Expertise In Echocardiography Research Articles

Improving pericardial effusion detection: evaluating sensitivity and specificity of traditional and novel ECG criteria

Abstract Funding Acknowledgements None. Introduction Historical studies have demonstrated that conventional low voltage criteria on ECG have high specificity but limited sensitivity in detecting pericardial effusion (PE). Indeed, an ECG showing low voltage can be indicative of several conditions, including PE. Purpose This study aimed to introduce new ECG criteria organized into a score called ARENA, to improve ECG diagnostic performance in detecting PE. Furthermore, we compared the sensitivity and specificity of traditional low voltage ECG criteria with the ARENA score using a test cohort. Methods This retrospective case-control study was conducted at our public University hospital from January 2022 to May 2023. Inclusion criteria were: age ≥18 years old, an available 12 lead ECG and an echocardiogram performed within 5 days from the recorded ECG. Exclusion criteria were: ventricular paced rhythm, heart transplantation, poor subcostal echocardiographic window. In the cohort of patients were selected: patients with ≥1.0 cm of PE on echocardiography and patients without effusion on echocardiography. ECG parameters were evaluated by two cardiologists blinded for the presence or absence of PE. Echocardiographic images were reviewed by an expert echocardiographer. Optimal ECG criteria with the highest ROC curve area were selected from the test cohort to develop the ARENA score. Patients were assessed using both traditional low voltage ECG criteria and the novel ARENA score. Statistical analysis showed that with 96 patients in the test cohort, there would be >90% power to detect a significant area under the curve (AUC) of 0.7 (vs. the null hypothesis of AUC of 0.5) to detect a moderate effect size. We used the AUC analysis to estimate the predicted performance of the proposed criteria. The McNemar test was used to assess the lack of agreement between the ECG criteria and the gold standard (echocardiography), and the results were reported as percentages with their respective 95% confidence intervals (CI). A p-value < 0.05 was considered statistically significant. Results The patient population consisted of 97 patients: 50 patients with PE, and 47 patients without it. The ARENA score showed a sensitivity of 52.0% (95% CI: 39.2% to 64.8%), compared to 14.0% (95% CI: 6.2% to 21.8%) of the traditional criteria. Specificity was 94% (95% CI: 84.1% to 100%) for ARENA Score and 100% (95% CI: 96.3% to 100%) for traditional criteria. Accuracy of the ARENA score was 70.0%, compared to 57% of the traditional criteria. The McNemar test indicated a p-value of <0.001 in the performance comparison. The AUC for the ARENA score was 0.781. Conclusion The novel ARENA score demonstrated an improved sensitivity while maintaining high specificity in detecting PE. This new score could become a valuable diagnostic tool for detecting PE by ECG in daily practice.ROC Curve ARENA ScoreECG criteria for pericardial effusion

RL-TEE: Autonomous Probe Guidance for Transesophageal Echocardiography Based on Attention-Augmented Deep Reinforcement Learning

Ultrasound image acquisition in conventional transesophageal echocardiography (TEE) requires complex manual operation of the probe in the esophagus based on the interpretation of ultrasound images and in-depth knowledge of the cardiac anatomy. In this work, we formulate the TEE probe guidance task as a reinforcement learning (RL) problem, and present the first learning-based solution to 3-DOF control of a TEE probe based on the ultrasound image feedback, named RL-TEE, in order to mimic the visual search and navigation strategies of expert echocardiographers. The probe-tissue interaction in TEE is carefully modeled in our framework by considering both the requirements for navigation towards the standard views and compliance in the esophageal environment. Furthermore, we propose a hybrid deep Q-network model that augments a convolutional neural network backbone with self-attention mechanisms to better capture spatial information in ultrasound images to guide navigation decisions. The presented methods are preliminarily validated in a TEE simulation environment built with data from 25 subjects to acquire four standard views of the heart. Our results show that the proposed method can effectively learn to accurately and compliantly guide the probe movement for TEE standard view acquisition tasks and has a good generalization ability to unseen patient data. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The motivation of this paper is to realize 3-DOF movement guidance of a TEE probe to acquire the standard views of the heart based on the real-time images, which can be applied to existing robotic control systems or used to assist novice echocardiographers in TEE examination, thereby relieving operator workload and improving ease of use. This paper suggests a novel approach that uses the deep RL technique to achieve automatic interpretation of TEE images and intelligent guidance of the probe movement. The RL framework is designed to take into account both the navigation efficiency and compliance with the esophageal environment for the targeted intracorporeal application. A hybrid deep Q-network model that augments a convolutional neural network with attention mechanisms is designed to better capture spatial information from ultrasound images to predict the probe movement. The effectiveness of the framework is preliminarily validated in extensive experiments in a simulation environment built with real patient data. The proposed method can be applied in clinical use to provide real-time TEE probe guidance for novice echocardiographers, and can be integrated with a robotic system to fully automate the TEE acquisition, thereby relieving the doctors from tedious manual operation to focus on the diagnosis and treatment.

