patients with acute ischemic stroke in a clinical setting. The CNN training was performed on a subset of 100 pre-processed perfusion CT dataset, while 15 samples were kept for testing. All the data used for the training/testing of the network and for generating ground truth (GT) maps, using a state-of-the-art deconvolution algorithm, were previously pre-processed using a pipeline for motion correction and filtering. Threefold cross validation had been used to estimate the performance of the model on unseen data, reporting Mean Squared Error (MSE). Maps accuracy had been checked through manual segmentation of infarct core and total hypo-perfused regions on both CNN-derived and GT maps. Concordance among segmented lesions was assessed using the Dice Similarity Coefficient (DSC). Correlation and agreement among different perfusion analysis methods were evaluated using mean absolute volume differences, Pearson correlation coefficients, Bland-Altman analysis, and coefficient of repeatability across lesion volumes. The MSE was very low for two out of three maps, and low in the remaining map, showing good generalizability. Mean Dice scores from two different raters and the GT maps ranged from 0.80 to 0.87. Inter-rater concordance was high, and a strong correlation was found between lesion volumes of CNN maps and GT maps (0.99, 0.98, respectively). The agreement between our CNN-based perfusion maps and the state-of-the-art deconvolution-algorithm perfusion analysis maps, highlights the potential of machine learning methods applied to perfusion analysis. CNN approaches can reduce the volume of data required by deconvolution algorithms to estimate the ischemic core, and thus might allow the development of novel perfusion protocols with lower radiation dose deployed to the patient.

Background During the cochlear implantation procedure, the receiver/stimulator (R/S) part of the implant is fixated to prevent postoperative device migration, which could have an adverse effect on the position of the electrode array in the cochlea. We aim to compare the migration rates of two fixation techniques, the bony recess versus the subperiosteal tight pocket without bony sutures. Methods and analysis This single-blind randomized controlled trial will recruit a total of 112 primary cochlear implantation adult patients, eligible for implantation according to the current standard of practice. Randomization will be performed by an electronic data capture system Castor EDC, with participants block randomized to either bony recess or standard subperiosteal tight pocket in a 1:1 ratio, stratified by age. The primary outcome of this study is the R/S device migration rate; secondary outcomes include patient-experienced burden using the validated COMPASS questionnaire, electrode migration rate, electrode impedance values, speech perception scores, correlation between R/S migration, electrode array migration and patient complaints, assessment of complication rates, and validation of an implant position measurement method. Data will be collected at baseline, 1 week, 4 weeks, 8 weeks, 3 months, and 12 months after surgery. All data analyses will be conducted according to the intention-to-treat principle. Discussion Cochlear implantation by means of creating a tight subperiosteal pocket without drilling a bony seat is a minimally invasive fixation technique with many advantages. However, the safety of this technique has not yet been proven with certainty. This is the first randomized controlled trial that directly compares the minimally invasive technique with the conventional method of drilling a bony seat. Trial registration Netherlands Trial Register NL9698. Registered on 31 August 2021.

Dual-energy CT (DECT) material decomposition techniques may better detect edema within cerebral infarcts than conventional non-contrast CT (NCCT). This study compared if Virtual Ischemia Maps (VIM) derived from non-contrast DECT of patients with acute ischemic stroke due to large-vessel occlusion (AIS-LVO) are superior to NCCT for ischemic core estimation, compared against reference-standard DWI-MRI. Only patients whose baseline ischemic core was most likely to remain stable on follow-up MRI were included, defined as those with excellent post-thrombectomy revascularization or no perfusion mismatch. Twenty-four consecutive AIS-LVO patients with baseline non-contrast DECT, CT perfusion (CTP), and DWI-MRI were analyzed. The primary outcome measure was agreement between volumetric manually segmented VIM, NCCT, and automatically segmented CTP estimates of the ischemic core relative to manually segmented DWI volumes. Volume agreement was assessed using Bland–Altman plots and comparison of CT to DWI volume ratios. DWI volumes were better approximated by VIM than NCCT (VIM/DWI ratio 0.68 ± 0.35 vs. NCCT/DWI ratio 0.34 ± 0.35; P < 0.001) or CTP (CTP/DWI ratio 0.45 ± 0.67; P < 0.001), and VIM best correlated with DWI (r VIM  = 0.90; r NCCT  = 0.75; r CTP  = 0.77; P < 0.001). Bland–Altman analyses indicated significantly greater agreement between DWI and VIM than NCCT core volumes (mean bias 0.60 [95%AI 0.39–0.82] vs. 0.20 [95%AI 0.11–0.30]). We conclude that DECT VIM estimates the ischemic core in AIS-LVO patients more accurately than NCCT.

