Reproducibility and Repeatability of Coronary Computed Tomography Angiography (CCTA) Image Segmentation in Detecting Atherosclerosis: A Radiomics Study.

Atherosclerosis is known as the leading factor in heart disease with the highest mortality rate among the Malaysian population. Usually, the gold standard for diagnosing atherosclerosis is by using the coronary computed tomography angiography (CCTA) technique to look for plaque within the coronary artery. However, qualitative diagnosis for noncalcified atherosclerosis is vulnerable to false-positive diagnoses, as well as inconsistent reporting between observers. In this study, we assess the reproducibility and repeatability of segmenting atherosclerotic lesions manually and semiautomatically in CCTA images to identify the most appropriate CCTA image segmentation method for radiomics analysis to quantitatively extract the atherosclerotic lesion. Thirty (30) CCTA images were taken retrospectively from the radiology image database of Hospital Canselor Tuanku Muhriz (HCTM), Kuala Lumpur, Malaysia. We extract 11,700 radiomics features which include the first-order, second-order and shape features from 180 times of image segmentation. The interest vessels were segmentized manually and semiautomatically using LIFEx (Version 7.0.15, Institut Curie, Orsay, France) software by two independent radiology experts, focusing on three main coronary blood vessels. As a result, manual segmentation with a soft-tissuewindowing setting yielded higher repeatability as compared to semiautomatic segmentation with a significant intraclass correlation coefficient (intra-CC) 0.961 for thefirst-order and shape features; intra-CC of 0.924 for thesecond-order features with p < 0.001. Meanwhile, the semiautomatic segmentation has higher reproducibility as compared to manual segmentation with significant interclass correlation coefficient (inter-CC) of 0.920 (first-order features) and a good interclass correlation coefficient of 0.839 for the second-order features with p < 0.001. The first-order, shape order and second-order features for both manual and semiautomatic segmentation have an excellent percentage of reproducibility and repeatability (intra-CC > 0.9). In conclusion, semi-automated segmentation is recommended for inter-observer study while manual segmentation with soft tissue-windowing can be used for single observer study.

Impact of Coronary Stent Architecture on Clinical Outcomes: Do Minor Changes in Stent Architecture Really Matter?

INTRODUCTION: The objective of this study was to compare the accumulated clinical outcomes of two Malaysian all-comers populations, each treated with different polymer-free sirolimus-eluting stents (PF-SES) of similar stent design. METHODS: The Malaysian subpopulation of two all-comers observational studies based on the same protocol (ClinicalTrials.gov Identifiers: NCT02629575 and NCT02905214) were combined and compared to a Malaysian-only cohort which was treated with a later-generation PF-SES. The PF-SES's used differed only in their bare-metal backbone architecture, with otherwise identical sirolimus coating. The primary endpoint was the accumulated target lesion revascularization (TLR) rate at 12 months. The rates of major adverse cardiac events (MACE), stent thrombosis (ST) and myocardial infarction (MI) were part of the secondary endpoints. RESULTS: A total of 643 patients were treated with either the first-generation PF-SES (413 patients) or second-generation PF-SES (230 patients). Patient demographics were similar in terms of age (p = 0.744), male gender (0.987), diabetes mellitus (p = 0.293), hypertension (p = 0.905) and acute coronary syndrome (ACS, 44.8% vs. 46.1%, p = 0.752) between groups. There were no differences between treatment groups in terms of lesion length (20.8 ± 7.3 mm vs. 22.9 ± 7.9, p = 0.111) or vessel diameter (2.87 ± 0.39 vs. 2.93 ± 0.40, p = 0.052) despite numerically smaller diameters in the first-generation PF-SES group. The second-generation PF-SES tended to have more complex lesions as characterized by calcification (10.3% vs. 16.2%, p = 0.022), severe tortuosity (3.5% vs. 6.9%, p = 0.041) and B2/C lesions (49.2% vs. 62.8%, p < 0.001). The accumulated TLR rates did not differ significantly between the first- and second-generation PF-SES (0.8% vs. 0.9%, p = 0.891). The accumulated MACE rates were not significantly different (p = 0.561), at 1.5% (6/413) and 2.2% (5/230), respectively. CONCLUSIONS: Modifications in coronary stent architecture which enhance the radial strength and radiopacity without gross changes in strut thickness and design do not seem to impact clinical outcomes. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov Identifiers: NCT02629575 and NCT02905214.

Real-time Detection of Aortic Valve in Echocardiography using Convolutional Neural Networks.

