Background: We determined the age-related changes in atrioventricular junction (AVJ) velocities a... more Background: We determined the age-related changes in atrioventricular junction (AVJ) velocities and displacements by feature tracking cardiovascular magnetic resonance (FT-CMR) in a healthy community-based population. We also investigated the importance of age-matching for the identification of altered AVJ dynamics. Methods: FT-CMR was performed in 230 controls (18-78 years) and in two patient groups each consisting of 40 subjects (group 1: 23-55 years, group 2: 56-80 years). AVJ dynamic parameters, including systolic velocity Sm, early diastolic velocity Em, late diastolic velocity Am, maximal systolic excursion MAPSE and the new parameter sweep surface area velocity SSAV were measured. Results: Increasing age in the control group was significantly associated with reductions in Sm, Em, MAPSE (r = −0.40, −0.76, −0.34, all P b 0.001) and an increase in Am (r = 0.45, P b 0.001). For patient group 1, the selection of an age-unmatched control group (56-76 years) underestimated the number of patients with abnormal AVJ dynamics during systole and early diastole (38% vs. 70% for Sm; 20% vs. 60% for Em; 35% vs. 50% for MAPSE). In contrast, for patient group 2, the number of patients with systolic and early diastolic AVJ dynamic abnormalities was overestimated (88% vs. 63% for Sm; 90% vs. 68% for Em; 73% vs. 58% for MAPSE) when compared with age-unmatched controls (24-55 years). Fifty-percent (20/40) of the subgroup of patients with normal left ventricular ejection fraction exhibited abnormal systolic Sm or MAPSE measurements. Conclusions: Significant correlations exist between age and AVJ dynamics. Age matching is important for evaluating AVJ long-axis function.
Journal of the Royal Society, Interface / the Royal Society, Jan 6, 2015
In this work, we present a method to assess left ventricle (LV) regional function from cardiac ma... more In this work, we present a method to assess left ventricle (LV) regional function from cardiac magnetic resonance (CMR) imaging based on the regional ejection fraction (REF) and regional area strain (RAS). CMR scans were performed for 30 patients after first-time myocardial infarction (MI) and nine age- and sex-matched healthy volunteers. The CMR images were processed to reconstruct three-dimensional LV geometry, and the REF and RAS in a 16-segment model were computed using our proposed methodology. The method of computing the REF was tested and shown to be robust against variation in user input. Furthermore, analysis of data was feasible in all patients and healthy volunteers without any exclusions. The REF correlated well with the RAS in a nonlinear manner (quadratic fit-R(2) = 0.88). In patients after first-time MI, the REF and RAS were significantly reduced across all 16 segments (REF: p < 0.05; RAS: p < 0.01). Moreover, the REF and RAS significantly decreased with the ext...
Introduction: The maximal rate of change of pressure-normalised wall stress dσ*/dtmax has been pr... more Introduction: The maximal rate of change of pressure-normalised wall stress dσ*/dtmax has been proposed as cardiac index of left ventricular (LV) contractility. In this study, we assessed the capacity of dσ*/dtmax to diagnose heart failure with normal ejection fraction (HFNEF). Materials and Methods: One hundred healthy normal controls and 140 patients admitted with heart failure (100, HFREF and 40, HFNEF) underwent echocardiography for stress-based contractility dσ*/dtmax. Patients with significant valvular heart disease were excluded. Tissue Doppler indices were also measured. Results: dσ*/dtmaxwas 4.43 ± 1.27 s-1 in control subjects; reduced in HFNEF, 3.02 ± 0.98 s-1; and HFREF, 2.00 ± 0.67 s-1 (P <0.001). In comparison with age- and sex-matched groups (n = 26 each), we found similar trend on reduction of dσ*/dtmax (normal control; 3.91 ± 0.87 s-1; HFNEF, 2.90 ± 0.84 s-1; HFREF, 1.84 ± 0.59 s-1, P <0.001). On multivariate analysis, dσ*/dtmax was found to be the independent ...