Assessment of Left Ventricular Hypertrophy Incidence and Antihypertensive Therapy Efficacy in Hypertensive Patients

Objectives: This study aimed to assess the incidence of left ventricular hypertrophy (LVH) in hypertensive individuals in Karachi, Pakistan, and to evaluate the effectiveness of antihypertensive therapy in preventing LVH. Methodology: A total of 200 hypertensive patients were referred from the Cardiac OPD to the Radiology department of Medicare Cardiac and General Hospital, Karachi, for echocardiography. Echocardiographic measurements of cardiac chambers were performed by expert echocardiographers, and left ventricular mass (LVM) was calculated using the formula recommended by the American Society of Echocardiography. Results: Among the hypertensive patients, 100 had uncontrolled blood pressure (≥140/90 mmHg) while 100 had controlled blood pressure (≤130/80 mmHg). The female-to-male ratio was 3:1, with ages ranging from 21 to 80 years. Patients with uncontrolled hypertension exhibited a higher mean body mass index (29 ± 5 kg/m2) and significantly elevated LVM (mean 154 ± 4 g/m2). Angiotensin receptor inhibitors were the most commonly prescribed and effective antihypertensive agents, either as monotherapy or in combination therapy (60%), demonstrating utility in preventing LVH. Conclusion: This study highlights a high incidence of LVH in uncontrolled hypertensive individuals compared to those with controlled hypertension. Transthoracic echocardiography emerged as a highly valuable, non-invasive tool for measuring LVM and assessing the effectiveness of antihypertensive therapy.

Hand-Held Echocardiography by Advanced Practice Providers in Patients with Congestive Heart Failure.

The performed hand-held echocardiography (HHE) was evaluated and interpreted by trained advanced practice providers (APPs) on hospitalized CHF patients for image quality and interpretation by comparing with expert echocardiographer and SE findings. Congestive heart failure (CHF) is associated with increased hospital admissions and mortality. While a standard echocardiogram (SE) is the gold standard for cardiac assessment, it is not readily available. Hospitalized CHF patients require rapid assessment for expedited treatment. Over 6 months, five trained APPs performed HHE on hospitalized CHF patients and interpreted: (a) left ventricular (LV) size, (b) LV ejection fraction (LVEF), and (c) right atrial pressure (RAP). The study echocardiographer reviewed and blindly interpreted the HHE images and compared them with APPs and SE findings. Kappa statistics determined the degree of agreement between APPs and the study echocardiographer's interpretation of the HHE images and SE. A total of 80 CHF patients (age 73 ± 14 years, 58% males; LVEF (by SE) 45 ± 19%; 36.3% body mass indexes ≥ 30 kg/m2) were enrolled. HHE interpretation by APPs had a good agreement for LVEF (kappa 0.79) with the study echocardiographer and SE (kappa 0.74) and a good agreement for RAP (kappa 0.67) with the study echocardiographer. The correlation between the absolute LVEF interpretation by the study echocardiographer on HHE and SE was r = 0.88 (p < 0.0001). Trained APPs obtained diagnostic-quality HHE images and interpreted the LV function and RAP in CHF patients in good agreement with the study echocardiographer. LVEF by HHE correlated with LVEF by SE. Our study suggests trained APPs can use HHE to evaluate LVEF and RAP in CHF patients, leading to expedited and optimized treatment.

Empowering Medical Students: Harnessing Artificial Intelligence for Precision Point-of-Care Echocardiography Assessment of Left Ventricular Ejection Fraction

Introduction. Point-of-care ultrasound (POCUS) use is now universal among nonexperts. Artificial intelligence (AI) is currently employed by nonexperts in various imaging modalities to assist in diagnosis and decision making. Aim. To evaluate the diagnostic accuracy of POCUS, operated by medical students with the assistance of an AI-based tool for assessing the left ventricular ejection fraction (LVEF) of patients admitted to a cardiology department. Methods. Eight students underwent a 6-hour didactic and hands-on training session. Participants used a hand-held ultrasound device (HUD) equipped with an AI-based tool for the automatic evaluation of LVEF. The clips were assessed for LVEF by three methods: visually by the students, by students + the AI-based tool, and by the cardiologists. All LVEF measurements were compared to formal echocardiography completed within 24 hours and were evaluated for LVEF using the Simpson method and eyeballing assessment by expert echocardiographers. Results. The study included 88 patients (aged 58.3 ± 16.3 years). The AI-based tool measurement was unsuccessful in 6 cases. Comparing LVEF reported by students’ visual evaluation and students + AI vs. cardiologists revealed a correlation of 0.51 and 0.83, respectively. Comparing these three evaluation methods with the echocardiographers revealed a moderate/substantial agreement for the students + AI and cardiologists but only a fair agreement for the students’ visual evaluation. Conclusion. Medical students’ utilization of an AI-based tool with a HUD for LVEF assessment achieved a level of accuracy similar to that of cardiologists. Furthermore, the use of AI by the students achieved moderate to substantial inter-rater reliability with expert echocardiographers’ evaluation.

