Perfusion Data Research Articles

Utilizing retrograde autologous priming for blood conservation in cardiac surgery.

The use of crystalloid priming for extracorporeal circuit in adult cardiac surgery causes inevitable haemodilution. The haemodilution can be reduced by using methods such as retrograde autologous priming (RAP) with the patient's blood. This study compares the RAP technique with standard priming with regards to safety and the impact on haemodilution. This was a retrospective cohort study between a control group (n = 100) consisting of consecutive patients undergoing first time isolated coronary artery bypass surgery (CABG) with crystalloid priming solution in the circuit, and the RAP group (n = 100) consisting of patients undergoing isolated first time CABG with the RAP method. All demographics, procedure and perfusion data were gathered from the local surgical and perfusion database. Despite starting with comparable mean pre-operative haemoglobin (Hb) levels (control 127 mg/dL versus RAP 129 mg/dL), the RAP group had significantly higher mean post-op Hb level (109 mg/dL versus 92 mg/dL, P < 0.01). Crystalloid use was also significantly lower in RAP group (3.15 L versus 4.17 L P < 0.02). Freedom from red blood cell transfusion (86% versus 76% P = 0.038) and freedom from blood products (78% versus 66%, P = 0.032) was also significantly better in the RAP group. This study demonstrates that retrograde autologous priming is a safe and effective method for priming the cardiopulmonary bypass circuit in adult cardiac surgery, with significantly beneficial effects on transfusion rates and intra operative fluid requirements. Given these results the RAP method should be considered as a routine step in priming an extracorporeal circuit for adult cardiac operations.

The incremental value of PET perfusion imaging over coronary CT angiography in predicting outcome. A study using machine learning

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): Academy of Finland, Turku University Hospital Background Coronary computed tomography angiography (CTA) and myocardial perfusion imaging (MPI) are powerful non-invasive tools to evaluate the patients with suspected coronary artery disease (CAD). Purpose The goal of this study was to evaluate the incremental value of these imaging methods in predicting short- and long-term cardiac events using machine learning (ML) approaches. Methods 2411 patients with clinically suspected CAD underwent coronary CTA and subsequent positron emission tomography (PET) MPI. Following our local routine, if obstructive CAD cannot be ruled-out by coronary CTA, PET MPI using 15O-water is performed. The incremental prognostic value of PET imaging was tested using several ML models of which XGBoost models consistently outperformed others and were chosen for further analyses. XGBoost models were trained by using clinical data from medical records and coronary imaging findings: one set of models used all the available variables, while a second set of models used only clinical and CTA-based variables. Differences in the performances of the models were then used to assess the incremental value of PET perfusion variables in outcome prediction. Results After the removal of incomplete data entries, data from 2284 patients was retained for further analysis. During 8-year follow-up period 210 patients had a major cardiac event (these endpoints correspond to 59 myocardial infarctions, 35 unstable angina pectoris, and 116 deaths). The PET perfusion imaging data improved the predictive power of CTA during the first 4 years of observation time. After that, no significant difference in the predictive power was observed between the considered sets of XGBoost models, which either included or did not include PET perfusion data in the input variables. The highest area under the receiver operating characteristic curve (AUC) was at the observation time of 2 years (0.81, 95% CI 0.805–0.823) when PET data were included. The corresponding AUC when PET data were not included was at the observation time of 2 years (0.79, 95% CI 0.785–0.802). Conclusions Based on ML approach PET perfusion imaging improves the power in predicting cardiac events over anatomical CTA imaging for the first 4 years. The results illustrate the differences and complementary nature of anatomic and perfusion information in predicting outcome of patients with suspected CAD.