In the presence of gestational diabetes mellitus (GDM), the fetus is exposed to a hyperinsulinemi... more In the presence of gestational diabetes mellitus (GDM), the fetus is exposed to a hyperinsulinemia environment. This environment can cause a wide range of metabolic and fetal cardiac structural alterations. Fetal myocardial hypertrophy predominantly affecting the interventricular septum possesses a morphology of disarray similar to hypertrophic cardiomyopathy, and may be present in some GDM neonates after birth. Myocardial thickness may increase in GDM fetuses independent of glycemic control status and fetal weight. Fetal echocardiography performed on fetuses with GDM helps in assessing cardiac structure and function, and to diagnose myocardial hypertrophy. There are few studies in the literature which have established evidence for morphologic variation associated with cardiac hypertrophy among fetuses of GDM mothers. In this study, fetal ultrasound images of normal, pregestational diabetes mellitus (preGDM) and GDM mothers were used to develop a computer aided diagnostic (CAD) tool. We proposed a new method called local preserving class separation (LPCS) framework to preserve the geometrical configuration of normal and preGDM/GDM subjects. The generated shearlet based texture features under LPCS framework showed promising results compared with deep learning algorithms. The proposed method achieved a maximum accuracy of 98.15% using a support vector machine (SVM) classifier. Hence, this paradigm can be helpful to physicians in detecting fetal myocardial hypertrophy in preGDM/GDM mothers. INDEX TERMS Fetal myocardial hypertrophy, gestational diabetes mellitus, local preserving class separation, computer-aided diagnosis, ultrasound images.
International Journal of Environmental Research and Public Health, 2021
Atrial fibrillation (AF) is a common arrhythmia that can lead to stroke, heart failure, and prema... more Atrial fibrillation (AF) is a common arrhythmia that can lead to stroke, heart failure, and premature death. Manual screening of AF on electrocardiography (ECG) is time-consuming and prone to errors. To overcome these limitations, computer-aided diagnosis systems are developed using artificial intelligence techniques for automated detection of AF. Various machine learning and deep learning (DL) techniques have been developed for the automated detection of AF. In this review, we focused on the automated AF detection models developed using DL techniques. Twenty-four relevant articles published in international journals were reviewed. DL models based on deep neural network, convolutional neural network (CNN), recurrent neural network, long short-term memory, and hybrid structures were discussed. Our analysis showed that the majority of the studies used CNN models, which yielded the highest detection performance using ECG and heart rate variability signals. Details of the ECG databases ...
International Journal of Environmental Research and Public Health, 2021
Arterial hypertension (HT) is a chronic condition of elevated blood pressure (BP), which may caus... more Arterial hypertension (HT) is a chronic condition of elevated blood pressure (BP), which may cause increased incidence of cardiovascular disease, stroke, kidney failure and mortality. If the HT is diagnosed early, effective treatment can control the BP and avert adverse outcomes. Physiological signals like electrocardiography (ECG), photoplethysmography (PPG), heart rate variability (HRV), and ballistocardiography (BCG) can be used to monitor health status but are not directly correlated with BP measurements. The manual detection of HT using these physiological signals is time consuming and prone to human errors. Hence, many computer-aided diagnosis systems have been developed. This paper is a systematic review of studies conducted on the automated detection of HT using ECG, HRV, PPG and BCG signals. In this review, we have identified 23 studies out of 250 screened papers, which fulfilled our eligibility criteria. Details of the study methods, physiological signal studied, database ...
This article presents a new deep learning approach for cardiac arrhythmia (17 classes) detection ... more This article presents a new deep learning approach for cardiac arrhythmia (17 classes) detection based on long-duration electrocardiography (ECG) signal analysis. Cardiovascular disease prevention is one of the most important tasks of any health care system as about 50 million people are at risk of heart disease in the world. Although automatic analysis of ECG signal is very popular, current methods are not satisfactory. The goal of our research was to design a new method based on deep learning to efficiently and quickly classify cardiac arrhythmias. Described research are based on 1000 ECG signal fragments from the MIT - BIH Arrhythmia database for one lead (MLII) from 45 persons. Approach based on the analysis of 10-s ECG signal fragments (not a single QRS complex) is applied (on average, 13 times less classifications/analysis). A complete end-to-end structure was designed instead of the hand-crafted feature extraction and selection used in traditional methods. Our main contributi...
