Lesion-Specific Coronary Artery Calcium Quantification for Predicting Cardiac Event with Multiple Instance Support Vector Machines

Liu, Qingshan; Qian, Zhen; Marvasty, Idean; Rinehart, Sarah; Vörös, Szilárd; Metaxas, Dimitris N.

doi:10.1007/978-3-642-15705-9_59

Cited by 10 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following calcium scoring, all patients received a single puff of Nitroglycerin 0.4 mg (GLYTRIN 0.4mg/dose, sanofi, MoretonWirral, UK). After 2 minutes, a bolus of 80 mL of Iohexol (Omniopaque, 350 mg I/mL, GE Healthcare, Ireland cork, Ireland) followed by 30 mL saline chaser, was injected into an antecubital vein via an 19-gauge catheter (injection rate 6 mL/s) ( 18 ). Contrast-agent application was controlled by bolus-tracking in the aortic root.…”

Section: Methodsmentioning

confidence: 99%

Diagnostic Efficacy of Vessel Specific Coronary Calcium Score in Detection of Coronary Artery Stenosis

Motevalli

Firouznia

Kargar

et al. 2014

Iran Red Crescent Med J

View full text Add to dashboard Cite

Background:Coronary artery calcification which is determined quantitatively by coronary calcium scoring has been known as a sign of coronary stenosis and thus future cardiac events; hence it has been noticed on spotlight of researchers in recent years. Developing different method for early and optimal detection of coronary artery disease (CAD) is really essential as CAD are the first cause of death in population.Objectives:To evaluate predictive value of vessel specific coronary artery calcium (CAC) score in predicting obstructive coronary artery disease.Patients and Methods:In this diagnostic test study we evaluated patients with coronary computed tomography angiography (CCTA) and CAC score which had been referred to two referral radiology center in Tehran, Iran and finally we selected 2525 patients in a single and sequential pattern to create a diagnostic study. The whole-heart CAC scores and vessel specific CAC scores were calculated individually for the 4 major epicardial coronary arteries in 2 distinct group; group A ( patients with previous history of CABG) and group B (patients without history of CABG). For evaluation of obstruction tree cut off points were described: 0 > ; at least 1 segment with any kind of stenosis, ≥ 50; at least 1 segment with stenosis ≥ 50, ≥ 70; at least 1 segment with stenosis ≥ 70.Results:Mean of coronary calcium scores in terms of each coronary artery vessel increase by increasing coronary stenosis grade in group B; LAD, RCA, LCX respectively have mean CAC score 6.06, 6.21 and 5.04 in normal patients and 221.6, 226.7 and 106.6 in patients with complete stenosis. As expected these findings don't work for group A. Also By increasing calcium score cutoff in all four vessels sensitivity decreased and specificity increased but steal LAD had higher sensitivity than other vessels and LM had higher specificity. Thus using calcium score method is useful for ruling out stenosis in LAD while calcium score of LM can predict existence of stenosis in LM. However none of the vessel specific CAC can reach to 100% sensitivity and specificity of CCTA method.Conclusions:CCTA is highly superior than vessel specific CAC score thus to minimize patients radiation does maybe we can eliminate CAC scan as a routinely perform procedure at the beginning of the CCTA.

show abstract

Section: Methodsmentioning

confidence: 99%

Diagnostic Efficacy of Vessel Specific Coronary Calcium Score in Detection of Coronary Artery Stenosis

Motevalli

Firouznia

Kargar

et al. 2014

Iran Red Crescent Med J

View full text Add to dashboard Cite

show abstract

“…Recently, MIL has been adopted in the medical imaging analysis domain (Bi and Liang, 2007; Liu et al, 2010; Lu et al, 2011; Xu et al, 2012; Tong et al, 2014; Xu et al, 2014; Yan et al, 2016). In Lu et al (2011) and Xu et al (2012), MIL-like methods were developed to perform medical image segmentation.…”

Section: Related Workmentioning

confidence: 99%

“…In Bi and Liang (2007), a MIL-based method was proposed to screen pulmonary embolisms among candidates. Liu et al (2010) developed a MIL-SVM method to predict cardiac events. Tong et al (2014) proposed a MIL-like model for dementia classification with brain MRI data, by first extracting multiple image patches and then constructing graph kernels for SVM-based classification.…”

