Classification Of Stenosis Research Articles

THE ROLE OF MAGNETIC RESONANCE IMAGING IN THE DIAGNOSIS AND TREATMENT ORIENTATION OF CERVICAL SPINAL STENOSIS SYNDROMES

Objectives: To study on the magnetic resonance imaging (MRI) to diagnose the causes of cervical spinal stenosis and contribution to therapeutic orientations in cervical spinal stenosis patients. Materials and methods: Descriptive cross-sectional studies of 55 consecutive patients with cervical canal stenosis on magnetic resonance imaging in Radiology department of Hue University Hospital of Medicine and Pharmacy. Results: The mean age of the patients was 49.6 ± 11.2, and 61.8% of cases were males. Age group 40-60 was highest ratio (67.3%). The cervical spine pain was the most common symptom of cervical canal stenosis. Abnormal reflexes of upper limb tendons were one of the more common symptoms (72.7%). The combined presentation of all three syndromes: cervical spine pain, cervical root compression, and cervical cord compression that were 54.5%. The cervical herniated disc (92.7%) is the most common cause of acquired cervical spinal stenosis. According to classical classification of cervical spinal stenosis, 52.7% of patients with absolute stenosis and 47.3% of patients with relative stenosis. According to the Kang classification in 2011, grade 1 stenosis is 50.9%, grade 2 stenosis is 36.4% and grade 3 stenosis 12.7%. Conclusion: There was a concordance between the clinical presentation and imaging manifestation of cervical canal stenosis. The concordance between the cervical canal stenosis by Kang grading system and clinical symptoms by Nurickscore were more than average with Kappa = 0.58. Therefore, it is necessary to determine the grade of injury on magnetic resonance imaging in combination with clinical symptoms to choose a reasonable treatment method for patients with acquired cervical spinal stenosis. Key words: Magnetic resonance imaging, cervical spine pain, cervical canal stenosis

The AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Extracranial Cerebrovascular System.

The AIUM Practice Parameter for the Performance of an Ultrasound Examination of the Extracranial Cerebrovascular System.

Abstract 11711: Quantification of Atherosclerotic Coronary Artery Disease in the Pre/Peri-Menopausal Female Using Cardiac CTA

Introduction: Coronary Artery Disease (CAD) occurs equally in males and females and is the most common cause of death in both genders worldwide. The pre/peri-menopausal female is believed to have minimal coronary atherosclerosis based primarily on non-invasive techniques, with limited coronary angiography data. Cardiac computed tomography angiography (CCTA) is a more sensitive, non-invasive anatomical diagnostic tool for quantifying coronary atherosclerosis, but pre/peri-menopausal women have been underrepresented and underdiagnosed. The goal of this study was to quantify coronary atherosclerosis in this population. Methods: We conducted a retrospective chart review on 510 females aged 35-60 presenting with atypical chest pain and shortness of breath. These were consecutive admissions in which CCTA was performed from January 2014 to December 2019. The degree of coronary atherosclerosis was quantified using the CAD-RADS scoring system. Results: Analysis showed 68.82% of patients had no evidence of coronary atherosclerosis. 4.31% had minimal CAD, 18.24% had mild CAD, 6.47% had moderate CAD and 2.16% were found to have severe CAD. Each category of stenosis typically had multiple vessels involved. Over 65% of patients were Non-Hispanic White, 27% Hispanic, 8% African American and 4% Asian. Less than 5% of patients were discharged with statin therapy. Conclusions: This is the first study to quantify coronary atherosclerosis in the pre/peri-menopausal female. Quantification was based on coronary angiograms obtained by CCTA, currently the most sensitive diagnostic imaging technique. Despite the perceived protective nature of the premenopausal state in women, 30% of patients had detectable CAD by CCTA. 10% had greater than 50% stenosis (moderate to severe CAD), suggesting the potential for future ischemic heart disease. So few patients were discharged on statin therapy, showing the paucity of guideline therapy used in this population. These results suggest targeting preventative therapy, even late in the pre or peri-menopausal stage, such as favorable lifestyle changes and reducing known modifiable cardiac risk factors, should significantly prevent the progression of coronary atherosclerosis and its subsequent sequelae in menopause.

