The use of artificial intelligence for predicting postinfarction myocardial viability in echocardiographic images.

BACKGROUND: Evaluation of standard echocardiographic examination with artificial intelligence may help in the diagnosis of myocardial viability and function recovery after acute coronary syndrome. METHODS: Sixty-one consecutive patients with acute coronary syndrome were enrolled in the present study (43 men, mean age 61 ± 9 years). All patients underwent percutaneous coronary intervention (PCI). 533 segments of the heart echo images were used. After 12 ± 1 months of follow-up, patients had an echocardiographic evaluation. After PCI each patient underwent cardiac magnetic resonance (CMR) with late enhancement and low-dose dobutamine echocardiographic examination. For texture analysis, custom software was used (MaZda 5.20, Institute of Electronics).Linear and non-linear (neural network) discriminative analyses were performed to identify the optimal analytic method correlating with CMR regarding the necrosis extent and viability prediction after follow-up. Texture parameters were analyzed using machine learning techniques: Artificial Neural Networks, Namely Multilayer Perceptron, Nonlinear Discriminant Analysis, Support Vector Machine, and Adaboost algorithm. RESULTS: The mean concordance between the CMR definition of viability and three classification models in Artificial Neural Networks varied from 42% to 76%. Echo-based detection of non-viable tissue was more sensitive in the segments with the highest relative transmural scar thickness: 51-75% and 76-99%. The best results have been obtained for images with contrast for red and grey components (74% of proper classification). In dobutamine echocardiography, the results of appropriate prediction were 67% for monochromatic images. CONCLUSIONS: Detection and semi-quantification of scar transmurality are feasible in echocardiographic images analyzed with artificial intelligence. Selected analytic methods yielded similar accuracy, and contrast enhancement contributed to the prediction accuracy of myocardial viability after myocardial infarction in 12 months of follow-up.

Feasibility and clinical utility of real-time tele-echocardiography using personal mobile device-based pocket echocardiograph.

Introduction: Recently introduced microUSB ultrasound probe, which can be connected to personal mobile device constitutes a new class of diagnostic pocket size imaging devices (PSID).The aim of this study was to assess the feasibility and clinical utility of real-time tele-echocardiography with the use such device. Material and methods: The study group comprised 30 patients (18 men and 12 women; mean age: 54 ±14 years; mean body mass index: 27 ±6 kg/m2), admitted to various hospital departments (infectious diseases, internal medicine and cardiology). All patients underwent focused echocardiographic examination performed by an inexperienced operator using PSID and collaborating remotely in real-time with cardiologist by means of tele-consultation. Before commencing the examination, the operators underwent 1 h training in use of the PSID. Results: In most of patients PSID imaging provided sufficient diagnostic image quality. The dimensions of left ventricle, left atrium and the aorta obtained during the this examination showed good correlation with standard echocardiography (r = 0.89, r = 0.82, r = 0.92 respectively), but the correlation for TAPSE measurements was less pronounced (r = 0.52). The agreement in detection of pathologies (wall motion abnormalities, morphological and functional valvular abnormalities) between the tele-echocardiography and standard echocardiographic examination was good to very good (κ ranged from 0.648 to 0.823). Conclusions: Pocket size imaging devices allows an inexperienced operator to perform a focused echocardiographic examination under a remote supervision of an experienced cardiologist. The introduction of new technology in the form of PSID with tele-echocardiography feature can revolutionize access to this imaging technique.

Feasibility and value of two-dimensional volumetric stress echocardiography.

BACKGROUND: Stroke volume response during stress is a major determinant of functional status in heart failure and can be measured by two-dimensional (2-D) volumetric stress echocardiography (SE). The present study hypothesis is that SE may identify mechanisms underlying the change in stroke volume by measuring preload reserve through end-diastolic volume (EDV) and left ventricular contractile reserve (LVCR) with systolic blood pressure and end-systolic volume (ESV). METHODS: We enrolled 4735 patients (age 63.6±11.3 years, 2800 male) referred to SE for known or suspected coronary artery disease (CAD) and/or heart failure (HF) in 21 SE laboratories in 8 countries. In addition to regional wall motion abnormalities (RWMA), force was measured at rest and peak stress as the ratio of systolic blood pressure by cuff sphygmomanometer/ESV by 2D with Simpson's or linear method. Abnormal values of LVCR (peak/rest) based on force were ≤1.10 for dipyridamole (N.=1992 patients) and adenosine (N.=18); ≤2.0 for exercise (N.=2087) or dobutamine (N.=638). RESULTS: Force-based LVCR was obtained in all 4735 patients. Lack of stroke volume increase during stress was due to either abnormal LVCR and/or blunted preload reserve, and 57% of patients with abnormal LVCR nevertheless showed increase in stroke volume. CONCLUSIONS: Volumetric SE is highly feasible with all stresses, and more frequently impaired in presence of ischemic RWMA, absence of viability and reduced coronary flow velocity reserve. It identifies an altered stroke volume response due to reduced preload and/or contractile reserve.