Blood–brain barrier damage, which can be quantified by measuring vascular permeability, is a potential predictor for hemorrhagic transformation in acute ischemic stroke. Permeability is commonly estimated by applying Patlak analysis to computed tomography (CT) perfusion data, but this method lacks precision. Applying more elaborate kinetic models by means of nonlinear regression (NLR) may improve precision, but is more time consuming and therefore less appropriate in an acute stroke setting. We propose a simplified NLR method that may be faster and still precise enough for clinical use. The aim of this study is to evaluate the reliability of in total 12 variations of Patlak analysis and NLR methods, including the simplified NLR method. Confidence intervals for the permeability estimates were evaluated using simulated CT attenuation-time curves with realistic noise, and clinical data from 20 patients. Although fixating the blood volume improved Patlak analysis, the NLR methods yielded significantly more reliable estimates, but took up to 12 x longer to calculate. The simplified NLR method was ~4 x faster than other NLR methods, while maintaining the same confidence intervals (CIs). In conclusion, the simplified NLR method is a new, reliable way to estimate permeability in stroke, fast enough for clinical application in an acute stroke setting.

Although CT scanners generally allow dynamic acquisition of thin slices (1 mm), thick slice (≥5 mm) reconstruction is commonly used for stroke imaging to reduce data, processing time, and noise level. Thin slice CT perfusion (CTP) reconstruction may suffer less from partial volume effects, and thus yield more accurate quantitative results with increased resolution. Before thin slice protocols are to be introduced clinically, it needs to be ensured that this does not affect overall CTP constancy. We studied the influence of thin slice reconstruction on average perfusion values by comparing it with standard thick slice reconstruction. From 50 patient studies, absolute and relative hemisphere averaged estimates of cerebral blood volume (CBV), cerebral blood flow (CBF), mean transit time (MTT), and permeability-surface area product (PS) were analyzed using 0.8, 2.4, 4.8, and 9.6 mm slice reconstructions. Specifically, the influence of Gaussian and bilateral filtering, the arterial input function (AIF), and motion correction on the perfusion values was investigated. Bilateral filtering gave noise levels comparable to isotropic Gaussian filtering, with less partial volume effects. Absolute CBF, CBV and PS were 22%, 14% and 46% lower with 0.8 mm than with 4.8 mm slices. If the AIF and motion correction were based on thin slices prior to reconstruction of thicker slices, these differences reduced to 3%, 4% and 3%. The effect of slice thickness on relative values was very small. This study shows that thin slice reconstruction for CTP with unaltered acquisition protocol gives relative perfusion values without clinically relevant bias. It does however affect absolute perfusion values, of which CBF and CBV are most sensitive. Partial volume effects in large arteries and veins lead to overestimation of these values. The effects of reconstruction slice thickness should be taken into account when absolute perfusion values are used for clinical decision making.

In brain CT perfusion (CTP), the arterial contrast bolus is scaled to have the same area under the curve (AUC) as the venous outflow to correct for partial volume effects (PVE). This scaling is based on the assumption that large veins are unaffected by PVE. Measurement of the internal carotid artery (ICA), usually unaffected by PVE due to its large diameter, may avoid the need for partial volume correction. The aims of this work are to examine i) the assumptions behind PVE correction and ii) the potential of selecting the ICA obviating correction for PVE. The AUC of the ICA and sagittal sinus were measured in CTP datasets from 52 patients. The AUCs were determined by i) using commercial CTP software based on a Gaussian curve-fitting to the time attenuation curve, and ii) by simple integration of the time attenuation curve over a time interval. In addition, frames acquired up to 3 minutes after first bolus passage were used to examine the ratio of arterial and venous enhancement. The impact of selecting the ICA without PVE correction was illustrated by reporting cerebral blood volume (CBV) measurements. In 49 of 52 patients, the AUC of the ICA was significantly larger than that of the sagittal sinus (p = 0.017). Measured after the first pass bolus, contrast enhancement remained 50% higher in the ICA just after the first pass bolus, and 30% higher 3 minutes later. CBV measurements were significantly lowered when the ICA was used without PVE correction. Contradicting the assumptions underlying PVE correction, contrast in the ICA was significantly higher than in the sagittal sinus, even 3 minutes after the first pass of the contrast bolus. PVE correction might lead to overestimation of CBV if the CBV is calculated using the AUC of the time attenuation curves.