Abstract 13587: Long-Term Outcomes of CT-Derived Fractional Flow Reserve in the Emergency Department

Introduction: In acute chest pain (ACP) patients, CT-derived fraction flow reserve (FFRCT) is feasible and adjudicates need for immediate coronary revascularization. There are no long-term outcome studies of FFRCT in ACP. Hypothesis: In ACP patients, FFRCT is safe and associated with favorable outcomes when negative. Methods: ACP patients with no known coronary artery disease or definitive acute coronary syndrome undergoing coronary CTA with FFRCT were compared to those having CTA alone for the composite major adverse cardiac event (MACE) endpoint of death, nonfatal myocardial infarction, unplanned percutaneous (PCI) or surgical (CABG) revascularization over 2 years. FFRCT ≤ 0.80 was considered positive for hemodynamically significant stenosis. Results: From September 2015 to July 2018, 2,568 ACP patients (mean age 55 years, 52% female) underwent CTA of which 236 (10%) had FFRCT (156 negative, 80 positive). Of the total, 1629 (63.2%) had normal coronaries, and 43 (1.7%) were uninterpretable. FFRCT patients were older (p<0.001) with more diabetes mellitus (18.2% vs. 12.2%, p = 0.009), hypertension (50.8% vs. 38.2%, p<0.001), hyperlipidemia (43.6% vs. 25.4%, p<0.001), and >25% stenosis on CTA (93% vs. 13.6%, p<0.001). Positive FFRCT was associated with higher MACE (10% vs. 2.1%, p<0.001). With comparable stenosis on CTA, positive FFRCT had higher MACE compared to negative FFRCT or CTA alone (Table 1). Conclusions: In ACP patients, negative FFRCT and CTA alone are comparable for MACE. Positive FFRCT portends higher MACE at 2 years within similar stenosis categories. FFRCT can provide incremental value to ACP evaluation with coronary CTA.

Abstract 13539: CT-Derived Fractional Flow Reserve in Patients With Non-ST Elevation Myocardial Infarction

Introduction: In patients with non-ST Elevation Myocardial Infarction (NSTEMI), coronary CTA has high diagnostic accuracy to diagnose obstructive coronary artery disease (CAD). The value of CT-derived fractional flow reserve (FFRCT) in NSTEMI patients is unclear. Hypothesis: In acute chest pain (ACP) patients diagnosed with an NSTEMI, FFRCT is feasible, safe and adjudicates the need for immediate revascularization. Methods: In ACP patients with NSTEMI and no prior coronary artery disease, those undergoing coronary CTA with FFRCT were compared to those having CTA alone for the composite major adverse cardiac event (MACE) endpoint of death, nonfatal myocardial infarction, unplanned percutaneous (PCI) or surgical (CABG) revascularization over 2 years. NSTEMI was defined as peak troponin I levels ≥ 0.04 ng/mL. FFRCT ≤ 0.80 was considered positive for hemodynamically significant stenosis. Results: From September 2015 to July 2018, 276 ACP patients with NSTEMI (mean age 59 years, 45% female) underwent CTA of which 35 (13%) had FFRCT (14 negative, 21 positive). Of the total, 126 (45.6%) had normal coronaries and 11 (4.0%) were uninterpretable. FFRCT patients were older (p<0.001) with no difference in diabetes (14.3% vs. 14.1%, p=1.0), hypertension (57.1% vs. 49.4%, p=0.391), hyperlipidemia (34.3% vs. 28.2%, p=0.460) and mean troponin I levels (0.49 ± 0.9 ng/mL vs. 1.10 ± 4.58 ng/mL, p = 0.061). All FFRCT patients had >25% stenosis on CTA (100% vs. 42%, p<0.001). Among patients with comparable stenosis on CTA, FFRCT and CTA had equivalent MACE (5.7% vs. 18.5%, p=0.137), with no difference between FFRCT negative and positive patients (Table 1). Conclusions: In this small study, FFRCT did not add incremental value to CTA in patients with NSTEMI across similar stenosis categories with respect to MACE. Larger prospective studies are needed to examine FFRCT in NSTEMI.