The clinical use of stress echocardiography in chronic coronary syndromes and beyond coronary artery disease: a clinical consensus statement from the European Association of Cardiovascular Imaging of the ESC.

Since the 2009 publication of the stress echocardiography expert consensus of the European Association of Echocardiography, and after the 2016 advice of the American Society of Echocardiography-European Association of Cardiovascular Imaging for applications beyond coronary artery disease, new information has become available regarding stress echo. Until recently, the assessment of regional wall motion abnormality was the only universally practiced step of stress echo. In the state-of-the-art ABCDE protocol, regional wall motion abnormality remains the main step A, but at the same time, regional perfusion using ultrasound-contrast agents may be assessed. Diastolic function and pulmonary B-lines are assessed in step B; left ventricular contractile and preload reserve with volumetric echocardiography in step C; Doppler-based coronary flow velocity reserve in the left anterior descending coronary artery in step D; and ECG-based heart rate reserve in non-imaging step E. These five biomarkers converge, conceptually and methodologically, in the ABCDE protocol allowing comprehensive risk stratification of the vulnerable patient with chronic coronary syndromes. The present document summarizes current practice guidelines recommendations and training requirements and harmonizes the clinical guidelines of the European Society of Cardiology in many diverse cardiac conditions, from chronic coronary syndromes to valvular heart disease. The continuous refinement of imaging technology and the diffusion of ultrasound-contrast agents improve image quality, feasibility, and reader accuracy in assessing wall motion and perfusion, left ventricular volumes, and coronary flow velocity. Carotid imaging detects pre-obstructive atherosclerosis and improves risk prediction similarly to coronary atherosclerosis. The revolutionary impact of artificial intelligence on echocardiographic image acquisition and analysis makes stress echo more operator-independent and objective. Stress echo has unique features of low cost, versatility, and universal availability. It does not need ionizing radiation exposure and has near-zero carbon dioxide emissions. Stress echo is a convenient and sustainable choice for functional testing within and beyond coronary artery disease.

Real-time guidance by deep learning of experienced operators to improve the standardization of echocardiographic acquisitions

Abstract Aims Impaired standardization of echocardiograms may increase inter-operator variability. This study aimed to determine whether the real-time guidance of experienced sonographers by deep learning (DL) could improve the standardization of apical recordings. Methods and results Patients (n = 88) in sinus rhythm referred for echocardiography were included. All participants underwent three examinations, whereof two were performed by sonographers and the third by cardiologists. In the first study period (Period 1), the sonographers were instructed to provide echocardiograms for the analyses of the left ventricular function. Subsequently, after brief training, the DL guidance was used in Period 2 by the sonographer performing the second examination. View standardization was quantified retrospectively by a human expert as the primary endpoint and the DL algorithm as the secondary endpoint. All recordings were scored in rotation and tilt both separately and combined and were categorized as standardized or non-standardized. Sonographers using DL guidance had more standardized acquisitions for the combination of rotation and tilt than sonographers without guidance in both periods (all P ≤ 0.05) when evaluated by the human expert and DL [except for the apical two-chamber (A2C) view by DL evaluation]. When rotation and tilt were analysed individually, A2C and apical long-axis rotation and A2C tilt were significantly improved, and the others were numerically improved when evaluated by the echocardiography expert. Furthermore, all, except for A2C rotation, were significantly improved when evaluated by DL (P &amp;lt; 0.01). Conclusion Real-time guidance by DL improved the standardization of echocardiographic acquisitions by experienced sonographers. Future studies should evaluate the impact with respect to variability of measurements and when used by less-experienced operators. ClinicalTrials.gov Identifier NCT04580095

Automated and semi-automated 3D echocardiographic software for aortic annulus sizing in transcatheter aortic valve implantation helps bridge the gap between expert and novice operators.