Diagnostic Value of Multi‐delay Multi‐parametric 3D pCASL in Early Alzheimer' Disease

AbstractBackgroundSingle post‐labeling delay (PLD) pseudo‐continuous arterial spin‐labeling (pCASL) measures cerebral blood Flow (CBF). The altered CBF is associated with cognitive decline in early Alzheimer’s disease (AD). However, multi‐delay pCASL that measures multi‐parametric measurements including CBF, arterial cerebral blood volume (aCBV) and arterial transit time (ATT) has rarely been applied to AD studies and its diagnostic performance to recognize AD in early stage remains unclear. The purpose of this study is to compare the diagnostic value between the CBF measured by single PLD pCASL and the CBF/aCBV/ATT measured by multi‐PLD pCASL for early‐stage AD.MethodBoth 1 PLD and 7 PLD 3D pCASL protocols on 3T GE scanner were performed within one MR session for 26 cognitively normal (CN), 37 mild cognitive impairment (MCI) and 39 AD subjects who were recruited in Chinese Imaging, Biomarkers, and Lifestyle Study of Ageing Project (CIBL) from April to October 2021. Comparisons on CBF/aCBV/ATT in 10 regions of interest (ROIs), including left and right regions of olfactory, post cingulate, hippocampus, cuneus and precuneus, were performed across different groups. The area under the curve (AUC) were assessed through receiver operating characteristic (ROC) analysis.ResultAll ROIs showed significant decrease in CBF and increase in ATT of AD groups compared with that of CN and MCI groups (P &lt; 0.05). There was no significant difference in aCBV between groups and CBF/ATT between CN and MCI (P &gt; 0.05). The area under the ROC curve (AUC) of CBF in ROIs for both3D pCASL protocols was similar for discriminating AD against HC (0.949 vs. 0.965 for 1 PLD vs. 7 PLD, respectively), AD against MCI (0.836 vs. 0.830), MCI against HC (0.686 vs. 0.786). However, the combination of CBF/aCBV/ATT in ROIs measured by 7 PLD 3D pCASL could significantly improve diagnostic performance, which AUC was 1.000, 0.967 and 0.918 for identifying AD against HC, AD against MCI and MCI against HC respectively.ConclusionMulti‐parametric perfusion data from multi‐delay pCASL can improve the diagnostic value of AD in early stage, which could be a promising imaging biomarker in AD patients.

Tissue Outcome Prediction in Patients with Proximal Vessel Occlusion and Mechanical Thrombectomy Using Logistic Models

Perfusion CT is established to aid selection of patients with proximal intracranial vessel occlusion for thrombectomy in the extended time window. Selection is mostly based on simple thresholding of perfusion parameter maps, which, however, does not exploit the full information hidden in the high-dimensional perfusion data. We implemented a multiparametric mass-univariate logistic model to predict tissue outcome based on data from 405 stroke patients with acute proximal vessel occlusion in the anterior circulation who underwent mechanical thrombectomy. Input parameters were acute multimodal CT imaging (perfusion, angiography, and non-contrast) as well as basic demographic and clinical parameters. The model was trained with the knowledge of recanalization status and final infarct localization. We found that perfusion parameter maps (CBF, CBV, and Tmax) were sufficient for tissue outcome prediction. Compared with single-parameter thresholding-based models, our logistic model had comparable volumetric accuracy, but was superior with respect to topographical accuracy (AUC of receiver operating characteristic). We also found higher spatial accuracy (Dice index) in an independent internal but not external cross-validation. Our results highlight the value of perfusion data compared with non-contrast CT, CT angiography and clinical information for tissue outcome-prediction. Multiparametric logistic prediction has high potential to outperform the single-parameter thresholding-based approach. In the future, the combination of tissue and functional outcome prediction might provide an individual biomarker for the benefit from mechanical thrombectomy in acute stroke care.

