Switch mode
Horus EchoNet

Automatic analysis of echocardiograms

Product based on Deep Learning techniques that employs a three-dimensional convolutional neural network architecture for video classification. This model is applied to echocardiography to semantically segment the left ventricle and to evaluate cardiac function by calculating the left ventricular ejection fraction or LVEF.

Problems to solve

Echocardiography, the most widely used and accessible cardiac imaging modality, plays a crucial role in the evaluation of cardiac structure and function.

Left ventricular ejection fraction, or LVEF, is a highly relevant clinical indicator. However, its estimation from echocardiograms still has an important manual and time-consuming component.

Despite advances in machine learning for biomedical image analysis, video-based medical imaging, such as echocardiography, has received less attention.

Horus Echonet closes this gap by employing a 3D convolutional neural network architecture to semantically classify and segment echocardiography videos, providing comprehensive cardiac analysis, including automatic LVEF estimation.

Experience the future of video-based echocardiogram image analysis. Elevate your diagnostic capabilities with Horus Echonet.

More information

Main features

Deep learning video analysis: Take advantage of state-of-the-art deep learning models, designed specifically for echocardiograms, to perform accurate segmentation and assess cardiac function.

Left ventricular segmentation: accurately segment the left ventricle from echocardiograms, enabling detailed measurements and analysis.

Left ventricular ejection fraction (LVEF): automatic calculation of LVEF, a critical parameter for assessing cardiac function and diagnosing various cardiovascular conditions.

Improved clinical decision making: provide clinicians with comprehensive cardiac health information, aiding in the diagnosis and treatment planning of heart disease.

Reduced cardiologist workload: streamline the analysis process by automating video-based cardiac image segmentation, saving time and reducing manual effort, as Horus Echonet takes less than 0.1 seconds per echocardiogram analysis.

No items found.

More projects

AI4TeenDep

Detection of depression in adolescents

Read more

Urban Physio

Personalized physiotherapy

Read more

Horus Predictor

Predictions of various clinical indicators

Read more