3D-transesophageal echocardiography (3D-TEE) is an alternative to multidetector row computed tomography (MDCT) for aortic annulus (AoA) sizing in preparation for Transcatheter aortic valve implantation (TAVI). We aim to evaluate how the fully automated (auto) and semi-automated (SA) TEE methods perform compared to conventional manual TEE method and the gold standard MDCT for annulus sizing both in expert and novice operators. In this prospective cohort study, eighty-nine patients with severe aortic stenosis underwent multimodality imaging with 3D-TEE and MDCT. Annular measurements were collected by expert echocardiographers using 3D auto, SA and manual methods and compared to MDCT. A novice in the field of echocardiography retrospectively measured the AoA for all patients using the same methods. TEE measurements, independently of the method used, had good to very good agreement to MDCT. They significantly underestimated aortic annular area and circumference vs. MDCT with the auto method underestimating it the most and the manual method the least (6.5% and 1.3% respectively for area and circumference). For experts, the manual TEE method offered the least systematic bias while the SA method had narrower limits of agreement (LOA). For the novice operator, SA method provided the least bias and narrower LOA vs. MDCT. There is good agreement between novice and experts for all 3 TEE methods but better agreement with auto and SA methods as opposed to manual one. Our study supports the use of 3D-TEE as a complementary method to MDCT for aortic annular sizing. The newer auto and SA software, that requires minimal operator intervention, is an easy to use, reliable and reproducible tool for aortic annulus sizing for experienced operators, and especially less experienced ones.

Controlled prospective study on ultrasound simulation training in fetal echocardiography: FESIM II.

To analyze the learning curves of ultrasound novices in fetal echocardiography during structured simulation-based ultrasound training (SIM-UT) including a virtual, randomly moving fetus. 11 medical students with minimal (< 10h) prior obstetric ultrasound experience underwent 12h of structured fetal echocardiography SIM-UT in individual hands-on sessions during a 6-week training program. Their learning progress was assessed with standardized tests after 2, 4, and 6weeks of SIM-UT. Participants were asked to obtain 11 fetal echocardiography standard planes (in accordance with ISUOG and AHA guidelines) as quickly as possible. All tests were carried out under real life, examination-like conditions on a healthy, randomly moving fetus. Subsequently, we analyzed the rate of correctly obtained images and the total time to completion (TTC). As reference groups, 10 Ob/Gyn physicians (median of 750 previously performed Ob/Gyn scans) and 10 fetal echocardiography experts (median of 15,000 previously performed Ob/Gyn scans) were examined with the same standardized tests. The students showed a consistent and steady improvement of their ultrasound performance during the training program. After 2weeks, they were able to obtain > 95% of the standard planes correctly. After 6weeks, they were significantly faster than the physician group (p < 0.001) and no longer significantly slower than the expert group (p = 0.944). SIM-UT is highly effective to learn fetal echocardiography. Regarding the acquisition of the AHA/ISUOG fetal echocardiography standard planes, the students were able to reach the same skill level as the expert group within 6 weeks.

Eye-tracking during simulation-based echocardiography: a feasibility study

IntroductionDue to the technical progress point-of-care ultrasound (POCUS) is increasingly used in critical care medicine. However, optimal training strategies and support for novices have not been thoroughly researched so far. Eye-tracking, which offers insights into the gaze behavior of experts may be a useful tool for better understanding. The aim of this study was to investigate the technical feasibility and usability of eye-tracking during echocardiography as well as to analyze differences of gaze patterns between experts and non-experts.MethodsNine experts in echocardiography and six non-experts were equipped with eye-tracking glasses (Tobii, Stockholm, Sweden), while performing six medical cases on a simulator. For each view case specific areas of interests (AOI) were defined by the first three experts depending on the underlying pathology. Technical feasibility, participants’ subjective experience on the usability of the eye-tracking glasses as well as the differences of relative dwell time (focus) inside the areas of interest (AOI) between six experts and six non-experts were evaluated.ResultsTechnical feasibility of eye-tracking during echocardiography was achieved with an accordance of 96% between the visual area orally described by participants and the area marked by the glasses. Experts had longer relative dwell time in the case specific AOI (50.6% versus 38.4%, p = 0.072) and performed ultrasound examinations faster (138 s versus 227 s, p = 0.068). Furthermore, experts fixated earlier in the AOI (5 s versus 10 s, p = 0.033).ConclusionThis feasibility study demonstrates that eye-tracking can be used to analyze experts and non-experts gaze patterns during POCUS. Although, in this study the experts had a longer fixation time in the defined AOIs compared to non-experts, further studies are needed to investigate if eye-tracking could improve teaching of POCUS.