Respiration-related variations in Pd/Pa ratio and fractional flow reserve in resting conditions and during intravenous adenosine administration.

AimsWe evaluated the occurrence and physiology of respiration‐related beat‐to‐beat variations in resting Pd/Pa and FFR during intravenous adenosine administration, and its impact on clinical decision‐making.Methods and ResultsCoronary pressure tracings in rest and at plateau hyperemia were analyzed in a total of 39 stenosis from 37 patients, and respiratory rate was calculated with ECG‐derived respiration (EDR) in 26 stenoses from 26 patients. Beat‐to‐beat variations in FFR occurred in a cyclical fashion and were strongly correlated with respiratory rate (R2 = 0.757, p < 0.001). There was no correlation between respiratory rate and variations in resting Pd/Pa. When single‐beat averages were used to calculate FFR, mean ΔFFR was 0.04 ± 0.02. With averaging of FFR over three or five cardiac cycles, mean ΔFFR decreased to 0.02 ± 0.02, and 0.01 ± 0.01, respectively. Using a FFR ≤ 0.80 threshold, stenosis classification changed in 20.5% (8/39), 12.8% (5/39) and 5.1% (2/39) for single‐beat, three‐beat and five‐beat averaged FFR. The impact of respiration was more pronounced in patients with pulmonary disease (ΔFFR 0.05 ± 0.02 vs 0.03 ± 0.02, p = 0.021).ConclusionBeat‐to‐beat variations in FFR during plateau hyperemia related to respiration are common, of clinically relevant magnitude, and frequently lead FFR to cross treatment thresholds. A five‐beat averaged FFR, overcomes clinically relevant impact of FFR variation.

Biological knowledge-slanted random forest approach for the classification of calcified aortic valve stenosis

BackgroundCalcific aortic valve stenosis (CAVS) is a fatal disease and there is no pharmacological treatment to prevent the progression of CAVS. This study aims to identify genes potentially implicated with CAVS in patients with congenital bicuspid aortic valve (BAV) and tricuspid aortic valve (TAV) in comparison with patients having normal valves, using a knowledge-slanted random forest (RF).ResultsThis study implemented a knowledge-slanted random forest (RF) using information extracted from a protein-protein interactions network to rank genes in order to modify their selection probability to draw the candidate split-variables. A total of 15,191 genes were assessed in 19 valves with CAVS (BAV, n = 10; TAV, n = 9) and 8 normal valves. The performance of the model was evaluated using accuracy, sensitivity, and specificity to discriminate cases with CAVS. A comparison with conventional RF was also performed. The performance of this proposed approach reported improved accuracy in comparison with conventional RF to classify cases separately with BAV and TAV (Slanted RF: 59.3% versus 40.7%). When patients with BAV and TAV were grouped against patients with normal valves, the addition of prior biological information was not relevant with an accuracy of 92.6%.ConclusionThe knowledge-slanted RF approach reflected prior biological knowledge, leading to better precision in distinguishing between cases with BAV, TAV, and normal valves. The results of this study suggest that the integration of biological knowledge can be useful during difficult classification tasks.

Surgical repair of a special category of arteriovenous fistula outflow stenosis caused by venous valve hyperplasia.