Stress-First Myocardial Perfusion Imaging

Stress-first approaches to myocardial perfusion imaging provide diagnostically and prognostically accurate perfusion data equivalent to a full rest-stress study while saving time in the imaging laboratory and reducing the radiation exposure to patients and laboratory staff. Unfortunately, implementing a stress-first approach in a nuclear cardiology laboratory involves significant challenges such as the need for attenuation correction, triage of patients to an appropriate protocol, real-time review of stress images, and consideration of differential reimbursement. Despite it being best practice for both the patient and the laboratory, these impediments have kept the proportions of studies performed stress-first relatively unchanged in North America and world-wide in the last 10years.

Safety and effectiveness of nano composite hydrogel stent implantation in the treatment of coronary cardiovascular disease: A preclinical study

With the improvement of people's quality of life, various cardiovascular diseases are the most common diseases. Therefore, the main site of disease atherosclerosis is blood vessels, so we can see that its flow rate has obvious changes. Through the analysis of coronary heart disease, this paper studies the relationship between coronary artery disease and cardiovascular disease, which is helpful to evaluate the risk of disease, and also provides the best prevention and treatment plan to overcome cardiovascular disease. As the material of artificial cartilage repair, nanocomposite hydrogel has excellent application value and attraction, because nanocomposite hydrogel has a structure similar to the extracellular matrix of natural chondrocytes. The patients in the experimental group were treated with nano composite hydrogel stent implantation. The other group of patients used the traditional way to carry out the comparative experiment. In the perfusion data of each ventricular wall in the coronary angiography and anterior wall perfusion group, the percentage of lateral wall in the normal proportion was the highest, 69.2%, 59.3% in the anterior wall, 39.5% in the inferior wall, and 19.7% in the apical value and interval. The percentage of LAD stenosis in anterior wall perfusion was O. The highest percentage in the lateral wall was 69.2%, and the lowest in the septum and apex was 19.7%. Nanocomposite hydrogel stent implantation can effectively treat coronary heart disease. The research shows that it is safe and effective in application.

Automated occlusion detection for the diagnosis of acute ischemic stroke: A detailed performance review

IntroductionStroke is a leading cause of adult disability and death worldwide. Automated detection of stroke on brain imaging has promise in a time critical environment. We present a method for the automated detection of intracranial occlusions on dynamic CT Angiography (CTA) causing acute ischemic stroke. MethodsWe derived dynamic CTA images from CT Perfusion (CTP) data and utilised advanced image processing to enhance and display major cerebral blood vessels for symmetry analysis. We reviewed the performance of the algorithm on a cohort of 207 patients from the International Stroke Perfusion Imaging Registry (INSPIRE), with Large Vessel Occlusion (LVO) and non-LVO strokes. Included in the data were images with chronic stroke, various artefacts, incomplete vessel occlusions, and images of poorer quality. All images were annotated by stroke experts. In addition, each image was graded in terms of the difficulty of the task of occlusion detection. Performance was evaluated on the overall cohort, and with respect to occlusion location, collateral grade, and task difficulty. We also evaluated the impact of including additional perfusion data. ResultsImages with a rating of lower difficulty achieved a sensitivity and specificity of 96% and 90%, respectively, while images with a moderate difficulty rating achieved 88% and 50%, respectively. For cases of high difficulty, where more than two experts or additional data were required to reach consensus, sensitivity and specificity was 53% and 11%. The addition of perfusion data to the dCTA images increased the specificity by 38%. ConclusionWe have provided an unbiased interpretation of algorithm performance. Further developments include generalising to conventional CTA and employing the algorithm in a clinical setting for prospective studies.

Using Laser Speckle Contrast Imaging to Quantify Perfusion Quality in Kidney and Pancreas Grafts on Vascular Reperfusion: A Proof-of-Principle Study.

