RESUMO
Aims: Machine-learning (ML)-based automated measurement of echocardiography images emerges as an option to reduce observer variability. The objective of the study is to improve the accuracy of a pre-existing automated reading tool ('original detector') by federated ML-based re-training. Methods and results: Automatisierte Vermessung der Echokardiographie was based on the echocardiography images of n = 4965 participants of the population-based Characteristics and Course of Heart Failure Stages A-B and Determinants of Progression Cohort Study. We implemented federated ML: echocardiography images were read by the Academic Core Lab Ultrasound-based Cardiovascular Imaging at the University Hospital Würzburg (UKW). A random algorithm selected 3226 participants for re-training of the original detector. According to data protection rules, the generation of ground truth and ML training cycles took place within the UKW network. Only non-personal training weights were exchanged with the external cooperation partner for the refinement of ML algorithms. Both the original detectors as the re-trained detector were then applied to the echocardiograms of n = 563 participants not used for training. With regard to the human referent, the re-trained detector revealed (i) superior accuracy when contrasted with the original detector's performance as it arrived at significantly smaller mean differences in all but one parameter, and a (ii) smaller absolute difference between measurements when compared with a group of different human observers. Conclusion: Population data-based ML in a federated ML set-up was feasible. The re-trained detector exhibited a much lower measurement variability than human readers. This gain in accuracy and precision strengthens the confidence in automated echocardiographic readings, which carries large potential for applications in various settings.
RESUMO
BACKGROUND: Although increased left ventricular (LV) mass is associated with adverse outcomes, measured values vary widely depending on the specific technique used. Moreover, the impact of sex, age, and race on LV mass remains controversial, further limiting the clinical use of this parameter. Accordingly, the authors studied LV mass using a variety of two-dimensional and three-dimensional echocardiographic techniques in a large population of normal subjects encompassing a wide range of ages. METHODS: Transthoracic echocardiograms obtained from 1,854 healthy adult subjects (52% men) enrolled in the World Alliance Societies of Echocardiography (WASE) Normal Values Study, were divided into three age groups (young, 18-35 years; middle aged, 36-55 years; and old, >55 years). LV mass was obtained using five conventional techniques, including linear and two-dimensional methods, as well as direct three-dimensional measurement. All LV mass values were indexed to body surface area, and differences according to sex, age, and race were analyzed for each technique. RESULTS: LV mass values differed significantly among the five techniques. Three-dimensional measurements were considerably smaller than those obtained using the other techniques and were closer to magnetic resonance imaging normal values reported in the literature. For all techniques, LV mass in men was significantly larger than in women, with and without body surface area indexing. These technique- and sex-related differences were larger than measurement variability. In women, age differences in LV mass were more pronounced and depicted significantly larger values in older age groups for all techniques, except three-dimensional echocardiography, which showed essentially no differences. LV mass was overall larger in black subjects than in white or Asian subjects. CONCLUSIONS: Significant differences in LV mass values exist across echocardiographic techniques, which are therefore not interchangeable. Sex-, race-, and age-related differences underscore the need for separate population specific normal values.