This study evaluated a special category of arteriovenous fistula outflow stenosis caused by venous valve hyperplasia and explored the effectiveness of surgical repair in dealing with this kind of stenosis. This retrospective cohort study was conducted from February 2016 to January 2020 in our center. Patients with arteriovenous fistula dysfunction, including flow rate insufficiency, venous hypertension, thrombosis, and aneurysm dilation enlargement, were selected. Stenosis lesions presenting with venous valve hyperplasia were selected after ultrasound screening. All patients underwent surgical repair and were followed up every 6months after surgery. Forty-three patients (median age, 54.5 ± 11.2years; 65.1% men) were included. All procedures were technically successful. Based on intraoperative exploration, 56.5% were reconstructed via autologous vein patch, 17.4% of patients were reconstructed with end-to-end reconstruction after cutting the stenotic segment, 13.0% of cases simply had the valve resected, and 13.0% of cases involved a longitudinal incision and transverse suture. All patients returned to routine dialysis the following day and avoided catheter insertion. The mean follow-up time was 22.5 ± 14.0 (range, 1.3-49.8) months. The patency rates at 2 and 4years were 92.2% and 79.0%, respectively. Valves harvested from patients were analyzed via Masson staining and immunohistochemical staining, indicating collagen fiber and myofibroblast hyperplasia in outflow venous valve hyperplasia (OVVH). Outflow venous valve hyperplasia can lead to fistula dysfunction. Ultrasound is the main method to diagnosis OVVH. Special surgical repair can preserve valuable vascular resources and relieve stenosis, is safe and effective, and has a high patency rate.

Spinal lumbar stenosis (analysis of the literature and own results)

The article is devoted to the issues of etiology, pathophysiology and clinical manifestations of spinal stenosis of the lumbar spine. The possibilities of modern instrumental diagnostics of the disease are described, the differential clinical diagnosis of lumbar stenosis is presented. The paper presents a detailed classification of spinal stenosis. The modern therapeutic tactics of the disease are described, indications for the surgical treatment of the disease are given. Modern methods of surgical treatment of stenosis are briefly and schematically described: decompression (microsurgical and endoscopic), interspinous fixation, stabilization (dorsal and ventral) and epidural stimulation of the spinal cord. The data on the immediate and long-term results of surgical methods of treatment are presented. The authors also presented their experience in the treatment of spinal stenosis of the lumbar spine. The use of differentiated surgical treatment tactics in 85–90 % of cases provides good and excellent results in the treatment of lumbar stenosis.

Value of semiquantitative assessment of high-risk plaque features on coronary CT angiography over stenosis in selection of studies for FFRct

IntroductionThe degree of stenosis on coronary CT angiography (CCTA) guides referral for CT-derived flow reserve (FFRct). We sought to assess whether semiquantitative assessment of high-risk plaque (HRP) features on CCTA improves selection of studies for FFRct over stenosis assessment alone. MethodsPer-vessel FFRct was computed in 1,395 vessels of 836 patients undergoing CCTA with 25–99% maximal stenosis. By consensus analysis, stenosis severity was graded as 25–49%, 50–69%, 70–89%, and 90–99%. HRPs including low attenuation plaque (LAP), positive remodeling (PR), and spotty calcification (SC) were assessed in lesions with maximal stenosis. Lesion FFRct was measured distal to the lesion with maximal stenosis, and FFRct<0.80 was defined as abnormal. Association of HRP and abnormal lesion FFRct was evaluated by univariable and multivariable logistic regression models. ResultsThe frequency of abnormal lesion FFRct increased with increase of stenosis severity across each stenosis category (25–49%:6%; 50–69%:30%; 70–89%:54%; 90–99%:91%, p ​< ​0.001). Univariable analysis demonstrated that stenosis severity, LAP, and PR were predictive of abnormal lesion FFRct, while SC was not. In multivariable analyses considering stenosis severity, presence of PR, LAP, and PR and/or LAP were independently associated with abnormal FFRct: Odds ratio 1.58, 1.68, and 1.53, respectively (p ​< ​0.02 for all). The presence of PR and/or LAP increased the frequency of abnormal FFRct with mild stenosis (p ​< ​0.05) with a similar trend with 70–89% stenosis. The combination of 2 HRP (LAP and PR) identified more lesions with FFR < 0.80 than only 1 HRP. ConclusionsSemiquantitative visual assessment of high-risk plaque features may improve the selection of studies for FFRct.

Refined staging classification of cardiac damage associated with aortic stenosis and outcomes after transcatheter aortic valve implantation.