This study was registered in ClinicalTrials.gov (NCT04202237). The Moor FLPI-2 blood flow imager was used in 4 patients (1 Simultaneous Pancreas and Kidney, 2 deceased, and 1 living donor kidney transplants) during reperfusion to capture reperfusion data. The following parameters were measured: flux (average speed × concentration of moving red blood cells in the sample volume), doppler centroid, total and valid pixels, valid rate, and total and valid area. Flux data were analyzed with Moor FLPI analysis software. The perfusion characteristics and flux images correlated with initial graft function. LSCI is a safe, noncontact imaging modality that provides real-time, accurate, high-resolution, full field blood flow images and a wide range of flux data to objectively quantify organ reperfusion intraoperatively in kidney/kidney-pancreas transplantation. This modality could be used to develop a robust numerical quantification system for the evaluation and reporting of intraoperative organ perfusion, and aid intraoperative decision-making. Perfusion data could be combined with biomarkers and immunological parameters to more accurately predict graft outcomes.

A proof-of-concept pipeline to guide evaluation of tumor tissue perfusion by dynamic contrast-agent imaging: Direct simulation and inverse tracer-kinetic procedures.

Dynamic contrast-enhanced (DCE) perfusion imaging has shown great potential to non-invasively assess cancer development and its treatment by their characteristic tissue signatures. Different tracer kinetics models are being applied to estimate tissue and tumor perfusion parameters from DCE perfusion imaging. The goal of this work is to provide an in silico model-based pipeline to evaluate how these DCE imaging parameters may relate to the true tissue parameters. As histology data provides detailed microstructural but not functional parameters, this work can also help to better interpret such data. To this aim in silico vasculatures are constructed and the spread of contrast agent in the tissue is simulated. As a proof of principle we show the evaluation procedure of two tracer kinetic models from in silico contrast-agent perfusion data after a bolus injection. Representative microvascular arterial and venous trees are constructed in silico. Blood flow is computed in the different vessels. Contrast-agent input in the feeding artery, intra-vascular transport, intra-extravascular exchange and diffusion within the interstitial space are modeled. From this spatiotemporal model, intensity maps are computed leading to in silico dynamic perfusion images. Various tumor vascularizations (architecture and function) are studied and show spatiotemporal contrast imaging dynamics characteristic of in vivo tumor morphotypes. The Brix II also called 2CXM, and extended Tofts tracer-kinetics models common in DCE imaging are then applied to recover perfusion parameters that are compared with the ground truth parameters of the in silico spatiotemporal models. The results show that tumor features can be well identified for a certain permeability range. The simulation results in this work indicate that taking into account space explicitly to estimate perfusion parameters may lead to significant improvements in the perfusion interpretation of the current tracer-kinetics models.

Perfusion Imaging Mismatch Profiles in the Early Thrombectomy Window: A Single-Center Analysis.