Flow-mediated skin fluorescence: A novel method for the estimation of sleep apnea risk in healthy persons and cardiac patients.

BACKGROUND: A pilot study revealed a relationship between the results of flow mediated skin fluorescence (FMSF) and of ECG-Holter-based estimated apnea/hypopnea index (eAHI) in asymptomatic individuals. The aim of this study was to test whether the results of FMSF show a relationship with the eAHI in patients with coronary artery disease or aortic stenosis. METHODS: Twenty-one patients (12 coronary disease, 9 aortic stenosis) and 37 healthy volunteers were included. FMSF was assessed before, during and after the pressure occlusion of the brachial artery, using a prototype device allowing the quantification of skin fluorescence. The values of FMSF expressed as baseline (BASE), maximum (MAX), and minimum (MIN) were analyzed. The percentages of ischemic response (IR) and hyperemic response (HR) were calculated. The eAHI was assessed from night ECG-Holter recordings. Differences between the groups and the relationships between the parameters were analyzed statistically. RESULTS: Mean ± standard deviation of BASE, MAX, MIN and IR were not significantly different in both groups (p > 0.05). HR was significantly lower in cardiac patients (14.7 ± 7.5 vs. 11.8 ± 5.1; p = 0.048), whose eAHI was significantly higher (11.0 ± 7.4 vs. 36.3 ± 16.5; p < 0.01). Negative correlation for MAX and eAHI was found in volunteers and patients: r = -0.38, p = 0.02 and r = -0.47, p = 0.03, respectively. In volunteers, HR had a negative correlation with eAHI: r = -0.34, p = 0.04. CONCLUSIONS: This pioneer study confirms that FMSF can be used to detect the negative correlation between MAX fluorescence and eAHI not only among healthy volunteers, but also among cardiac patients with coronary artery disease or aortic stenosis.

Complexity of changes in right ventricular morphology and function in patients undergoing cardiac surgery - 3D echocardiographic study.

BACKGROUND: An impairment of standard echocardiographic parameters of right ventricular (RV) function is a known phenomenon in patients undergoing cardiac surgery, but its significance remains unclear. AIMS: This study aimed to assess changes in RV function in patients undergoing cardiac surgery using speckle tracking and 3D echocardiography. METHODS: The study population comprised 122 patients referred for cardiac surgery. Transthoracic echocardiographic (TTE) examinations were performed: before the surgery (TTE1), 1 week after surgery (TTE2), and 1 year after surgery (TTE 3). Parameters measured during these examinations included both standard and advanced indices of the RV size and function, as well as a new parameter introduced by our team - RV shortening fraction (RV SF). RESULTS: TTE1 was performed on average (standard deviation [SD]) 24 (15) hours before surgery, whereas TTE2 and TTE3 were performed on average 7.2 (3) days and 346 (75) days after the surgery, respectively. A postoperative impairment of parameters of RV longitudinal function was observed (P <0.001). However, neither the RV size assessed by both 2D and 3D techniques changed, nor the global RV function measured with the use of fractional area change and ejection fraction. Additionally, during the postoperative period, an increase in the value of an RV SF by 12.9% was observed. After 12 months we observed an improvement in the parameters of the longitudinal RV function. CONCLUSIONS: Uncomplicated cardiac surgery causes transient impairment of the longitudinal systolic RV function, with no influence on the global RV function. The preservation of global function results from increased RV SF. After 12 months, an improvement of the longitudinal function can be observed.

Artificial Intelligence-Powered Measurement of Left Ventricular Ejection Fraction Using a Handheld Ultrasound Device.

The aim of this study was to assess the accuracy of an algorithm for automated measurement of left ventricular ejection fraction (LVEF) available on handheld ultrasound devices (HUDs). One hundred twelve patients admitted to the cardiology department underwent assessment performed with an HUD. In each case, the four-chamber apical view was obtained, and LVEF was calculated with LVivo software. Subsequently, during the examination performed with the use of the stationary echocardiograph, the 3-D measurement of LVEF was recorded. The average LVEFs measured with LVivo and the 3-D reference method were 46 ± 14% and 48 ± 14%, respectively. The correlation between the measurements obtained with the HUD and 3-D evaluation was high (r = 0.92, 95% confidence interval: 0.87-0.95, p < 0.0001). The mean difference between the LVEF obtained with LVivo and the 3-D LVEF was not significant (mean difference: -0.61%, 95% confidence interval: -1.89 to 0.68, p = 0.31). The LVivo software despite its limitations is capable of the accurate LVEF measurement when the acquired views are of at least good imaging quality.