A new staging classification of aortic stenosis (AS) characterizing the extent of cardiac damage was established and validated in patients undergoing transcatheter aortic valve implantation (TAVI). We aimed to validate an updated classification system in patients undergoing TAVI. In a prospective TAVI registry, AS patients were categorized into the following stages: no cardiac damage (Stage 0), left ventricular damage (Stage 1), left atrial or mitral valve damage (Stage 2), pulmonary vasculature or tricuspid valve damage (Stage 3), or right ventricular (RV) damage or low-flow state (Stage 4). Stage 3 was sub-divided into Stage 3a (≤moderate pulmonary hypertension) and Stage 3b (severe pulmonary hypertension). Stage 4 was sub-divided into Stage 4a (low-flow without RV dysfunction), Stage 4b (RV dysfunction without low-flow), and Stage 4c (RV dysfunction with low-flow). The primary endpoint was all-cause death at 1 year. Among 1156 eligible patients, 14 were classified as Stage 0, 38 as Stage 1, 105 as Stage 2278 as Stage 3, and 721 as Stage 4. There was a stepwise increase in mortality according to advancing stages of cardiac damage: 3.9% (Stage 0-1), 9.6% (Stage 2), 14.1% (Stage 3), and 17.4% (Stage 4) (P = 0.002). After multivariable adjustment, only Stage 3b, Stage 4b, and Stage 4c conferred a significantly increased risk of mortality compared to Stage 0-1. More than one-third of patients had advanced cardiac damage (severe pulmonary hypertension or RV dysfunction) before TAVI, associating with a five- to seven-fold increased risk of mortality at 1 year. https://www.clinicaltrials.gov. NCT01368250.

Deep Learning Model for Automated Detection and Classification of Central Canal, Lateral Recess, and Neural Foraminal Stenosis at Lumbar Spine MRI.

Background Assessment of lumbar spinal stenosis at MRI is repetitive and time consuming. Deep learning (DL) could improve -productivity and the consistency of reporting. Purpose To develop a DL model for automated detection and classification of lumbar central canal, lateral recess, and neural -foraminal stenosis. Materials and Methods In this retrospective study, lumbar spine MRI scans obtained from September 2015 to September 2018 were included. Studies of patients with spinal instrumentation or studies with suboptimal image quality, as well as postgadolinium studies and studies of patients with scoliosis, were excluded. Axial T2-weighted and sagittal T1-weighted images were used. Studies were split into an internal training set (80%), validation set (9%), and test set (11%). Training data were labeled by four radiologists using predefined gradings (normal, mild, moderate, and severe). A two-component DL model was developed. First, a convolutional neural network (CNN) was trained to detect the region of interest (ROI), with a second CNN for classification. An internal test set was labeled by a musculoskeletal radiologist with 31 years of experience (reference standard) and two subspecialist radiologists (radiologist 1: A.M., 5 years of experience; radiologist 2: J.T.P.D.H., 9 years of experience). DL model performance on an external test set was evaluated. Detection recall (in percentage), interrater agreement (Gwet κ), sensitivity, and specificity were calculated. Results Overall, 446 MRI lumbar spine studies were analyzed (446 patients; mean age ± standard deviation, 52 years ± 19; 240 women), with 396 patients in the training (80%) and validation (9%) sets and 50 (11%) in the internal test set. For internal testing, DL model and radiologist central canal recall were greater than 99%, with reduced neural foramina recall for the DL model (84.5%) and radiologist 1 (83.9%) compared with radiologist 2 (97.1%) (P < .001). For internal testing, dichotomous classification (normal or mild vs moderate or severe) showed almost-perfect agreement for both radiologists and the DL model, with respective κ values of 0.98, 0.98, and 0.96 for the central canal; 0.92, 0.95, and 0.92 for lateral recesses; and 0.94, 0.95, and 0.89 for neural foramina (P < .001). External testing with 100 MRI scans of lumbar spines showed almost perfect agreement for the DL model for dichotomous classification of all ROIs (κ, 0.95-0.96; P < .001). Conclusion A deep learning model showed comparable agreement with subspecialist radiologists for detection and classification of central canal and lateral recess stenosis, with slightly lower agreement for neural foraminal stenosis at lumbar spine MRI. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Hayashi in this issue.

Staging cardiac damage associated with aortic stenosis in patients undergoing transcatheter aortic valve implantation.