Assuntos
Ecocardiografia Tridimensional , Ventrículos do Coração , Adulto , Masculino , Pessoa de Meia-Idade , Humanos , Feminino , Idoso , Adolescente , Adulto Jovem , Valores de Referência , Ventrículos do Coração/diagnóstico por imagem , Hipertrofia Ventricular Esquerda , Ecocardiografia Tridimensional/métodos , Ecocardiografia , Função Ventricular EsquerdaRESUMO
BACKGROUND: Left ventricular (LV) circumferential strain has received less attention than longitudinal deformation, which has recently become part of routine clinical practice. Among other reasons, this is because of the lack of established normal values. Accordingly, the aim of this study was to establish normative values for LV circumferential strain and determine sex-, age-, and race-related differences in a large cohort of healthy adults. METHODS: Complete two-dimensional transthoracic echocardiograms were obtained in 1,572 healthy subjects (51% men), enrolled in the World Alliance Societies of Echocardiography Normal Values Study. Subjects were divided into three age groups (<35, 35-55, and >55 years) and stratified by sex and by race. Vendor-independent semiautomated speckle-tracking software was used to determine LV regional circumferential strain and global circumferential strain (GCS) values. Limits of normal for each measurement were defined as 95% of the corresponding sex and age group falling between the 2.5th and 97.5th percentiles. Intergroup differences were analyzed using unpaired t tests. RESULTS: Circumferential strain showed a gradient, with lower magnitude at the mitral valve level, increasing progressively toward the apex. Compared with men, women had statistically higher magnitudes of regional and global strain. Older age was associated with a stepwise increase in GCS despite an unaffected ejection fraction, a decrease in LV volume, and relatively stable global longitudinal strain in men, with a small gradual decrease in women. Asian subjects demonstrated significantly higher GCS magnitudes than whites of both sexes and blacks among women only. In contrast, no significant differences in GCS were found between white and black subjects of either sex. Importantly, despite statistical significance of these differences across sex, age, and race, circumferential strain values were similar in all groups, with variations of the order of magnitude of 1% to 2%. Notably, no differences in GCS were found among brands of imaging equipment. CONCLUSION: This study established normal values of LV regional circumferential strain and GCS and identified sex-, age-, and race-related differences when present.
Assuntos
Disfunção Ventricular Esquerda , Função Ventricular Esquerda , Adulto , Masculino , Humanos , Feminino , Pessoa de Meia-Idade , Valores de Referência , Ecocardiografia/métodos , Ventrículos do Coração/diagnóstico por imagem , Voluntários SaudáveisRESUMO
AIMS: Aortic valve area (AVA) used for echocardiographic assessment of aortic stenosis (AS) has been traditionally interpreted independently of sex, age and race. As differences in normal values might impact clinical decision-making, we aimed to establish sex-, age- and race-specific normative values for AVA and Doppler parameters using data from the World Alliance Societies of Echocardiography (WASE) Study. METHODS AND RESULTS: Two-dimensional transthoracic echocardiographic studies were obtained from 1903 healthy adult subjects (48% women). Measurements of the left ventricular outflow tract (LVOT) diameter and Doppler parameters, including AV and LVOT velocity time integrals (VTIs), AV mean pressure gradient, peak velocity, were obtained according to ASE/EACVI guidelines. AVA was calculated using the continuity equation. Compared with men, women had smaller LVOT diameters and AVA values, and higher AV peak velocities and mean gradients (all P < 0.05). LVOT and AV VTI were significantly higher in women (P < 0.05), and both parameters increased with age in both sexes. AVA differences persisted after indexing to body surface area. According to the current diagnostic criteria, 13.5% of women would have been considered to have mild AS and 1.4% moderate AS. LVOT diameter and AVA were lower in older subjects, both men and women, and were lower in Asians, compared with whites and blacks. CONCLUSION: WASE data provide clinically relevant information about significant differences in normal AVA and Doppler parameters according to sex, age, and race. The implementation of this information into clinical practice should involve development of specific normative values for each ethnic group using standardized methodology.