Curvature-based three-dimensional cardiovascular magnetic resonance (CMR) allows regional functio... more Curvature-based three-dimensional cardiovascular magnetic resonance (CMR) allows regional function characterization without an external spatial frame of reference. However, introduction of this modality into clinical practice is hampered by lack of reference values. We aim to establish normal ranges for 3D left ventricular (LV) regional parameters in relation to age and gender for 171 healthy subjects. LV geometrical reconstruction and automatic calculation of regional parameters were implemented by in-house software (CardioWerkz) using stacks of short-axis cine slices. Parameter normal ranges were stratified by gender and age categories (≤44, 45–64, 65–74 and 75–84 years). Our software had excellent intra- and inter-observer agreement. Ageing was significantly associated with increases in end-systolic (ES) curvedness (CES) and area strain (AS) with higher rates of increase in males, end-diastolic (ED) and ES wall thickness (WTED, WTES) with higher rates of increase in females, and ...
A new framework is proposed for the auto-segmentation of the left ventricle (LV) from cardiac mag... more A new framework is proposed for the auto-segmentation of the left ventricle (LV) from cardiac magnetic resonance (CMR) images. The segmentation method is based on the random walk (RW) algorithm, which requires user-selected background and foreground seeds. In this paper, the seeds are initialized automatically. The first image frame of a short-axis slice is first partitioned into different regions using the fuzzy c-means (FCM) algorithm, and the LV region is identified using a heuristic method. Two circular region of interests (ROIs) are then defined based on the estimated centre of the partitioned LV region, which are used as the RW seeds initialization to segment the LV of the first image frame. The centre pixel of the adjacent image frame is then computed using the segmented LV of the previous frame. The foreground and background circular ROIs can then be defined and used as initialization of the RW algorithm to segment the adjacent image. The effectiveness of the proposed framework is verified by the experimental results on real CMR images.
Background: Abnormal left atrial (LA) function is a marker of cardiac dysfunction and adverse car... more Background: Abnormal left atrial (LA) function is a marker of cardiac dysfunction and adverse cardiovascular outcome, but is difficult to assess, and hence not, routinely quantified. We aimed to determine the feasibility and effectiveness of a fast method to measure long-axis LA strain and strain rate (SR) with standard cardiovascular magnetic resonance (CMR) compared to conventional feature tracking (FT) derived longitudinal strain. Methods: We studied 50 normal controls, 30 patients with hypertrophic cardiomyopathy, and 100 heart failure (HF) patients, including 40 with reduced ejection fraction (HFrEF), 30 mid-range ejection fraction (HFmrEF) and 30 preserved ejection fraction (HFpEF). LA longitudinal strain and SR parameters were derived by tracking the distance between the left atrioventricular junction and a user-defined point at the mid posterior LA wall on standard cine CMR two-and four-chamber views. LA performance was analyzed at three distinct cardiac phases: reservoir function (reservoir strain ε s and strain rate SR s), conduit function (conduit strain ε e and strain rate SR e) and booster pump function (booster strain ε a and strain rate SR a). Results: There was good agreement between LA longitudinal strain and SR assessed using the fast and conventional FT-CMR approaches (r = 0.89 to 0.99, p < 0.001). The fast strain and SRs showed a better intra-and inter-observer reproducibility and a 55% reduction in evaluation time (85 ± 10 vs. 190 ± 12 s, p < 0.001) compared to FT-CMR. Fast LA measurements in normal controls were 35.3 ± 5.2% for ε s , 18.1 ± 4.3% for ε e , 17.2 ± 3.5% for ε a , and 1.8 ± 0.4, − 2.0 ± 0.5, − 2.3 ± 0.6 s − 1 for the respective phasic SRs. Significantly reduced LA strains and SRs were observed in all patient groups compared to normal controls. Patients with HFpEF and HFmrEF had significantly smaller ε s , SR s , ε e and SR e than hypertrophic cardiomyopathy, and HFmrEF had significantly impaired LA reservoir and booster function compared to HFpEF. The fast LA strains and SRs were similar to FT-CMR for discriminating patients from controls (area under the curve (AUC) = 0.79 to 0.96 vs. 0.76 to 0.93, p = NS). Conclusions: Novel quantitative LA strain and SR derived from conventional cine CMR images are fast assessable parameters for LA phasic function analysis.