A new staging classification of aortic stenosis (AS) characterizing the extent of cardiac damage was established and validated in patients undergoing transcatheter aortic valve implantation (TAVI). The present study was aimed to refine the staging system by integrating a quantitative evaluation of right ventricular (RV) dysfunction defined by current echocardiographic guideline recommendations. In a prospective TAVI registry, patients were categorized into the stages: no cardiac damage (Stage 0), left ventricular damage (Stage 1), left atrial or mitral valve damage (Stage 2), pulmonary vasculature or tricuspid valve damage (Stage 3), or RV damage (Stage 4) based on baseline echocardiography. Among 1133 eligible patients undergoing TAVI, 8 (3.4%) patients were categorized as Stage 0, 113 (10.0%) as Stage 1, 397 (35.0%) as Stage 2, 239 (21.1%) as Stage 3, and 346 (30.5%) as Stage 4. There was a stepwise increase in all-cause and cardiovascular mortality rates at 1year according to increasing stages of secondary cardiac damage: 5.4% and 0% in Stage 0, 5.3% and 1.8% in Stage 1, 8.9% and 5.9% in Stage 2, 17.7% and 12.9% in Stage 3, and 25.8% and 19.9% in Stage 4, respectively. After multivariable adjustment, increasing stages of cardiac damage gradually correlated with all-cause and cardiovascular mortality. A significant number of patients with AS underwent TAVI only once cardiac damage has already occurred. Integrating a guideline-based definition of RV dysfunction increased the sensitivity of the staging system to identify patients at increased risk of death after TAVI.

Age and functional relevance of coronary stenosis: a post hoc analysis of the ADVISE II trial.

The influence of age-dependent changes on fractional flow reserve (FFR) or instantaneous wave-free ratio (iFR) and the response to pharmacological hyperaemia has not been investigated. We investigated the impact of age on these indices. This is a post hoc analysis of the ADVISE II trial, including a total of 690 pressure recordings (in 591 patients). Age-dependent correlations with FFR and iFR were calculated and adjusted for stenosis severity. Patients were stratified into three age terciles. The hyperaemic response to adenosine, calculated as the difference between resting and hyperaemic pressure ratios, and the prevalence of FFR-iFR discordance were assessed. Age correlated positively with FFR (r=0.08, 95% CI: 0.01 to 0.15, p=0.015), but not with iFR (r=-0.03, 95% CI: -0.11 to 0.04, p=0.411). The hyperaemic response to adenosine decreased with patient age (0.12±0.07, 0.11±0.06, 0.09±0.05, for the 1st [33-58 years], 2nd [59-69 years] and 3rd [70-94 years] age tertiles, respectively, p<0.001) and showed significant correlation with age (r=-0.14, 95% CI: -0.21 to -0.06, p<0.001). The proportion of patients with FFR ≤0.80+iFR >0.89 discordance doubled in the first age tercile (14.1% vs 7.1% vs 7.0%, p=0.005). The hyperaemic response of the microcirculation to adenosine administration is age dependent. FFR values increase with patient age, while iFR values remain constant across the age spectrum. These findings contribute to explaining differences observed in functional stenosis classification with hyperaemic and non-hyperaemic coronary indices.

Myocardial extracellular volume quantification by computed tomography predicts outcomes in patients with severe aortic stenosis.

The extent of myocardial fibrosis in patients with severe aortic stenosis might have an important prognostic value. Non-invasive imaging to quantify myocardial fibrosis by measuring extracellular volume fraction might have an important clinical utility prior to aortic valve intervention. Seventy-five consecutive patients with severe aortic stenosis, and 19 normal subjects were prospectively recruited and underwent pre- and post-contrast computed tomography for estimating myocardial extracellular volume fraction. Serum level of galectin-3 was measured and 2-dimensional echocardiography was performed to characterize the extent of cardiac damage using a recently published aortic stenosis staging classification. Extracellular volume fraction was higher in patients with aortic stenosis compared to normal subjects (40.0±11% vs. 21.6±5.6%; respectively, p<0.001). In patients with aortic stenosis, extracellular volume fraction correlated with markers of left ventricular decompensation including New York Heart Association functional class, left atrial volume, staging classification of aortic stenosis and lower left ventricular ejection fraction. Out of 75 patients in the AS group, 49 underwent TAVI, 6 surgical AVR, 2 balloon valvuloplasty, and 18 did not undergo any type of intervention. At 12-months after aortic valve intervention, extracellular volume fraction predicted the combined outcomes of stroke and hospitalization for heart failure with an area under the curve of 0.77 (95% confidence interval: 0.65-0.88). A trend for correlation between serum galectin-3 and extracellular volume was noted. In patients with severe aortic stenosis undergoing computed tomography before aortic valve intervention, quantification of extracellular volume fraction correlated with functional status and markers of left ventricular decompensation, and predicted the 12-months composite adverse clinical outcomes. Implementation of this novel technique might aid in the risk stratification process before aortic valve interventions.