Short period of training in 3D echocardiography provides good feasibility and reproducibility of right ventricular assessment in heart failure

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): PROHEART 3D. Background Right ventricular (RV) dysfunction is an independent risk factor for increased mortality and morbidity. Standard approach for assessing is 2-dimensional echocardiography (2DE). However, 2DE is prone to errors due to increased variability of the anatomical cut-planes and geometrical assumptions due to the complex triangular and "wrapped-around" RV shape. Meanwhile, 2DE reflects mainly the RV longitudinal function, at the expense of the radial and global RV functions. 3D echocardiography (3DE) enabled itself as a more accurate method for assessment of cardiac volumes and calculation of ejection fraction (EF), but with a cumbersome learning time-curve. Objectives To establish if trainees in 3DE provide feasible and reproducible 3D measurements of RV size and function, by comparison to 2DE assessment, after a short period of training by an expert in echocardiography. Methods 161 consecutive patients (58±17 years, 71% males, 3D left ventricular EF 35±10%), hospitalized for heart failure (HF), underwent standard precordial 2DE acquisitions and 3DE multi-beat full-volumes of the RV. One expert in both 2DE and 3DE, and one fellow trained for 2 years in 2DE and for 3 months in 3DE, performed measurements of both 2DE and 3DE parameters for RV size and function, in a blinded fashion. 2DE parameters were measured first, in the following order: RV diameter, RV end-diastolic (RVED) and end-systolic (RVES) areas, RV fractional area change (RV FAC), tricuspid annular plane systolic excursion (TAPSE), tricuspid annular systolic velocity (RV S’), and RV free wall longitudinal strain (RV LS). Then, 3D RVED and RVES volumes were measured, and RV ejection fraction (RVEF) was calculated, using a semi-automated software package. Endocardial borders were corrected, as needed. Results Feasibility of 2DE and 3DE was 97% and 92%, respectively, patients being excluded due to poor quality images, similarly for expert vs. trainee. Mean RVED volume was 94.0±30.6 ml/m2, and mean RVEF was 37.3±9.6%. Reproducibility of 2DE parameters obtained by the expert vs. trainee were good for the assessment of RV longitudinal function (TAPSE, RV S', and RV LS), but suboptimal for the assessment of RV size and global function (RV diameter, RV areas, and RV FAC) (Table). Reproducibility of 3DE parameters obtained by the expert vs. trainee were also good, even after a short period of training (3 months) (Table). Conclusions A short period of training in 3DE provides a good feasibility and reproducibility of the assessment of RV size and function, in a large cohort of hospitalized patients with HF. 3DE might be a more reliable method than 2D for the initial assessment of patients with HF.

Fetal Echocardiographic Z Score Pilot Project: Study Design and Impact of Gestational Age and Variable Type on Reproducibility of Measurements Within and Across Investigators

Fetal echocardiography is widely available, but normative data are not robust. In this pilot study, the authors evaluated (1) the feasibility of prespecified measurements in a normal fetal echocardiogram to inform study design and (2) measurement variability to assign thresholds of clinical significance and guide analyses in larger fetal echocardiographic Z score initiatives. Images from predefined gestational age groups (16-20, >20-24, >24-28, and >28-32weeks) were retrospectively analyzed. Fetal echocardiography expert raters attended online group training and then independently analyzed 73 fetal studies (18 per age group) in a fully crossed design of 53 variables; each observer repeated measures for 12 fetuses. Kruskal-Wallis tests were used to compare measurements across centers and age groups. Coefficients of variation (CoVs) were calculated at the subject level for each measurement as the ratio of SD to mean. Intraclass correlation coefficients were used to show inter- and intrarater reliabilities. Cohen's d>0.8 was used to define clinically important differences. Measurements were plotted against gestational age, biparietal diameter, and femur length. Expert raters completed each set of measurements in a mean of 23±9 min/fetus. Missingness ranged from 0% to 29%. CoVs were similar across age groups for all variables (P<.05) except ductus arteriosus mean velocity and left ventricular ejection time, which were both higher at older gestational age. CoVs were >15% for right ventricular systolic and diastolic widths despite fair to good repeatability (intraclass correlation coefficient>0.5); ductal velocities and two-dimensional measures, left ventricular short-axis dimensions, and isovolumic times all had high CoVs and high interobserver variability despite good to excellent intraobserver agreement (intraclass correlation coefficient>0.6). CoVs did not improve when ratios (e.g., tricuspid/mitral annulus) were used instead of linear measurements. Overall, 27 variables had acceptable inter- and intraobserver repeatability, while 14 had excessive variability between readers despite good intraobserver agreement. There is considerable variability in fetal echocardiographic quantification in clinical practice that may affect the design of multicenter fetal echocardiographic Z score studies, and not all measurements may be feasible for standard normalization. As missingness was substantial, a prospective design will be needed. Data from this pilot study may aid in the calculation of sample sizes and inform thresholds for distinguishing clinically significant from statistically significant effects.