Assuntos
Estenose da Valva Aórtica , Valva Aórtica , Masculino , Humanos , Feminino , Idoso , Valva Aórtica/diagnóstico por imagem , Ecocardiografia/métodos , Estenose da Valva Aórtica/diagnóstico por imagem , Ultrassonografia Doppler , Ventrículos do Coração/diagnóstico por imagemRESUMO
BACKGROUND: Normal values for three-dimensional (3D) right ventricular (RV) size and function are not well established, as they originate from small studies that involved predominantly white North American and European populations, did not use RV-focused views, and relied on older 3D RV analysis software. The World Alliance Societies of Echocardiography study was designed to generate reference ranges for normal subjects around the world. The aim of this study was to assess the worldwide capability of 3D imaging of the right ventricle and report size and function measurements, including their dependency on age, sex, and ethnicity. METHODS: Healthy subjects free of cardiac, pulmonary, and renal disease were prospectively enrolled at 19 centers in 15 countries, representing six continents. Three-dimensional wide-angle RV data sets were obtained and analyzed using dedicated RV software (TomTec) to measure end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume, and ejection fraction (EF). Results were categorized by sex, age (18-40, 41-65, and >65 years) and ethnicity. RESULTS: Of the 2,007 subjects with attempted 3D RV acquisitions, 1,051 had adequate image quality for confident measurements. Upper and lower limits for body surface area-indexed EDV, ESV, and EF were 48 and 95 mL/m2, 19 and 43 mL/m2, and 44% and 58%, respectively, for men and 42 and 81 mL/m2, 16 and 36 mL/m2, and 46% and 61%, respectively, for women. Men had significantly larger EDVs, ESVs, and stroke volumes (even after body surface area indexing) and lower EFs than women (P < .05). EDV and ESV did not show any meaningful differences among age groups. Three-dimensional RV volumes were smallest in Asians. CONCLUSIONS: Reliability of 3D RV acquisition is low worldwide, underscoring the importance of future improvements in imaging techniques. Sex and race must be taken into consideration in the assessment of both RV volumes and EF.
Assuntos
Ecocardiografia Tridimensional , Ventrículos do Coração , Masculino , Humanos , Feminino , Idoso , Ventrículos do Coração/diagnóstico por imagem , Valores de Referência , Reprodutibilidade dos Testes , Volume Sistólico , Ecocardiografia , Ecocardiografia Tridimensional/métodos , Função Ventricular DireitaRESUMO
BACKGROUND: Echocardiography remains the most widely used modality to assess left ventricular (LV) chamber size and function. Currently this assessment is most frequently performed using two-dimensional (2D) echocardiography. However, three-dimensional (3D) echocardiography has been shown to be more accurate and reproducible than 2D echocardiography. Current normative reference values for 3D LV analysis are based predominantly on data from North America and Europe. The World Alliance Societies of Echocardiography study was designed to sample normal subjects from around the world to provide more universal global reference ranges. The aim of this study was to assess the worldwide feasibility of LV 3D echocardiography and report on size and functional measurements. METHODS: A total of 2,262 healthy subjects were prospectively enrolled from 19 centers in 15 countries. Three-dimensional LV full-volume data sets were obtained and analyzed offline using vendor-neutral software. Measurements included LV end-diastolic and end-systolic volumes, LV ejection fraction (LVEF), global longitudinal strain (GLS), and global circumferential strain. Results were categorized by age (18-40, 41-65, and >65 years), sex, and race. RESULTS: A total of 1,589 subjects (feasibility 70%) had adequate LV data sets for analysis. Mean normal values for indexed end-diastolic volume, end-systolic volume, and LVEF in men and women were 70 ± 15 and 65 ± 12 mL/m2, 28 ± 7 and 25 ± 6 mL/m2, and 60 ± 5% and 62 ± 5%, respectively. Men had larger LV volumes and lower LVEFs than women. GLS and global circumferential strain were higher in magnitude in women. In both sexes, LV volumes were lower and LVEF tended to be higher with increasing age, especially considering the differences between the youngest and oldest age groups. Although GLS was similar across age groups in men, in women, the youngest and middle-age cohorts revealed higher magnitudes of GLS compared with the oldest age group. Global circumferential strain was higher in magnitude at older age in both men and women. Finally, Asians had smaller chamber sizes and higher LVEFs and absolute strain values than both blacks and whites. CONCLUSIONS: Age, sex, and race should be considered when defining normal reference values for LV dimension and functional parameters obtained by 3D echocardiography.