Optimal Carotid Plaque Features on Computed Tomography Angiography Associated With Ischemic Stroke.

BackgroundStenosis has historically been the major factor used to determine carotid stroke sources. Recent evidence suggests that specific plaque features detected on imaging may be more highly associated with ischemic stroke than stenosis. We sought to determine computed tomography angiography (CTA) imaging features of carotid plaque that optimally discriminate ipsilateral stroke sources.Methods and ResultsIn this institutional review board–approved retrospective cross‐sectional study, 494 ipsilateral carotid CTA‐brain magnetic resonance imaging pairs were available for analysis after excluding patients with alternative stroke sources. Carotid CTA and clinical markers were recorded, a multivariable Poisson regression model was fitted, and backward elimination was performed with a 2‐sided threshold of P<0.10. Discriminatory value was determined using receiver operating characteristic analysis, area under the curve, and bootstrap validation. The final CTA carotid‐source stroke prediction model included intraluminal thrombus (prevalence ratio, 2.8 [P<0.001]; 95% CI, 1.6–4.9), maximum soft plaque thickness (prevalence ratio, 1.2 [P<0.001]; 95% CI, 1.1–1.4), and the rim sign (prevalence ratio, 2.0 [P=0.007]; 95% CI, 1.2–3.3). The final discriminatory value (area under the curve=78.3%) was higher than intraluminal thrombus (56.4%, P<0.001), maximum soft plaque thickness (76.4%, P=0.007), or rim sign alone (69.9%, P=0.001). Furthermore, NASCET (North American Symptomatic Carotid Endarterectomy Trial) stenosis categories (cutoffs of 50% and 70%) had lower stroke discrimination (area under the curve=67.4%, P<0.001).ConclusionsOptimal discrimination of ipsilateral carotid sources of stroke requires information on intraluminal thrombus, maximum soft plaque thickness, and the rim sign. These results argue against the sole use of carotid stenosis to determine stroke sources on CTA, and instead suggest these alternative markers may better diagnose vulnerable carotid plaque and guide treatment decisions.

Deep learning-based classification of lower extremity arterial stenosis in computed tomography angiography

PurposeThe purpose of this study is to develop and evaluate a deep learning model to assist radiologists in classifying lower extremity arteries based on the degree of arterial stenosis caused by plaque in lower extremity computed tomography angiography (CTA) of patients with peripheral artery disease. MethodsIn this retrospective study, 265 patients who underwent lower-extremity CTA between January 1, 2016 and October 31, 2019 were selected. A total of 17050 axial images of iliac, femoropopliteal and infrapopliteal artery from these patients were used for the training and validation of the parallel efficient network (p-EffNet), a kind of supervised convolutional neural network, to classify the lower-extremity artery segments according to the degree of stenosis with digital subtraction angiography as reference standard. The classification results of the p-EffNet were then compared with those obtained from radiologists. Receiver operating characteristic curve (ROC) was used to evaluate the performance of the p-EffNet and accuracy, specificity, sensitivity and area under the curve (AUC) were used as measure metrics to compare the performance of the p-EffNet and that of radiologists. ResultsThe p-EffNet exhibited a good performance of 91.5 % accuracy, 0.987 AUC and 90.2 % sensitivity and 97.7 % specificity in classifying above-knee artery and 90.9 % accuracy, 0.981 AUC, 91.3 % sensitivity and 95.2 % specificity in classifying below-knee artery. When compared with human readers, for both above-knee and below-knee artery, the p-EffNet had comparable accuracy (p = 0.266 and p = 0.808, respectively) and specificity (p = 0.118 and p = 0.971, respectively) but lower sensitivity (p < 0.001 and p = 0.022, respectively). ConclusionsThe p-EffNet demonstrates promising diagnostic performance and has the potential to reduce the workload of radiologists and help to find the plaques that might otherwise have been missed or misjudged.