A data-efficient zero-shot and few-shot Siamese approach for automated diagnosis of left ventricular hypertrophy

Left ventricular hypertrophy (LVH) is a life-threatening condition in which the muscle of the left ventricle thickens and enlarges. Echocardiography is a test performed by cardiologists and echocardiographers to diagnose this condition. The manual interpretation of echocardiography tests is time-consuming and prone to errors. To address this issue, we have developed an automated LVH diagnosis technique using deep learning. However, the availability of medical data is a significant challenge due to varying industry standards, privacy laws, and legal constraints. To overcome this challenge, we have proposed a data-efficient technique for automated LVH classification using echocardiography. Firstly, we collected our own dataset of normal and LVH echocardiograms from 70 patients in collaboration with a clinical facility. Secondly, we introduced novel zero-shot and few-shot algorithms based on a modified Siamese network to classify LVH and normal images. Unlike traditional zero-shot learning approaches, our proposed method does not require text vectors, and classification is based on a cutoff distance. Our model demonstrates superior performance compared to state-of-the-art techniques, achieving up to 8% precision improvement for zero-shot learning and up to 11% precision improvement for few-shot learning approaches. Additionally, we assessed the inter-observer and intra-observer reliability scores of our proposed approach against two expert echocardiographers. The results revealed that our approach achieved better inter-observer and intra-observer reliability scores compared to the experts.

Evaluating Independent Echocardiography Interpretation Skills: A Novel Assessment Tool.

Echocardiography is an essential imaging modality for clinical care. Currently, by virtue of graduating from cardiology residency and performing/interpreting an arbitrary number of echocardiograms, cardiology trainees are deemed competent to independently practice echocardiography. There is no formal assessment to ensure their competence. Recognizing this unmet need, we developed a novel echocardiography assessment tool and piloted it in a sample of 38 participants (including 7 expert echocardiographers). The tool was found to be valid and reliable in assessing independent echocardiography interpretation skills; thus can be used during training to ensure cardiology residents achieve competence in echocardiography before completing their training.

Serial image interpretation tasks improve accuracy and increase confidence in Level 1 echocardiography reporting: a pilot study

BackgroundFocused echocardiography is increasingly used in acute and emergency care, with point-of-care ultrasound integrated into several specialist training curricula (e.g. Emergency Medicine, Cardiology, Critical Care). Multiple accreditation pathways support development of this skill but there is scant empirical evidence to inform selection of teaching methods, accreditation requirements or quality assurance of education in focussed echocardiography. It has also been noted that access to in-person teaching can be a barrier to completing accreditation programmes, and that this may affect learners disproportionately depending on the location or nature of their institution. The purpose of the study was to determine whether serial image interpretation tasks as a distinct learning tool improved novice echocardiographers’ ability to accurately identify potentially life-threatening pathology from focused scans. We also aimed to describe the relationship between accuracy of reporting and participants’ confidence in those reports, and to assess users’ satisfaction with a learning pathway that could potentially be delivered remotely.Methods27 participants from a variety of healthcare roles completed a program of remote lectures and 2 in-person study days. During the program they undertook 4 ‘packets’ of 10 focused echocardiography reporting tasks (total = 40) based on images from a standardised dataset. Participants were randomized to view the scans in varying orders. Reporting accuracy was compared with consensus reports from a panel of expert echocardiographers, and participants self-reported confidence in their image interpretation and their satisfaction with the learning experience.ResultsThere was a stepwise improvement in reporting accuracy with each set of images reported, from an average reporting score of 66% for the 1st packet to 78% for the 4th packet. Participants felt more confident in identifying common life-threatening pathologies as they reported more echocardiograms. The correlation between report accuracy and confidence in the report was weak and did not increase during the study (rs = 0.394 for the 1st packet, rs = 0.321 for the 4th packet). Attrition during the study related primarily to logistical issues. There were high levels of satisfaction amongst participants, with most reporting that they would use and / or recommend a similar teaching package to colleagues.ConclusionsHealthcare professionals undertaking remote training with recorded lectures, followed by multiple reporting tasks were capable of interpreting focused echocardiograms. Reporting accuracy and confidence in identifying life-threatening pathology increased with the number of scans interpreted. The correlation between accuracy and confidence for any given report was weak (and this relationship should be explored further given the potential safety considerations). All components of this package could be delivered via distance learning to enhance the flexibility of echocardiography education.

Gaze patterns reflect and predict expertise in dynamic echocardiographic imaging.