Quantitative assessment of the rotation and twist of the left ventricle during dobutamine stress echocardiography: a comparison of patients with and without significant coronary artery disease.

BACKGROUND: The rotation and twist of the left ventricle (LV) have been comprehensively evaluated at rest. However, little is known about rotational mechanics during dobutamine stress echocardiography (DSE). AIMS: We aimed to quantify and compare the basal and apical rotation and twist of the LV at rest as well as at the peak and recovery stages of DSE in patients with and without coronary artery disease (CAD). METHODS: We enrolled 91 patients, including 48 patients with CAD and 43 patients without CAD (mean [SD] age, 62 [9] years and 61 [10] years, respectively). Coronary artery disease was defined as the presence of stenoses of 50% or more in the left main coronary artery and/or stenoses of 70% or more in other epicardial arteries. Rotation was measured by 2­dimensional speckle­tracking echocardiography, and twist was calculated as the difference between the basal and apical rotation. RESULTS: Neither rotation nor twist differed between patients with and without CAD at rest, although apical rotation was significantly greater in the CAD group at peak DSE (mean [SD], 5.43° [3.45°] vs 3.71° [3.52°], P = 0.01) and at recovery (mean [SD], 5.05° [3.65°] vs 2.87° [2.73°], P <0.01). On the contrary, the absolute value for basal rotation at recovery was higher in patients without CAD (mean [SD], 3.87° [3.37°] vs 2.63° [2.43°], P = 0.03). In both groups, the rotation and twist did not change significantly during the dobutamine challenge. CONCLUSIONS: During DSE, we observed differences in LV rotation between patients with and without CAD, revealing the effect of ischemia on deformation parameters.

Incremental value of left atrial mechanical dispersion over CHA2 DS2 -VASc score in predicting risk of thrombus formation.

PURPOSE: To assess the potential ability of two-dimensional speckle tracking analysis (STE) during atrial fibrillation (AF) to identify patients with LA appendage thrombi (LAAT). METHODS: This study involved 93 patients with AF (39% female, 67.1 ± 9.5 years) who were referred for a clinical indication for transesophageal echocardiography (TEE). TEE revealed LAAT in 39 (42%) patients. We analyzed standard parameters of the left ventricle (LV) and LA in transthoracic echocardiography. Using STE, we assessed LV global longitudinal strain (LVGLS), peak atrial longitudinal strain (PALS), and intra-atrial asynchrony. The PALS was calculated using the global strain curve (GPALS) and as the mean of peaks derived from segmental strain curves (MPALS). RESULTS: Patients were comparable with regard to the clinical data. A subgroup with LAAT had lower LV ejection fraction (LVEF) and a lower absolute value of the LVGLS, as well as greater impairment in the LA standard parameters, PALS, and asynchrony. Receiver operating characteristic curve analysis revealed that the LVEF of 30% (P < .001), the LVGLS of -7% (P < .0001), the GPALS of 11% (P < .005), and the LA asynchrony of 22% (P < .01) were the optimal cutoff values for distinguishing both subgroups. LA asynchrony, LVEF, and LVGLS were independently associated with the presence of LAAT in multivariate analyses, and PALS had additional significance over the CHA2 DS2 -VASc score. CONCLUSIONS: Left ventricular systolic dysfunction characterized by both LVEF and LVGLS is an independent factor for LAAT. LA asynchrony provides additional diagnostic value for discriminating between patients with and without LAAT.

Bicuspid aortic valve morphology and its impact on aortic diameters-A systematic review with meta-analysis and meta-regression.

AIM: To evaluate the impact of the 2 most common bicuspid aortic valve (BAV) morphology patterns [right-left (RL) vs right-noncoronary (RN) cusp fusion] on the aortic diameters and the impact of gender, aortic stenosis (AS), aortic regurgitation (AR), and age on the observed effects. METHODS: The PubMed databases was searched up to December 31, 2016 to identify studies investigating the morphology of BAV and aortic diameters. Inclusion criteria were as follows: the data on diameter of sinuses of Valsalva (SVD) and/or ascending aorta (AAD) and BAV morphology. The additional characteristics [gender, AS and AR (% of patients with moderate or severe AS/AR) and mean age] were collected to perform a meta-regression analysis. RESULTS: A total of 12 studies with 2192 patients with indexed AAD, 15 studies with 3104 patients with nonindexed AAD and 8 studies with 1271 patients with indexed SVD, and 16 studies with 3454 patients with nonindexed SVD were included. There was no difference between RL and RN group in indexed/nonindexed AAD-mean difference (MD): 0.06 mm/m2 (95% CI: -0.65 to 0.77 mm/m2 , P = .87) and -0.06 mm (95% CI: 1.10-0.97 mm, P = .91). Differently, the RL BAV was associated with larger indexed/nonindexed SVD than RN phenotype-MD: 1.66 mm/m2 (95% CI: 0.83-2.49 mm/m2 , P < .001) and 2.03 mm (95% CI: 0.97-3.09 mm, P < .001). Age, gender, AS, and AR had no influence on observed differences. CONCLUSIONS: RL BAV phenotype is associated with larger SVD than RN BAV, and the observed differences are independent from aortic valve dysfunction degree, age, and gender.