Deep learning pipeline for automated detection and classification of central canal, lateral recess and neural foraminal stenosis on lumbar spine MRI

Deep learning pipeline for automated detection and classification of central canal, lateral recess and neural foraminal stenosis on lumbar spine MRI

Follow-up after carotid endarterectomy and stenting: What to look for and why

Duplex ultrasound testing after open or endovascular extracranial carotid artery interventions is a clinical practice guideline with a strong recommendation from the Society for Vascular Surgery. Neurologic outcomes are improved by the recognition of repair site stenosis or atherosclerotic disease progression in the unoperated carotid artery. The benefit of surveillance outweighs its risk because duplex testing is free of complications and accurate in the detection of internal carotid artery (ICA) stenosis or occlusion. Surveillance for >70% ICA stenosis is recommended within 30 days of the procedure, then every 6 months for 2 years, and annually thereafter. Repair site and contralateral ICA stenosis classification should be based on angle-corrected pulsed Doppler measurements of peak systolic velocity (PSV), end-diastolic velocity (EDV), and the ratio of PSV at the stenosis to a proximal, nondiseased common carotid artery (CCA) segment (ICA/CCA ratio). Interpretation criteria of PSV >300 cm/s, EDV >125 cm/s, and ICA/CCA ratio >4 predicts >70% repair site stenosis. Endovascular intervention is recommended for a carotid repair site stenosis based on the occurrence of an ipsilateral neurologic event and appropriate anatomy for angioplasty. For asymptomatic restenosis, intervention is based on stenosis progression to elevated PSV and EDV >70% stenosis threshold values and the patient is deemed high risk for stroke due to contralateral ICA occlusion or incomplete functional patency of the circle of Willis.

Automatic stenosis recognition from coronary angiography using convolutional neural networks

Background and objectiveCoronary artery disease, which is mostly caused by atherosclerotic narrowing of the coronary artery lumen, is a leading cause of death. Coronary angiography is the standard method to estimate the severity of coronary artery stenosis, but is frequently limited by intra- and inter-observer variations. We propose a deep-learning algorithm that automatically recognizes stenosis in coronary angiographic images. MethodsThe proposed method consists of key frame detection, deep learning model training for classification of stenosis on each key frame, and visualization of the possible location of the stenosis. Firstly, we propose an algorithm that automatically extracts key frames essential for diagnosis from 452 right coronary artery angiography movie clips. Our deep learning model is then trained with image-level annotations to classify the areas narrowed by over 50 %. To make the model focus on the salient features, we apply a self-attention mechanism. The stenotic locations are visualized using the activated area of feature maps with gradient-weighted class activation mapping. ResultsThe automatically detected key frame was very close to the manually selected key frame (average distance (1.70 ± 0.12) frame per clip). The model was trained with key frames on internal datasets, and validated with internal and external datasets. Our training method achieved high frame-wise area-under-the-curve of 0.971, frame-wise accuracy of 0.934, and clip-wise accuracy of 0.965 in the average values of cross-validation evaluations. The external validation results showed high performances with the mean frame-wise area-under-the-curve of (0.925 and 0.956) in the single and ensemble model, respectively. Heat map visualization shows the location for different types of stenosis in both internal and external data sets. With the self-attention mechanism, the stenosis could be precisely localized, which helps to accurately classify the stenosis by type. ConclusionsOur automated classification algorithm could recognize and localize coronary artery stenosis highly accurately. Our approach might provide the basis for a screening and assistant tool for the interpretation of coronary angiography.