Background: We determined the age-related changes in atrioventricular junction (AVJ) velocities a... more Background: We determined the age-related changes in atrioventricular junction (AVJ) velocities and displacements by feature tracking cardiovascular magnetic resonance (FT-CMR) in a healthy community-based population. We also investigated the importance of age-matching for the identification of altered AVJ dynamics. Methods: FT-CMR was performed in 230 controls (18-78 years) and in two patient groups each consisting of 40 subjects (group 1: 23-55 years, group 2: 56-80 years). AVJ dynamic parameters, including systolic velocity Sm, early diastolic velocity Em, late diastolic velocity Am, maximal systolic excursion MAPSE and the new parameter sweep surface area velocity SSAV were measured. Results: Increasing age in the control group was significantly associated with reductions in Sm, Em, MAPSE (r = −0.40, −0.76, −0.34, all P b 0.001) and an increase in Am (r = 0.45, P b 0.001). For patient group 1, the selection of an age-unmatched control group (56-76 years) underestimated the number of patients with abnormal AVJ dynamics during systole and early diastole (38% vs. 70% for Sm; 20% vs. 60% for Em; 35% vs. 50% for MAPSE). In contrast, for patient group 2, the number of patients with systolic and early diastolic AVJ dynamic abnormalities was overestimated (88% vs. 63% for Sm; 90% vs. 68% for Em; 73% vs. 58% for MAPSE) when compared with age-unmatched controls (24-55 years). Fifty-percent (20/40) of the subgroup of patients with normal left ventricular ejection fraction exhibited abnormal systolic Sm or MAPSE measurements. Conclusions: Significant correlations exist between age and AVJ dynamics. Age matching is important for evaluating AVJ long-axis function.
Journal of the Royal Society, Interface / the Royal Society, Jan 6, 2015
In this work, we present a method to assess left ventricle (LV) regional function from cardiac ma... more In this work, we present a method to assess left ventricle (LV) regional function from cardiac magnetic resonance (CMR) imaging based on the regional ejection fraction (REF) and regional area strain (RAS). CMR scans were performed for 30 patients after first-time myocardial infarction (MI) and nine age- and sex-matched healthy volunteers. The CMR images were processed to reconstruct three-dimensional LV geometry, and the REF and RAS in a 16-segment model were computed using our proposed methodology. The method of computing the REF was tested and shown to be robust against variation in user input. Furthermore, analysis of data was feasible in all patients and healthy volunteers without any exclusions. The REF correlated well with the RAS in a nonlinear manner (quadratic fit-R(2) = 0.88). In patients after first-time MI, the REF and RAS were significantly reduced across all 16 segments (REF: p < 0.05; RAS: p < 0.01). Moreover, the REF and RAS significantly decreased with the ext...
Introduction: The maximal rate of change of pressure-normalised wall stress dσ*/dtmax has been pr... more Introduction: The maximal rate of change of pressure-normalised wall stress dσ*/dtmax has been proposed as cardiac index of left ventricular (LV) contractility. In this study, we assessed the capacity of dσ*/dtmax to diagnose heart failure with normal ejection fraction (HFNEF). Materials and Methods: One hundred healthy normal controls and 140 patients admitted with heart failure (100, HFREF and 40, HFNEF) underwent echocardiography for stress-based contractility dσ*/dtmax. Patients with significant valvular heart disease were excluded. Tissue Doppler indices were also measured. Results: dσ*/dtmaxwas 4.43 ± 1.27 s-1 in control subjects; reduced in HFNEF, 3.02 ± 0.98 s-1; and HFREF, 2.00 ± 0.67 s-1 (P <0.001). In comparison with age- and sex-matched groups (n = 26 each), we found similar trend on reduction of dσ*/dtmax (normal control; 3.91 ± 0.87 s-1; HFNEF, 2.90 ± 0.84 s-1; HFREF, 1.84 ± 0.59 s-1, P <0.001). On multivariate analysis, dσ*/dtmax was found to be the independent ...