Assuntos
Ecocardiografia Tridimensional , Disfunção Ventricular Esquerda , Idoso , Ecocardiografia/métodos , Ecocardiografia Tridimensional/métodos , Feminino , Ventrículos do Coração/diagnóstico por imagem , Humanos , Masculino , Pessoa de Meia-Idade , Valores de Referência , Volume Sistólico , Função Ventricular EsquerdaRESUMO
BACKGROUND: Left atrial (LA) evaluation includes volumetric and functional parameters with an abundance of diagnostic and prognostic implications. Solid normal reference ranges are compulsory for accurate interpretation in individual patients, but previous studies have yielded mixed conclusions regarding the effects of age, sex, and/or race. The present report from the World Alliance Societies of Echocardiography study focuses on two-dimensional (2D) and three-dimensional (3D) measures of LA structure and function, with subgroup analysis by age, sex, and race. METHODS: Transthoracic 2D and 3D echocardiographic images were obtained in 1,765 healthy individuals (901 men, 864 women) evenly distributed among age subgroups: 18 to 40 years (n = 745), 41 to 65 years (n = 618), and >65 years (n = 402); the racial distribution was 38.4% white, 39.9% Asian, and 9.7% black. Images were analyzed using dedicated LA analysis software to measure LA volumes and phasic function from 3D volume and 2D strain curves. RESULTS: Three-dimensional maximum and minimum LA volumes adjusted for body surface area were nearly identical for men and women, but women demonstrated higher 3D total and passive emptying fractions (EFs). Two-dimensional reservoir strain was similar for both sexes. Age was associated with an incremental rise in LA volumes alongside characteristic shifts in functional indices. Total 2D EF and reservoir and conduit strain varied inversely with age, counteracted by higher booster strain, with a greater magnitude of effect in women. Active 3D EF was significantly higher, while total and passive EFs decreased with age. Interracial differences were noted in LA volumes, without substantial differences in functional indices. CONCLUSION: Although similar normal values for LA volumes and strain can be applied to both sexes, meaningful differences in LA size occur with aging. Indices of function also shift with age, with a compensatory rise in booster function, which may serve to counteract observed lower total and passive EFs. Defining age-associated normal values may help differentiate age-associated "healthy" LA aging from pathologic processes.
Assuntos
Apêndice Atrial , Ecocardiografia Tridimensional , Adolescente , Adulto , Função do Átrio Esquerdo , Ecocardiografia , Feminino , Átrios do Coração/diagnóstico por imagem , Humanos , Masculino , Valores de Referência , Adulto JovemRESUMO
BACKGROUND: Accurate measurements of the aortic annulus and root are important for guiding therapeutic decisions regarding the need for aortic surgery. Current echocardiographic guidelines for identification of aortic root dilatation are limited because current normative values were derived predominantly from white individuals in narrow age ranges, and based partially on M-mode measurements. Using data from the World Alliance Societies of Echocardiography study, the authors sought to establish normal ranges of aortic dimensions across sexes, races, and a wide range of ages. METHODS: Adult individuals free of heart, lung, and kidney disease were prospectively enrolled from 15 countries, with even distributions among sexes and age groups: young (18-40 years), middle aged (41-65 years) and old (>65 years). Transthoracic two-dimensional echocardiograms of 1,585 subjects (mean age, 47 ± 17 years; 50.4% men; mean body surface area [BSA], 1.77 ± 0.22 m2) were analyzed in a core laboratory following American Society of Echocardiography guidelines. Measurements, indexed separately by BSA and by height, included the aortic annulus, sinuses of Valsalva, and sinotubular junction. Differences among age, sex, and racial groups were evaluated using unpaired two-tailed Student's t tests. RESULTS: All aortic root dimensions were larger in men compared with women. After indexing to BSA, all measured dimensions were significantly larger in women, whereas men continued to show larger dimensions after indexing to height. Of note, the upper limits of normal for all aortic dimensions were lower across all age groups, compared with the guidelines. Aortic dimensions were larger in older age groups in both sexes, a trend that persisted regardless of BSA or height adjustment. Last, differences in aortic dimensions were also observed according to race: Asians had the smallest nonindexed aortic dimensions at all levels. CONCLUSIONS: There are significant differences in aortic dimensions according to sex, age, and race. Thus, current guideline-recommended normal ranges may need to be adjusted to account for these differences.