Little is known on the role of mismatch profile in patients undergoing early endovascular treatment (EVT). We aimed to describe pretreatment perfusion parameters and mismatch profiles in anterior circulation large vessel occlusion acute ischemic stroke undergoing EVT in the early time window and assess their association with time from stroke onset and outcomes. Retrospective single-center study, including early (<6 hours) EVT-treated large vessel occlusion acute ischemic stroke with baseline perfusion data, assessing perfusion parameters (ischemic core volume, mismatch volume and mismatch ratio) and mismatch profiles (favorable versus unfavorable, based on criteria adopted in EXTEND-IA [Extending the Time for Thrombolysis in Emergency Neurological Deficits - Intra-Arterial], SWIFT PRIME [Solitaire With the Intention for Thrombectomy as Primary Endovascular Treatment], DEFUSE 3 [Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3], and DAWN [Clinical Mismatch in the Triage of Wake Up and Late Presenting Strokes Undergoing Neurointervention With Trevo] trials). We evaluated their association with time from stroke onset (rs [for parameters] or χ2 for trend [for profiles]) and association with modified Rankin Scale score >2, symptomatic intracranial hemorrhage, and mortality (multivariate regression analyses [each parameter/profile entered into a separate logistic regression model, adjusted for baseline variables associated with each outcome in the univariate analysis at the P<0.1 level]). Among 357 patients, unfavorable mismatch profiles ranged from 21% to 60%, depending on the criterion, and were not correlated with time from stroke onset (P=0.490). All individual perfusion parameters and unfavorable mismatch profiles were associated with poor functional outcome: ischemic core volume adjusted odds ratio (aOR), 1.49 ([95% CI, 1.13-1.97] P=0.005); penumbral volume aOR, 0.30 ([95% CI, 0.10-0.84] P=0.022); mismatch ratio aOR, 0.67 ([95% CI, 0.50-0.90] P=0.007); EXTEND-IA aOR, 2.61 ([95% CI, 1.23-5.51] P=0.012); SWIFT PRIME aOR, 2.50 ([95% CI, 1.30-4.57] P=0.006); DEFUSE 3 aOR, 2.28 ([95% CI, 1.14-4.57] P=0.020); and DAWN aOR, 4.19 ([95% CI, 2.13-8.26] P<0.001). EXTEND-IA and DEFUSE 3 unfavorable profiles were also independently associated with symptomatic intracranial hemorrhage (aOR, 3.82 [95% CI, 1.42-10.3]; P=0.008 and aOR, 2.83 [95% CI, 1.09-7.36]; P=0.033) and death (aOR, 3.26 [95% CI, 1.33-8.02]; P=0.010 and aOR, 2.52 [95% CI, 1.10-5.82]; P=0.030). Pretreatment perfusion parameters and mismatch profiles in early EVT-treated patients were not correlated with time from stroke onset but were independently associated with functional outcome. Mismatch assessment in the early time window may improve EVT patient selection, independently of onset-to-treatment time.

Application of perfusion computed tomography in renal diseases (review of literature)

Purpose. To analyze the literature data on the use of CT perfusion in kidney diseases and to assess the future prospects of using the technique in clinical practice.Materials and methods. In electronic databases (PubMed, E-library, Web of Science, Google Scholar), a search was conducted for published studies evaluating the possibilities of using CT perfusion in both neoplastic and non-neoplastic kidney diseases. The article analyzes the results of 40 most relevant works of Russian and foreign researchers devoted to this topic.Results. According to the analysis of the data obtained, perfusion CT is an effective diagnostic tool in oncology: the technique allows noninvasively assessing the nature of the tumour, including differentiating benign nodes (fat-poor angiomyolipoma and oncocytoma) from renal cell carcinoma; to establish the histological variant of renal cell carcinoma and Fuhrman grade, to characterize the effectiveness of ablative techniques and systemic treatment of renal cell carcinoma. Based on the correlation of CT kidney perfusion data and the results of various methods for determining organ function, the possibility of using perfusion CT as one of the prognostic factors for determining the tactics of treatment of patients with obstructive uropathies, aortomesenteric compression, and also shows the potential of using the technique in transplantology both in patients after surgery and during the examination of donors.Conclusions. Despite the fact that the role of CT kidney perfusion in various fields of urology and nephrology has been sufficiently studied, some important aspects of the likely application of this technique remain underestimated. Taking into account the high incidence rates and a significant percentage of localized forms of tumors, the study of the role of CT perfusion in planning and evaluating the results of nephron-sparing treatment of renal cell carcinoma may open up new prospects in optimizing surgical tactics.

Application of CT functional imaging in the assessment of chemotherapy efficacy in lung cancer