Conscious sedation for transcatheter implantation of atrial septal occluders with two- and three-dimensional transoesophageal echocardiography guidance - a feasibility and safety study.

BACKGROUND: General anaesthesia may have negative impact on patient mortality and morbidity, as well as overall procedure costs, in atrial septal occluder (ASO) implantation. AIM: We sought to evaluate the safety, efficacy, and feasibility of conscious sedation for transcatheter implantation of ASOs. METHODS: A total of 122 patients referred for transcatheter implantation of ASO were included. Mean patient age was 51 ± 15 years, and 43 (35%) patients were male. The initial dose of midazolam was 2 mg and fentanyl dose was 25 µg. Additional doses of midazolam and fentanyl were administered, if necessary. Patient responsiveness was assessed every 10 min, and the sedatives doses were titrated in order not to exceed grade 3 sedation in the Ramsey scale. RESULTS: Atrial septal occluders were successfully implanted in the majority of patients (98.4%). In two (1.6%) cases the proce-dure failed because of too small patent foramen ovale (PFO) diameter (n = 1, 0.8%) or device instability (n = 1, 0.8%). The mean duration of procedure was 47.6 ± 28.4 min and was similar for ASD and PFO closure (p = 0.522). The overall mean dose of midazolam was 4.7 ± 2.2 mg (63.9 ± 32.5 µg/kg) and fentanyl was 30.0 ± 11.9 µg (0.43 ± 0.17 µg/kg). Median entrance dose of radiation at the patient plane was 25 (interquartile range: 16-57) mGy, and did not differ between ASD and PFO procedures (p = 0.614). The majority of patients were free of complications (91.0%). The following early complications were observed: transient ischaemic attack (n = 2, 1.6%), supraventricular arrhythmias (n = 4, 3.3%), left atrial thrombus formation (n = 1, 0.8%), symptomatic bradycardia (n = 1, 0.8%), and femoral venous bleeding (n = 5, 4.1%). After mean follow-up of 386 days residual shunt was observed in eight (6.6%) patients. CONCLUSIONS: Conscious sedation for transcatheter implantation of ASO is a feasible, safe, and efficient technique, allowing successful PFO and ASD closure in the majority of patients.

Conscious sedation during cryoballoon ablation of atrial fibrillation: a feasibility and safety study.

BACKGROUND: Cryoballoon ablation for atrial fibrillation (cryoAF) is relatively simple, cost-effective and easy procedure. However, general anesthesia during this procedure may have negative impact on patients' mortality and morbidity, as well as procedure costs. We sought to assess the feasibility and safety of conscious sedation during cryoAF. METHODS: Patients with paroxysmal or persistent, drug-refractory AF who underwent a first procedure of cryoAF were included in our single-center, retrospective study. The loading dose of midazolam was 2 mg and loading fentanyl dose was 25 µg intravenously. Additional doses of midazolam and fentanyl were administered, if necessary. Midazolam and fentanyl were administered to maintain amnesia and analgesia, and patients' responsiveness in Ramsay Sedation Score was assessed every 10 minutes with sedation kept at the Ramsey Sedation Scale not exceeding Ramsey 3 Grade. RESULTS: A total of 71 patients with mean age 59.5±11.1 years were enrolled (48 males, 68%). The TEE and the TEE-guided transseptal puncture was feasible in all patients. The mean duration of cryoAF procedure was 136.3±36.0 minutes. The overall mean midazolam and fentanyl used doses were: 5.1±3.1 mg (55.7±35.1 µg/kg) and 98.9±51.1 µg (1.13±0.60 µg/kg). In 3 patients (4.2%), vascular complications occurred (femoral vein bleeding with hematoma formation without communication with femoral artery). In 1 case (1.4%) transient right phrenic palsy was observed; symptoms disappeared completely within 12 hours after procedure. The overall acute procedural success rate of cryoAF (defined as electrical isolation of all pulmonary veins) was 68/71 (95.8%). CONCLUSIONS: Conscious sedation is a safe, efficacious and feasible during cryoablation of pulmonary veins for AF.