Assuntos
Aorta , Ecocardiografia , Adolescente , Adulto , Idoso , Aorta/diagnóstico por imagem , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Grupos Raciais , Valores de Referência , População Branca , Adulto JovemRESUMO
BACKGROUND: Although the assessment of right ventricular (RV) diastolic function is feasible, it has garnered far less momentum for use compared with its left ventricular counterpart. The scarcity of data defining normative RV diastolic function and the fact that implications of RV diastolic dysfunction in different disease states on outcomes are less well known both hinder integration into routine clinical assessment. The aim of this study was to establish normal values of RV diastolic parameters stratified by sex, age, and race using data from the World Alliance Societies of Echocardiography study. METHODS: A subset of 888 normal subjects from the World Alliance Societies of Echocardiography database were analyzed, including measurements of tricuspid valve (TV) inflow E- and A-wave velocities, E-wave deceleration time, and TV annular tissue Doppler e' and a' velocities. Additionally, right atrial (RA) maximal volume and RA peak reservoir strain were measured. Patients were grouped by age (<40, 41-65, and >65 years) and stratified by sex and race. Differences were analyzed using unpaired t tests. RESULTS: Compared with men, women had significantly higher TV e' and E-wave and A-wave velocities, though differences were modest. Increasing age was associated with stepwise lower TV E wave, e' velocity, and TV E/A ratio and higher a' velocity and E/e' ratio. RA peak reservoir strain was also lower, and RA end-systolic volume trended toward being smaller for older age groups. Asian subjects demonstrated significantly higher a' velocities, lower E wave, the smallest RA end-systolic volumes, and the lowest RA peak strain values compared with white subjects of both sexes. CONCLUSIONS: This study provides normal values for parameters used in the assessment of RV diastolic function stratified by race, sex, and age. The results demonstrate significant differences in RV diastolic parameters between age groups, which manifest in both individual parameters and composite ratios of TV inflow and annular velocities. Although limited sex- and race-related differences were also noted, age appears to have the most significant impact on RV diastolic parameters. These findings may aid in refining current normative values.
Assuntos
Ventrículos do Coração , Função Ventricular Direita , Idoso , Diástole , Ecocardiografia , Feminino , Ventrículos do Coração/diagnóstico por imagem , Humanos , Masculino , Valores de ReferênciaRESUMO
BACKGROUND: Recent advances in mitral valve (MV) percutaneous interventions have escalated the need for a more quantitative and comprehensive assessment of the MV, which can be best achieved using three-dimensional echocardiography. Understanding normal valve size, structure, and function is essential for differentiation of healthy from disease states. The aims of this study were to establish normative values for MV apparatus size and morphology and to determine how they vary across age, sex, and race groups using data from the World Alliance Societies of Echocardiography Normal Values Study. METHODS: Three-dimensional volumetric data sets obtained on transthoracic echocardiography in 748 normal subjects (51% men) were analyzed using commercial MV analysis software (TomTec Imaging Systems) to determine annular and leaflet dimensions and areas. The subjects were divided into groups by sex (378 men and 370 women) and age (18 to 40 years [n = 266], 41 to 65 years [n = 249], and >65 years [n = 233]) to identify sex- and age-related differences. In addition, differences among black, white, and Asian populations were studied. Inter- and intraobserver variability was assessed in a subset of 30 subjects and expressed as mean absolute difference between pairs of repeated measurements. RESULTS: Compared with women, men had larger annular size measurements, larger tenting size parameters, and larger leaflet length and area. Compared with the black and white populations, the Asian population showed significantly smaller mitral annular size. Although many of the age, sex, and race differences in MV parameters were statistically significant, they were comparable with or smaller than the corresponding measurement variability. Indexing to body surface area and height did not eliminate these differences consistently, suggesting that parameters may need to be indexed according to their dimensionality. CONCLUSIONS: This analysis of the World Alliance Societies of Echocardiography data provides normative values of mitral apparatus size and morphology. Although sex- and age-related differences were noted, they need to be interpreted with caution in view of the associated measurement variability.