ABSTRACT In this research, we aim to investigate the feasibility of a one-stop CT energy spectrum perfusion imaging technique for chemotherapy efficacy assessment of lung cancer patients by obtaining both functional imaging parameters of energy spectrum and perfusion in one scan. From November 2018 to February 2020, a group of 23 patients with pathologically confirmed lung cancer were chosen to undergo CT energy spectrum scans both before and after treatment. The post-treatment CT perfusion data was acquired one week after the second conventional chemotherapy session. Out of the 23 patients, 15 were in the chemotherapy effective group and the remaining 8 were in the ineffective group. The reason for this group was according to recist criteria. Arterial phase iodine concentration (icap) and intravenous phase iodine concentration (icpp) of the lesions were measured, and standardized iodine base values (nic) were calculated. The maximum diameter of the tumor before and after treatment was compared to the perfusion parameters and energy spectrum parameters before and after chemotherapy in the effective group and the invalid group was compared by two tests that p<0.05. The differences between the maximum diameter of the tumor before and after chemotherapy. 2 of the 15 patients in the effective group had liquefied necrotic areas in their lesions. One-stop ct energy-spectrum perfusion imaging can show the disease progression from a functional perspective and assess the efficacy early according to the changes in perfusion parameters and energy-spectrum parameters after lung cancer treatment.

Transplacental passage of hyperforin, hypericin, and valerenic acid.

Safe medications for mild mental diseases in pregnancy are needed. Phytomedicines from St. John's wort and valerian are valid candidates, but safety data in pregnancy are lacking. The transplacental transport of hyperforin and hypericin (from St. John's wort), and valerenic acid (from valerian) was evaluated using the ex vivo cotyledon perfusion model (4h perfusions, term placentae) and, in part, the in vitro Transwell assay with BeWo b30 cells. Antipyrine was used for comparison in both models. U(H)PLC-MS/MS bioanalytical methods were developed to quantify the compounds. Perfusion data obtained with term placentae showed that only minor amounts of hyperforin passed into the fetal circuit, while hypericin did not cross the placental barrier and valerenic acid equilibrated between the maternal and fetal compartments. None of the investigated compounds affected metabolic, functional, and histopathological parameters of the placenta during the perfusion experiments. Data from the Transwell model suggested that valerenic acid does not cross the placental cell layer. Taken together, our data suggest that throughout the pregnancy the potential fetal exposure to hypericin and hyperforin - but not to valerenic acid - is likely to be minimal.

Spatial CT perfusion data helpful in automatically locating vessel occlusions for acute ischemic stroke patients.

Locating a vessel occlusion is important for clinical decision support in stroke healthcare. The advent of endovascular thrombectomy beyond proximal large vessel occlusions spurs alternative approaches to locate vessel occlusions. We explore whether CT perfusion (CTP) data can help to automatically locate vessel occlusions. We composed an atlas with the downstream regions of particular vessel segments. Occlusion of these segments should result in the hypoperfusion of the corresponding downstream region. We differentiated between seven-vessel occlusion locations (ICA, proximal M1, distal M1, M2, M3, ACA, and posterior circulation). We included 596 patients from the DUtch acute STroke (DUST) multicenter study. Each patient CTP data set was processed with perfusion software to determine the hypoperfused region. The downstream region with the highest overlap with the hypoperfused region was considered to indicate the vessel occlusion location. We assessed the indications from CTP against expert annotations from CTA. Our atlas-based model had a mean accuracy of 86% and could achieve substantial agreement with the annotations from CTA according to Cohen's kappa coefficient (up to 0.68). In particular, anterior large vessel occlusions and occlusions in the posterior circulation could be located with an accuracy of 80 and 92%, respectively. The spatial layout of the hypoperfused region can help to automatically indicate the vessel occlusion location for acute ischemic stroke patients. However, variations in vessel architecture between patients seemed to limit the capacity of CTP data to distinguish between vessel occlusion locations more accurately.