In the presence of gestational diabetes mellitus (GDM), the fetus is exposed to a hyperinsulinemi... more In the presence of gestational diabetes mellitus (GDM), the fetus is exposed to a hyperinsulinemia environment. This environment can cause a wide range of metabolic and fetal cardiac structural alterations. Fetal myocardial hypertrophy predominantly affecting the interventricular septum possesses a morphology of disarray similar to hypertrophic cardiomyopathy, and may be present in some GDM neonates after birth. Myocardial thickness may increase in GDM fetuses independent of glycemic control status and fetal weight. Fetal echocardiography performed on fetuses with GDM helps in assessing cardiac structure and function, and to diagnose myocardial hypertrophy. There are few studies in the literature which have established evidence for morphologic variation associated with cardiac hypertrophy among fetuses of GDM mothers. In this study, fetal ultrasound images of normal, pregestational diabetes mellitus (preGDM) and GDM mothers were used to develop a computer aided diagnostic (CAD) tool. We proposed a new method called local preserving class separation (LPCS) framework to preserve the geometrical configuration of normal and preGDM/GDM subjects. The generated shearlet based texture features under LPCS framework showed promising results compared with deep learning algorithms. The proposed method achieved a maximum accuracy of 98.15% using a support vector machine (SVM) classifier. Hence, this paradigm can be helpful to physicians in detecting fetal myocardial hypertrophy in preGDM/GDM mothers. INDEX TERMS Fetal myocardial hypertrophy, gestational diabetes mellitus, local preserving class separation, computer-aided diagnosis, ultrasound images.
International Journal of Environmental Research and Public Health, 2021
Atrial fibrillation (AF) is a common arrhythmia that can lead to stroke, heart failure, and prema... more Atrial fibrillation (AF) is a common arrhythmia that can lead to stroke, heart failure, and premature death. Manual screening of AF on electrocardiography (ECG) is time-consuming and prone to errors. To overcome these limitations, computer-aided diagnosis systems are developed using artificial intelligence techniques for automated detection of AF. Various machine learning and deep learning (DL) techniques have been developed for the automated detection of AF. In this review, we focused on the automated AF detection models developed using DL techniques. Twenty-four relevant articles published in international journals were reviewed. DL models based on deep neural network, convolutional neural network (CNN), recurrent neural network, long short-term memory, and hybrid structures were discussed. Our analysis showed that the majority of the studies used CNN models, which yielded the highest detection performance using ECG and heart rate variability signals. Details of the ECG databases ...
International Journal of Environmental Research and Public Health, 2021
Arterial hypertension (HT) is a chronic condition of elevated blood pressure (BP), which may caus... more Arterial hypertension (HT) is a chronic condition of elevated blood pressure (BP), which may cause increased incidence of cardiovascular disease, stroke, kidney failure and mortality. If the HT is diagnosed early, effective treatment can control the BP and avert adverse outcomes. Physiological signals like electrocardiography (ECG), photoplethysmography (PPG), heart rate variability (HRV), and ballistocardiography (BCG) can be used to monitor health status but are not directly correlated with BP measurements. The manual detection of HT using these physiological signals is time consuming and prone to human errors. Hence, many computer-aided diagnosis systems have been developed. This paper is a systematic review of studies conducted on the automated detection of HT using ECG, HRV, PPG and BCG signals. In this review, we have identified 23 studies out of 250 screened papers, which fulfilled our eligibility criteria. Details of the study methods, physiological signal studied, database ...
This article presents a new deep learning approach for cardiac arrhythmia (17 classes) detection ... more This article presents a new deep learning approach for cardiac arrhythmia (17 classes) detection based on long-duration electrocardiography (ECG) signal analysis. Cardiovascular disease prevention is one of the most important tasks of any health care system as about 50 million people are at risk of heart disease in the world. Although automatic analysis of ECG signal is very popular, current methods are not satisfactory. The goal of our research was to design a new method based on deep learning to efficiently and quickly classify cardiac arrhythmias. Described research are based on 1000 ECG signal fragments from the MIT - BIH Arrhythmia database for one lead (MLII) from 45 persons. Approach based on the analysis of 10-s ECG signal fragments (not a single QRS complex) is applied (on average, 13 times less classifications/analysis). A complete end-to-end structure was designed instead of the hand-crafted feature extraction and selection used in traditional methods. Our main contributi...