BACKGROUND: Valvular heart disease is a serious disease leading to mortality and increasing medical care cost. The aortic valve is the most common valve affected by this disease. Doctors rely on echocardiogram for diagnosing and evaluating valvular heart disease. However, the images from echocardiogram are poor in comparison to Computerized Tomography and Magnetic Resonance Imaging scan. This study proposes the development of Convolutional Neural Networks (CNN) that can function optimally during a live echocardiographic examination for detection of the aortic valve. An automated detection system in an echocardiogram will improve the accuracy of medical diagnosis and can provide further medical analysis from the resulting detection. METHODS: Two detection architectures, Single Shot Multibox Detector (SSD) and Faster Regional based Convolutional Neural Network (R-CNN) with various feature extractors were trained on echocardiography images from 33 patients. Thereafter, the models were tested on 10 echocardiography videos. RESULTS: Faster R-CNN Inception v2 had shown the highest accuracy (98.6%) followed closely by SSD Mobilenet v2. In terms of speed, SSD Mobilenet v2 resulted in a loss of 46.81% in framesper- second (fps) during real-time detection but managed to perform better than the other neural network models. Additionally, SSD Mobilenet v2 used the least amount of Graphic Processing Unit (GPU) but the Central Processing Unit (CPU) usage was relatively similar throughout all models. CONCLUSION: Our findings provide a foundation for implementing a convolutional detection system to echocardiography for medical purposes.

Systematic Scoring Balloon Lesion Preparation for Drug-Coated Balloon Angioplasty in Clinical Routine: Results of the PASSWORD Observational Study.

INTRODUCTION: Scoring balloon angioplasty (SBA) for lumen gain prior to stent implantations or drug-coated balloon angioplasty (DCB) is considered an essential interventional tool for lesion preparation. Recent evidence indicates that SBA may play a pivotal role in enhancing the angiographic and clinical outcomes of DCB angioplasty. METHODS: We studied the systematic use of SBA with a low profile, non-slip element device prior to DCB angioplasty in an unselected, non-randomized patient population. This prospective, all-comers study enrolled patients with de novo lesions as well as in-stent restenotic lesions in bare metal stents (BMS-ISR) and drug-eluting stents (DES-ISR). The primary endpoint was the target lesion failure (TLF) rate at 9 months (ClinicalTrials.gov Identifier NCT02554292). RESULTS: A total of 481 patients (496 lesions) were recruited to treat de novo lesions (78.4%, 377), BMS-ISR (4.0%, 19), and DES-ISR (17.6%, 85). Overall risk factors were acute coronary syndrome (ACS, 20.6%, 99), diabetes mellitus (46.8%, 225), and atrial fibrillation (8.5%, 41). Average lesion lengths were 16.7 ± 10.4 mm in the de novo group, and 20.1 ± 8.9 mm (BMS-ISR) and 16.2 ± 9.8 mm (DES-ISR) in the ISR groups. Scoring balloon diameters were 2.43 ± 0.41 mm (de novo), 2.71 ± 0.31 mm (BMS-ISR), and 2.92 ± 0.42 mm (DES-ISR) whereas DCB diameters were 2.60 ± 0.39 mm (de novo), 3.00 ± 0.35 mm (BMS-ISR), and 3.10 ± 0.43 mm (DES-ISR), respectively. The overall accumulated TLF rate of 3.0% (14/463) was driven by significantly higher target lesion revascularization rates in the BMS-ISR (5.3%, 1/19) and the DES-ISR group (6.0%, 5/84). In de novo lesions, the TLF rate was 1.1% (4/360) without differences between calcified and non-calcified lesions (p = 0.158) and small vs. large reference vessel diameters with a cutoff value of 3.0 mm (p = 0.901). CONCLUSIONS: The routine use of a non-slip element scoring balloon catheter to prepare lesions suitable for drug-coated balloon angioplasty is associated with high procedural success rates and low TLF rates in de novo lesions.

Analysis of Gln223Agr polymorphism of Leptin Receptor Gene in type II diabetic mellitus subjects among Malaysians.

Leptin is known as the adipose peptide hormone. It plays an important role in the regulation of body fat and inhibits food intake by its action. Moreover, it is believed that leptin level deductions might be the cause of obesity and may play an important role in the development of Type 2 Diabetes Mellitus (T2DM), as well as in cardiovascular diseases (CVD). The Leptin Receptor (LEPR) gene and its polymorphisms have not been extensively studied in relation to the T2DM and its complications in various populations. In this study, we have determined the association of Gln223Agr loci of LEPR gene in three ethnic groups of Malaysia, namely: Malays, Chinese and Indians. A total of 284 T2DM subjects and 281 healthy individuals were recruited based on International Diabetes Federation (IDF) criteria. Genomic DNA was extracted from the buccal specimens of the subjects. The commercial polymerase chain reaction (PCR) method was carried out by proper restriction enzyme MSP I to both amplify and digest the Gln223Agr polymorphism. The p-value among the three studied races was 0.057, 0.011 and 0.095, respectively. The values such as age, WHR, FPG, HbA1C, LDL, HDL, Chol and Family History were significantly different among the subjects with Gln223Agr polymorphism of LEPR (p < 0.05).