Longitudinal Lung Function Assessment of Patients Hospitalized With COVID-19 Using 1H and 129Xe Lung MRI

Microvascular abnormalities and impaired gas transfer have been observed in patients with COVID-19. The progression of pulmonary changes in these patients remains unclear. Do patients hospitalized with COVID-19 without evidence of architectural distortion on structural imaging exhibit longitudinal improvements in lung function measured by using 1H and 129Xe MRI between 6 and 52weeks following hospitalization? Patients who were hospitalized with COVID-19 pneumonia underwent a pulmonary 1H and 129Xe MRI protocol at 6, 12, 25, and 51weeks following hospital admission in a prospective cohort study between November 2020 and February 2022. The imaging protocol was as follows: 1H ultra-short echo time, contrast-enhanced lung perfusion, 129Xe ventilation, 129Xe diffusion-weighted, and 129Xe spectroscopic imaging of gas exchange. Nine patients were recruited (age 57 ± 14 [median ± interquartile range] years; six of nine patients were male). Patients underwent MRI at 6 (n= 9), 12 (n= 9), 25 (n= 6), and 51 (n= 8) weeks following hospital admission. Patients with signs of interstitial lung damage were excluded. At 6weeks, patients exhibited impaired 129Xe gas transfer (RBC to membrane fraction), but lung microstructure was not increased (apparent diffusion coefficient and mean acinar airway dimensions). Minor ventilation abnormalities present in four patients were largely resolved in the 6- to 25-week period. At 12weeks, all patients with lung perfusion data (n= 6) showed an increase in both pulmonary blood volume and flow compared with 6weeks, although this was not statistically significant. At 12weeks, significant improvements in 129Xe gas transfer were observed compared with 6-week examinations; however, 129Xe gas transfer remained abnormally low at weeks 12, 25, and51. 129Xe gas transfer was impaired up to 1 year following hospitalization in patients who were hospitalized with COVID-19 pneumonia, without evidence of architectural distortion on structural imaging, whereas lung ventilation was normal at 52weeks.

Neural network-derived perfusion maps: A model-free approach to computed tomography perfusion in patients with acute ischemic stroke.

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.

Automated identification of myocardial perfusion defects in dynamic cardiac computed tomography using deep learning

PurposeThe purpose of this study was to develop and evaluate deep convolutional neural network (CNN) models for quantifying myocardial blood flow (MBF) as well as for identifying myocardial perfusion defects in dynamic cardiac computed tomography (CT) images. MethodsAdenosine stress cardiac CT perfusion data acquired from 156 patients having or being suspected with coronary artery disease were considered for model development and validation. U-net-based deep CNN models were developed to segment the aorta and myocardium and to localize anatomical landmarks. Color-coded MBF maps were obtained in short-axis slices from the apex to the base level and were used to train a deep CNN classifier. Three binary classification models were built for the detection of perfusion defect in the left anterior descending artery (LAD), the right coronary artery (RCA), and the left circumflex artery (LCX) territories. ResultsMean Dice scores were 0.94 (±0.07) and 0.86 (±0.06) for the aorta and myocardial deep learning-based segmentations, respectively. With the localization U-net, mean distance errors were 3.5 (±3.5) mm and 3.8 (±2.4) mm for the basal and apical center points, respectively. The classification models identified perfusion defects with the accuracy of mean area under the receiver operating curve (AUROC) values of 0.959 (±0.023) for LAD, 0.949 (±0.016) for RCA, and 0.957 (±0.021) for LCX. ConclusionThe presented method has the potential to fully automate the quantification of MBF and subsequently identify the main coronary artery territories with myocardial perfusion defects in dynamic cardiac CT perfusion.

Machine learning segmentation of core and penumbra from acute stroke CT perfusion data.

Computed tomography perfusion (CTP) imaging is widely used in cases of suspected acute ischemic stroke to positively identify ischemia and assess suitability for treatment through identification of reversible and irreversible tissue injury. Traditionally, this has been done via setting single perfusion thresholds on two or four CTP parameter maps. We present an alternative model for the estimation of tissue fate using multiple perfusion measures simultaneously. We used machine learning (ML) models based on four different algorithms, combining four CTP measures (cerebral blood flow, cerebral blood volume, mean transit time and delay time) plus 3D-neighborhood (patch) analysis to predict the acute ischemic core and perfusion lesion volumes. The model was developed using 86 patient images, and then tested further on 22 images. XGBoost was the highest-performing algorithm. With standard threshold-based core and penumbra measures as the reference, the model demonstrated moderate agreement in segmenting core and penumbra on test images. Dice similarity coefficients for core and penumbra were 0.38 ± 0.26 and 0.50 ± 0.21, respectively, demonstrating moderate agreement. Skull-related image artefacts contributed to lower accuracy. Further development may enable us to move beyond the current overly simplistic core and penumbra definitions using single thresholds where a single error or artefact may lead to substantial error.