Echocardiography is the most important modality in cardiac imaging. Rapid valid visual assessment is a critical skill for image interpretation. However, it is unclear how skilled viewers assess echocardiographic images. Therefore, guidance and implicit advice are needed for learners to achieve valid image interpretation. Using a signal detection approach, we compared 15 certified experts with 15 medical students in their diagnostic decision-making and viewing behavior. To quantify attention allocation, we recorded eye movements while viewing dynamic echocardiographic imaging loops of patients with reduced ejection fraction and healthy controls. Participants evaluated left ventricular ejection fraction and image quality (as diagnostic and visual control tasks, respectively). Experts were much better at discriminating between patients and healthy controls ( of 2.58, versus 0.98 for novices). Eye tracking revealed that experts fixated diagnostically relevant areas earlier and more often, whereas novices were distracted by visually salient task-irrelevant stimuli. We show that expertise status can be almost perfectly classified either based on judgments or purely on eye movements and that an expertise score derived from viewing behavior predicts diagnostic quality. Judgments and eye tracking revealed significant differences between echocardiography experts and novices that can be used to derive numerical expertise scores. Experts have implicitly learned to ignore the salient motion cue presented by the mitral valve and to focus on the diagnostically more relevant left ventricle. These findings have implications for echocardiography training, objective characterization of echocardiographic expertise, and the design of user-friendly interfaces for echocardiography.

Surgical treatment for obstructive hypertrophic cardiomyopathy: a five-year single-center experience of 421 cases

Objectives: To examine the short-term and mid-term effects of surgical treatment of obstructive hypertrophic cardiomyopathy (HCM) in one center. Methods: The perioperative data and short-term follow-up outcomes of 421 patients with obstructive HCM who received surgical treatment at Department of Cardiac Surgery, Zhongshan Hospital, Fudan University from January 2017 to December 2021 were analyzed retrospectively. There were 207 males and 214 females, aged (56.5±11.7) years (range: 19 to 78 years). Preoperative New York Heart Association (NYHA) classification included 45 cases of class Ⅱ, 328 cases in class Ⅲ, and 48 cases in class Ⅳ. Fifty-eight patients were diagnosed with latent obstructive HCM and 257 patients had moderate or more mitral regurgitation with 56 patients suffering from intrinsic mitral valve diseases. All procedures were completed by a multidisciplinary team, including professional echocardiologists involving in preoperative planning for proper mitral valve management strategies and intraoperative monitoring. A total of 338 patients underwent septal myectomy alone, and 59 patients underwent mitral valve surgery along with myectomy. A single transaortic approach was used in 355 patients, and a right atrial-atrial septal/atrial sulcus approach was used in 51 other patients. Long-handled minimally invasive surgical instruments were used for the procedures. Student t test, Wilcoxon rank sum test, χ2 test or Fisher exact test were used to compare the data before and after surgery. Results: The aortic cross-clamping time of septal myectomy alone was (34.3±8.5) minutes (range: 21 to 94 minutes). Eighteen patients had intraoperative adverse events and underwent immediate reoperation, including residual obstruction (10 patients), left ventricular free wall rupture (4 patients), ventricular septal perforation (3 patients), and aortic valve perforation (1 patient). Four patients died during hospitalization, and 11 patients developed complete atrioventricular block requiring permanent pacemaker implantation. After discharge, 384 (92.1%) patients received a follow-up visit with a median duration of 9 months. All follow-up patients survived with significantly improved NYHA classifications: 216 patients in class Ⅰ and 168 patients in class Ⅱ (χ2=662.73, P<0.01 as compared to baseline). At 6 months after surgery, follow-up echocardiography showed that the thickness of the ventricular septum ((13.6±2.5) mm vs. (18.2±3.0) mm, t=23.51, P<0.01) and the peak left ventricular outflow tract gradient ((12.0±6.3) mmHg vs. (93.4±19.8) mmHg, 1 mmHg=0.133 kPa, t=78.29, P<0.01) were both significantly lower than baseline values. Conclusion: The construction of the surgical team (including echocardiography experts), proper mitral valve management strategies, identification and management of sub-mitral-valve abnormalities, and application of long-handled minimally invasive surgical instruments are important for the successful implementation of septal myectomy with satisfactory short-and medium-term outcomes.

Abbreviated Echocardiographic Screening for Rheumatic Heart Disease by Nonexperts with and without Offsite Expert Review: A Diagnostic Accuracy Study