Accurate detection and quantification of unruptured intracranial aneurysms (UIAs) is important for rupture risk assessment and to allow an informed treatment decision to be made. Currently, 2D manual measures used to assess UIAs on Time-of-Flight magnetic resonance angiographies (TOF-MRAs) lack 3D information and there is substantial inter-observer variability for both aneurysm detection and assessment of aneurysm size and growth. 3D measures could be helpful to improve aneurysm detection and quantification but are time-consuming and would therefore benefit from a reliable automatic UIA detection and segmentation method. The Aneurysm Detection and segMentation (ADAM) challenge was organised in which methods for automatic UIA detection and segmentation were developed and submitted to be evaluated on a diverse clinical TOF-MRA dataset. A training set (113 cases with a total of 129 UIAs) was released, each case including a TOF-MRA, a structural MR image (T1, T2 or FLAIR), annotation of any present UIA(s) and the centre voxel of the UIA(s). A test set of 141 cases (with 153 UIAs) was used for evaluation. Two tasks were proposed: (1) detection and (2) segmentation of UIAs on TOF-MRAs. Teams developed and submitted containerised methods to be evaluated on the test set. Task 1 was evaluated using metrics of sensitivity and false positive count. Task 2 was evaluated using dice similarity coefficient, modified hausdorff distance (95th percentile) and volumetric similarity. For each task, a ranking was made based on the average of the metrics. In total, eleven teams participated in task 1 and nine of those teams participated in task 2. Task 1 was won by a method specifically designed for the detection task (i.e. not participating in task 2). Based on segmentation metrics, the top two methods for task 2 performed statistically significantly better than all other methods. The detection performance of the top-ranking methods was comparable to visual inspection for larger aneurysms. Segmentation performance of the top ranking method, after selection of true UIAs, was similar to interobserver performance. The ADAM challenge remains open for future submissions and improved submissions, with a live leaderboard to provide benchmarking for method developments at https://adam.isi.uu.nl/.

Purpose Early infarcts are hard to diagnose on non-contrast head CT. Dual-energy CT (DECT) may potentially increase infarct differentiation. The optimal DECT settings for differentiation were identified and evaluated. Methods One hundred and twenty-five consecutive patients who presented with suspected acute ischemic stroke (AIS) and underwent non-contrast DECT and subsequent DWI were retrospectively identified. The DWI was used as reference standard. First, virtual monochromatic images (VMI) of 25 patients were reconstructed from 40 to 140 keV and scored by two readers for acute infarct. Sensitivity, specificity, positive, and negative predictive values for infarct detection were compared and a subset of VMI energies were selected. Next, for a separate larger cohort of 100 suspected AIS patients, conventional non-contrast CT (NCT) and selected VMI were scored by two readers for the presence and location of infarct. The same statistics for infarct detection were calculated. Infarct location match was compared per vascular territory. Subgroup analyses were dichotomized by time from last-seen-well to CT imaging. Results A total of 80–90 keV VMI were marginally more sensitive (36.3–37.3%) than NCT (32.4%; p  > 0.680), with marginally higher specificity (92.2–94.4 vs 91.1%; p  > 0.509) for infarct detection. Location match was superior for VMI compared with NCT (28.7–27.4 vs 19.5%; p  < 0.010). Within 4.5 h from last-seen-well, 80 keV VMI more accurately detected infarct (58.0 vs 54.0%) and localized infarcts (27.1 vs 11.9%; p  = 0.004) than NCT, whereas after 4.5 h, 90 keV VMI was more accurate (69.3 vs 66.3%). Conclusion Non-contrast 80–90 keV VMI best differentiates normal from infarcted brain parenchyma.