A deep learning approach to predict collateral flow in stroke patients using radiomic features from perfusion images.

Collateral circulation results from specialized anastomotic channels which are capable of providing oxygenated blood to regions with compromised blood flow caused by arterial obstruction. The quality of collateral circulation has been established as a key factor in determining the likelihood of a favorable clinical outcome and goes a long way to determining the choice of a stroke care model. Though many imaging and grading methods exist for quantifying collateral blood flow, the actual grading is mostly done through manual inspection. This approach is associated with a number of challenges. First, it is time-consuming. Second, there is a high tendency for bias and inconsistency in the final grade assigned to a patient depending on the experience level of the clinician. We present a multi-stage deep learning approach to predict collateral flow grading in stroke patients based on radiomic features extracted from MR perfusion data. First, we formulate a region of interest detection task as a reinforcement learning problem and train a deep learning network to automatically detect the occluded region within the 3D MR perfusion volumes. Second, we extract radiomic features from the obtained region of interest through local image descriptors and denoising auto-encoders. Finally, we apply a convolutional neural network and other machine learning classifiers to the extracted radiomic features to automatically predict the collateral flow grading of the given patient volume as one of three severity classes - no flow (0), moderate flow (1), and good flow (2). Results from our experiments show an overall accuracy of 72% in the three-class prediction task. With an inter-observer agreement of 16% and a maximum intra-observer agreement of 74% in a similar experiment, our automated deep learning approach demonstrates a performance comparable to expert grading, is faster than visual inspection, and eliminates the problem of grading bias.

CT perfusion extended window ischemic core estimation: Bayesian algorithm versus oscillation index singular value decomposition.

Ischemic core estimation by CT perfusion (CTp) is a diagnostic challenge, mainly because of the intrinsic noise associated with perfusion data. However, an accurate and reliable quantification of the ischemic core is critical in the selection of patients for reperfusion therapies. Our study aimed at assessing the diagnostic accuracy of two different CTp postprocessing algorithms, that is, the Bayesian Method and the oscillation index singular value decomposition (oSVD). All the consecutive stroke patients studied in the extended time window (>4.5 hours from stroke onset) by CTp and diffusion-weighted imaging (DWI), between October 2019 and December 2021, were enrolled. The agreement between both algorithms and DWI was assessed by the Bland-Altman plot, Wilcoxon signed-rank test, Spearman's rank correlation coefficient, and the intraclass correlation coefficient (ICC). Twenty-four patients were enrolled (average age: 72±15years). The average National Institutes of Health Stroke Scale was 14.42±6.75, the median Alberta Stroke Program Early CT score was 8.50 (interquartile range [IQR]=7.75-9), and median time from stroke onset to neuroimaging was 7.5 hours (IQR=6.5-8). There was an excellent correlation between DWI and oSVD (ρ=.87, p-value<.001) and DWI and Bayesian algorithm (ρ=.94, p-value<.001). There was a stronger ICC between DWI and Bayesian algorithm (.97, 95% confidence interval [CI]: .92-.99, p-value<.001) than between DWI and oSVD (.59, 95% CI: .26-.8, p-value<.001). The agreement between Bayesian algorithm and DWI was greater than between oSVD and DWI in the extended window. The more accurate estimation of the ischemic core offered by the Bayesian algorithm may well play a critical role in the accurate selection of patients for reperfusion therapies.