Early detection of rheumatic heart disease (RHD) through echocardiographic screening can facilitate early access to effective treatment, which reduces the risk for progression. Accurate, feasible approaches to echocardiographic screening that can be incorporated into routine health services are needed. The authors hypothesized that offsite expert review could improve the diagnostic accuracy of nonexpert-obtained echocardiographic images. This prospective cross-sectional study was performed to evaluate the diagnostic accuracy of health worker-conducted single parasternal long-axis view with a sweep of the heart using hand-carried ultrasound for the detection of RHD in high-risk populations in Timor-Leste and Australia. In the primary analysis, the presence of any mitral or aortic regurgitation met the criteria for a positive screening result. Sensitivity and specificity were calculated for a screen-and-refer approach based on nonexpert practitioner assessment (approach 1) and for an approach using offsite expert review of nonexpert practitioner-obtained images to decide onward referral (approach 2). Each participant had a reference test performed by an expert echocardiographer on the same day as the index test. Diagnosis of RHD was determined by a panel of three experts, using 2012 World Heart Federation criteria. The prevalence of borderline or definite RHD among 3,329 participants was 4.0% (95% CI, 3.4%-4.7%). The sensitivity of approach 1 for borderline or definite RHD was 86.5% (95% CI, 79.5%-91.8%), and the specificity was 61.4% (95% CI, 59.7%-63.1%). Approach 2 achieved similar sensitivity (88.4%; 95% CI, 81.5%-93.3%) and improved specificity (77.1%; 95% CI, 75.6%-78.6%). Nonexpert practitioner-obtained single parasternal long-axis view with a sweep of the heart images, reviewed by an offsite expert, can detect borderline and definite RHD on screening with reasonable sensitivity and specificity. Brief training of nonexpert practitioners with ongoing support could be used as an effective strategy for scaling up echocardiographic screening for RHD in high-risk settings.

Machine learning for the real-time assessment of left ventricular ejection fraction in critically ill patients: a bedside evaluation by novices and experts in echocardiography

BackgroundMachine learning algorithms have recently been developed to enable the automatic and real-time echocardiographic assessment of left ventricular ejection fraction (LVEF) and have not been evaluated in critically ill patients.MethodsReal-time LVEF was prospectively measured in 95 ICU patients with a machine learning algorithm installed on a cart-based ultrasound system. Real-time measurements taken by novices (LVEFNov) and by experts (LVEFExp) were compared with LVEF reference measurements (LVEFRef) taken manually by echo experts.ResultsLVEFRef ranged from 26 to 80% (mean 54 ± 12%), and the reproducibility of measurements was 9 ± 6%. Thirty patients (32%) had a LVEFRef < 50% (left ventricular systolic dysfunction). Real-time LVEFExp and LVEFNov measurements ranged from 31 to 68% (mean 54 ± 10%) and from 28 to 70% (mean 54 ± 9%), respectively. The reproducibility of measurements was comparable for LVEFExp (5 ± 4%) and for LVEFNov (6 ± 5%) and significantly better than for reference measurements (p < 0.001). We observed a strong relationship between LVEFRef and both real-time LVEFExp (r = 0.86, p < 0.001) and LVEFNov (r = 0.81, p < 0.001). The average difference (bias) between real time and reference measurements was 0 ± 6% for LVEFExp and 0 ± 7% for LVEFNov. The sensitivity to detect systolic dysfunction was 70% for real-time LVEFExp and 73% for LVEFNov. The specificity to detect systolic dysfunction was 98% both for LVEFExp and LVEFNov.ConclusionMachine learning-enabled real-time measurements of LVEF were strongly correlated with manual measurements obtained by experts. The accuracy of real-time LVEF measurements was excellent, and the precision was fair. The reproducibility of LVEF measurements was better with the machine learning system. The specificity to detect left ventricular dysfunction was excellent both for experts and for novices, whereas the sensitivity could be improved.Trial registration: NCT05336448. Retrospectively registered on April 19, 2022.

Right ventricular systolic strain in patients with pulmonary hypertension: clinical feasibility, reproducibility, and correlation with ejection fraction.

Right ventricular (RV) systolic function is the major determinant of prognosis in patients with pulmonary hypertension (PH) with quantitative assessment by speckle-tracking strain echocardiography emerging as a viable candidate measure. We evaluated a prospective cohort of 231 patients with known or suspected PH referred for clinical echocardiography. All underwent measurement of RV free-wall systolic strain by sonographer staff. Digital images were recorded for blinded offline assessment by an expert echocardiographer. Reproducibility was assessed using the analysis methods of Bland-Altman and the Cohen's-Kappa coefficient. RV strain was feasible in 213 (92%). The average RV systolic pressure was 59 ± 22mmHg. RV systolic strain correlated with functional class, NT-proBNP, and the degree of RV enlargement. The average free-wall systolic strain was -20 ± 7% (range -2 to -37%). The RV strain measures (clinical practice versus blinded expert) had an excellent correlation with a normal distribution (R2 0.87, p < 0.0001). By Bland-Altman analysis, the mean difference in measurement was -1.7% (95% CI -1.4 to -2.1) with a correlation of 0.93, p value of < 0.0001. The reproducibility of RV strain for clinically relevant thresholds was also excellent (Kappa coefficients 0.68-0.83). There was no effect on the variability of strain measures across body mass, pulmonary pressures, or rhythm. RV strain correlated with RV diastolic volumes and ejection fraction with RV free wall strain being the best echo predictor for a reduction in ejection fraction. Here RV systolic strain was found to be highly feasible and reproducible in clinical practice with excellent levels of agreement for clinically relevant thresholds.