Curvature-based three-dimensional cardiovascular magnetic resonance (CMR) allows regional functio... more Curvature-based three-dimensional cardiovascular magnetic resonance (CMR) allows regional function characterization without an external spatial frame of reference. However, introduction of this modality into clinical practice is hampered by lack of reference values. We aim to establish normal ranges for 3D left ventricular (LV) regional parameters in relation to age and gender for 171 healthy subjects. LV geometrical reconstruction and automatic calculation of regional parameters were implemented by in-house software (CardioWerkz) using stacks of short-axis cine slices. Parameter normal ranges were stratified by gender and age categories (≤44, 45–64, 65–74 and 75–84 years). Our software had excellent intra- and inter-observer agreement. Ageing was significantly associated with increases in end-systolic (ES) curvedness (CES) and area strain (AS) with higher rates of increase in males, end-diastolic (ED) and ES wall thickness (WTED, WTES) with higher rates of increase in females, and ...
A new framework is proposed for the auto-segmentation of the left ventricle (LV) from cardiac mag... more A new framework is proposed for the auto-segmentation of the left ventricle (LV) from cardiac magnetic resonance (CMR) images. The segmentation method is based on the random walk (RW) algorithm, which requires user-selected background and foreground seeds. In this paper, the seeds are initialized automatically. The first image frame of a short-axis slice is first partitioned into different regions using the fuzzy c-means (FCM) algorithm, and the LV region is identified using a heuristic method. Two circular region of interests (ROIs) are then defined based on the estimated centre of the partitioned LV region, which are used as the RW seeds initialization to segment the LV of the first image frame. The centre pixel of the adjacent image frame is then computed using the segmented LV of the previous frame. The foreground and background circular ROIs can then be defined and used as initialization of the RW algorithm to segment the adjacent image. The effectiveness of the proposed framework is verified by the experimental results on real CMR images.
Background: Abnormal left atrial (LA) function is a marker of cardiac dysfunction and adverse car... more Background: Abnormal left atrial (LA) function is a marker of cardiac dysfunction and adverse cardiovascular outcome, but is difficult to assess, and hence not, routinely quantified. We aimed to determine the feasibility and effectiveness of a fast method to measure long-axis LA strain and strain rate (SR) with standard cardiovascular magnetic resonance (CMR) compared to conventional feature tracking (FT) derived longitudinal strain. Methods: We studied 50 normal controls, 30 patients with hypertrophic cardiomyopathy, and 100 heart failure (HF) patients, including 40 with reduced ejection fraction (HFrEF), 30 mid-range ejection fraction (HFmrEF) and 30 preserved ejection fraction (HFpEF). LA longitudinal strain and SR parameters were derived by tracking the distance between the left atrioventricular junction and a user-defined point at the mid posterior LA wall on standard cine CMR two-and four-chamber views. LA performance was analyzed at three distinct cardiac phases: reservoir function (reservoir strain ε s and strain rate SR s), conduit function (conduit strain ε e and strain rate SR e) and booster pump function (booster strain ε a and strain rate SR a). Results: There was good agreement between LA longitudinal strain and SR assessed using the fast and conventional FT-CMR approaches (r = 0.89 to 0.99, p < 0.001). The fast strain and SRs showed a better intra-and inter-observer reproducibility and a 55% reduction in evaluation time (85 ± 10 vs. 190 ± 12 s, p < 0.001) compared to FT-CMR. Fast LA measurements in normal controls were 35.3 ± 5.2% for ε s , 18.1 ± 4.3% for ε e , 17.2 ± 3.5% for ε a , and 1.8 ± 0.4, − 2.0 ± 0.5, − 2.3 ± 0.6 s − 1 for the respective phasic SRs. Significantly reduced LA strains and SRs were observed in all patient groups compared to normal controls. Patients with HFpEF and HFmrEF had significantly smaller ε s , SR s , ε e and SR e than hypertrophic cardiomyopathy, and HFmrEF had significantly impaired LA reservoir and booster function compared to HFpEF. The fast LA strains and SRs were similar to FT-CMR for discriminating patients from controls (area under the curve (AUC) = 0.79 to 0.96 vs. 0.76 to 0.93, p = NS). Conclusions: Novel quantitative LA strain and SR derived from conventional cine CMR images are fast assessable parameters for LA phasic function analysis.
Uploads
Papers by R Tan