Assuntos
Ecocardiografia Tridimensional , Insuficiência da Valva Mitral , Adolescente , Adulto , Ecocardiografia , Ecocardiografia Tridimensional/métodos , Ecocardiografia Transesofagiana/métodos , Feminino , Humanos , Masculino , Valva Mitral/diagnóstico por imagem , Software , Adulto JovemRESUMO
The objective of this study was to evaluate the use of the generation of 3D models and 3D prints of complex cases for physicians at the example of an intricate left ventricular outflow tract obstruction (LVOTO). LVOTO is a known complication of mitral valve surgery. A 38-year-old female patient with increasing dyspnoea after mitral valve replacement was referred to our centre. Echocardiography showed a strut of the bioprosthetic heart valve protruding into the left ventricular outflow tract. However, the diagnosis of a LVOTO was difficult based on echocardiography alone. Therefore, we fabricated a physical model of the left ventricular outflow tract, the mitral valve, the aortic valve and the left ventricle. With this physical model in hand, we were able to visualize the LVOTO and to discuss potential therapeutic options. Moreover, we were able to plan the subsequent redo surgery in detail using the model. This case shows the benefit of 3D printing technologies for surgeons and patients, not only for analysis, but also during the decision-making and pre-operative planning process.
Assuntos
Procedimentos Cirúrgicos Cardíacos , Insuficiência da Valva Mitral , Impressão Tridimensional , Adulto , Feminino , Humanos , Valva Mitral/diagnóstico por imagem , Valva Mitral/cirurgia , Obstrução do Fluxo Ventricular Externo/diagnóstico por imagem , Obstrução do Fluxo Ventricular Externo/etiologia , Obstrução do Fluxo Ventricular Externo/cirurgiaRESUMO
BACKGROUND: Three-dimensional echocardiographic (3DE) imaging and cardiac computed tomographic (CCT) imaging are important cardiac imaging tools. Despite the three-dimensional nature of these image acquisitions and reconstructions, they are visualized on two-dimensional monitors with shading and coloring to create the illusion of three dimensions. Virtual reality (VR) is a novel tool that allows true three-dimensional visualization and manipulation. The aims of this study were to test the feasibility of converting 3DE and CCT data into three-dimensional VR models, compare the variability of measurements performed in VR and conventional software, assess the diagnostic quality of VR models, and understand the value of VR over conventional viewing. METHODS: Custom software with clinically relevant postprocessing tools (interactively adjustable visualization parameters, multiplanar reconstructions, cropping planes, and nonplanar measurements) was developed to convert 3DE and CCT data into VR models. Anatomic measurements of 15 3DE and 15 CCT data sets of the mitral valve were compared using conventional software and in the VR environment. Additionally, the diagnostic quality of the VR models created from 3DE and CCT data sets was assessed. RESULTS: The 3DE and CCT data sets were successfully converted into VR models in <3 min. The measurement variabilities were reduced by 40% (20.1% vs 12.2%) for 3DE imaging and 34% (15.3% vs 10.1%) for CCT imaging by using VR. The mean time needed for measurements was reduced by 31% (from 61 to 42 sec) for 3DE imaging and 39% (from 37 to 23 sec) for CCT imaging. Most users reported facile manipulation of VR models, diagnostic quality visualization of the anatomy, and high confidence in the measurements. CONCLUSIONS: This study demonstrates the feasibility of converting 3DE and CCT data into diagnostic-quality VR models. Compared with conventional imaging, VR analysis is associated with faster navigation and accurate measurements with lower variability.