Section: Related Workmentioning

confidence: 99%

Landmark-based deep multi-instance learning for brain disease diagnosis

Liu

Zhang

Adeli

et al. 2018

Medical Image Analysis

374

204

View full text Add to dashboard Cite

In conventional Magnetic Resonance (MR) image based methods, two stages are often involved to capture brain structural information for disease diagnosis, i.e., 1) manually partitioning each MR image into a number of regions-of-interest (ROIs), and 2) extracting pre-defined features from each ROI for diagnosis with a certain classifier. However, these pre-defined features often limit the performance of the diagnosis, due to challenges in 1) defining the ROIs and 2) extracting effective disease-related features. In this paper, we propose a landmark-based deep multi-instance learning (LDMIL) framework for brain disease diagnosis. Specifically, we first adopt a data-driven learning approach to discover disease-related anatomical landmarks in the brain MR images, along with their nearby image patches. Then, our LDMIL framework learns an end-to-end MR image classifier for capturing both the local structural information conveyed by image patches located by landmarks and the global structural information derived from all detected landmarks. We have evaluated our proposed framework on 1526 subjects from three public datasets (i.e., ADNI-1, ADNI-2, and MIRIAD), and the experimental results show that our framework can achieve superior performance over state-of-the-art approaches.

show abstract

“…Recently, coronary computed tomographic angiography (CCTA) which is a noninvasive imaging method with high diagnostic accuracy ( 2 ) is used for detecting coronary artery stenosis. Coronary artery calcium (CAC) which is a good marker of atherosclerosis is evaluated at the beginning of CCTA with less radiation and it is an established predictor of cardiovascular events ( 3 - 6 ). Most frequently, the Agatston score has been used to measure the total coronary calcified plaque ( 7 ).…”

Section: Introductionmentioning

confidence: 99%

Diagnostic Accuracy of Coronary Calcium Score Less than 100 in Excluding Coronary Artery Disease

et al. 2016

View full text Add to dashboard Cite

BackgroundCoronary artery calcium score (CACS) is a quantitative assessment of calcifications and an established predictor of cardiovascular events.ObjectivesIn this study, we evaluated the diagnostic accuracy, negative predictive value (NPV), positive predictive value (PPV), specificity and sensitivity of CACS less than 100 in predicting significant coronary artery stenosis in patients with risk of coronary artery disease (CAD) in a vessel-based analysis.Patients and MethodsA cross sectional study was carried out on a study population of 2527 consecutive stable patients with symptoms suggestive of CAD who were referred for coronary computed tomographic angiography (CCTA). We performed 1343 studies with 256 slice machine in Shahid Rajaee hospital and the other studies were carried out with 64 slice machine in Imam Khomeini hospital and the calcium score was quantified according to the Agatston method.ResultsAt the cutoff point of 100 for coronary calcium scoring, there was high specificity (87%), high sensitivity (79%), high efficiency (84%), high PPV (79%), and high NPV (87%) in the diagnosis of significant stenosis in the whole heart. The frequency of zero calcium scoring was 59% in normal or nonsignificant stenosis and 7.6% in significant stenosis in the whole heart. Calcium scoring increased with greater severity of the arterial stenosis (P values < 0.001).ConclusionWe conclude that coronary calcium scoring provided useful information in the management of patients. In CACS less than 100, it has a NPV of 87% in excluding significant stenosis in patients with the risk of CAD but it does not have enough diagnostic accuracy for surely excluding coronary stenosis, so we should perform a combination of CACS and coronary CT angiography for patients.

show abstract

Lesion-Specific Coronary Artery Calcium Quantification for Predicting Cardiac Event with Multiple Instance Support Vector Machines

Cited by 10 publications

References 12 publications

Diagnostic Efficacy of Vessel Specific Coronary Calcium Score in Detection of Coronary Artery Stenosis

Diagnostic Efficacy of Vessel Specific Coronary Calcium Score in Detection of Coronary Artery Stenosis

Landmark-based deep multi-instance learning for brain disease diagnosis

Diagnostic Accuracy of Coronary Calcium Score Less than 100 in Excluding Coronary Artery Disease

Contact Info

Product

Resources

About