Assuntos
Ecocardiografia Tridimensional , Realidade Virtual , Técnicas de Imagem Cardíaca , Humanos , Valva Mitral , Tomografia Computadorizada por Raios XRESUMO
RESUMEN Introducción: El número creciente de estudios ecocardiográficos y la necesidad de cumplir rigurosamente con las recomendaciones de guías internacionales de cuantificación, ha llevado a que los cardiólogos deban realizar tareas sumamente extensas y repetitivas, como parte de la interpretación y análisis de cantidades de información cada vez más abrumadoras. Novedosas técnicas de machine learning (ML), diseñadas para reconocer imágenes y realizar mediciones en las vistas adecuadas, están siendo cada vez más utilizadas para responder a esta necesidad evidente de automatización de procesos. Objetivos: Nuestro objetivo fue evaluar un modelo alternativo de interpretación y análisis de estudios ecocardiográficos, basado fundamentalmente en la utilización de software de ML, capaz de identificar y clasificar vistas y realizar mediciones estandarizadas de forma automática. Material y métodos: Se utilizaron imágenes obtenidas en 2000 sujetos normales, libres de enfermedad, de los cuales 1800 fueron utilizados para desarrollar los algoritmos de ML y 200 para su validación posterior. Primero, una red neuronal convolucional fue desarrollada para reconocer 18 vistas ecocardiográficas estándar y clasificarlas de acuerdo con 8 grupos (stacks) temáticos. Los resultados de la identificación automática fueron comparados con la clasificación realizada por expertos. Luego, algoritmos de ML fueron desarrollados para medir automáticamente 16 parámetros de eco Doppler de evaluación clínica habitual, los cuales fueron comparados con las mediciones realizadas por un lector experto. Finalmente, comparamos el tiempo necesario para completar el análisis de un estudio ecocardiográfico con la utilización de métodos manuales convencionales, con el tiempo necesario con el empleo del modelo que incorpora ML en la clasificación de imágenes y mediciones ecocardiográficas iniciales. La variabilidad inter e intraobservador también fue analizada. Resultados: La clasificación automática de vistas fue posible en menos de 1 segundo por estudio, con una precisión de 90 % en imágenes 2D y de 94 % en imágenes Doppler. La agrupación de imágenes en stacks tuvo una precisión de 91 %, y fue posible completar dichos grupos con las imágenes necesarias en 99% de los casos. La concordancia con expertos fue excelente, con diferencias similares a las observadas entre dos lectores humanos. La incorporación de ML en la clasificación y medición de imágenes ecocardiográficas redujo un 41 % el tiempo de análisis y demostró menor variabilidad que la metodología de interpretación convencional. Conclusión: La incorporación de técnicas de ML puede mejorar significativamente la reproducibilidad y eficiencia de las interpretaciones y mediciones ecocardiográficas. La implementación de este tipo de tecnologías en la práctica clínica podría resultar en reducción de costos y aumento en la satisfacción del personal médico.
ABSTRACT Background: The growing number of echocardiographic tests and the need for strict adherence to international quantification guidelines have forced cardiologists to perform highly extended and repetitive tasks when interpreting and analyzing increasingly overwhelming amounts of data. Novel machine learning (ML) techniques, designed to identify images and perform measurements at relevant visits, are becoming more common to meet this obvious need for process automation. Objectives: Our objective was to evaluate an alternative model for the interpretation and analysis of echocardiographic tests mostly based on the use of ML software in order to identify and classify views and perform standardized measurements automatically. Methods: Images came from 2000 healthy subjects, 1800 of whom were used to develop ML algorithms and 200 for subsequent validation. First, a convolutional neural network was developed in order to identify 18 standard echocardiographic views and classify them based on 8 thematic groups (stacks). The results of automatic identification were compared to classification by experts. Later, ML algorithms were developed to automatically measure 16 Doppler scan parameters for regular clinical evaluation, which were compared to measurements by an expert reader. Finally, we compared the time required to complete the analysis of an echocardiographic test using conventional manual methods with the time needed when using the ML model to classify images and perform initial echocardiographic measurements. Inter- and intra-observer variability was also analyzed. Results: Automatic view classification was possible in less than 1 second per test, with a 90% accuracy for 2D images and a 94% accuracy for Doppler scan images. Stacking images had a 91% accuracy, and it was possible to complete the groups with any necessary images in 99% of cases. Expert agreement was outstanding, with discrepancies similar to those found between two human readers. Applying ML to echocardiographic imaging classification and measurement reduced time of analysis by 41% and showed lower variability than conventional reading methods. Conclusion: Application of ML techniques may significantly improve reproducibility and efficiency of echocardiographic interpretations and measurements. Using this type of technologies in clinical practice may lead to reduced costs and increased medical staff satisfaction.