Deep Learning
Studies in this Category
SCG With Your Phone: Diagnosis of Rhythmic Spectrum Disorders in Field Conditions
This study shows that smartphones can reliably monitor heart rhythms using vibrations from the chest, thanks to advanced AI that works even in noisy, real-world conditions.
Deep Learning Predicts Cardiac Output from Seismocardiographic Signals in Heart Failure
This study shows that wearable sensors using chest vibrations and heart signals can estimate heart function as accurately as invasive tests, offering a safer and more accessible option for heart failure patients.
From video to vital signs: a new method for contactless multichannel seismocardiography
This study shows that smartphone videos can track heart vibrations using QR stickers on the chest, offering a low-cost way to monitor heart health and detect issues early, with accuracy comparable to clinical tools.
Deep learning-based beat-to-beat delineation of heart sounds and fiducial points in seismocardiography
This study developed an AI tool that accurately detects key heart vibration points, enabling better heart monitoring for patients with or without heart disease.
Extracting Cardiovascular-Induced Chest Vibrations from Ordinary Chest Videos: A Comparative Study
This study shows that smartphone videos can accurately track heart vibrations using advanced computer vision methods, offering a comfortable and non-invasive way to monitor heart health.
Deep Learning for identifying systolic complexes in SCG traces: a cross-dataset analysis
This study shows how deep learning can identify heart activity from chest vibrations, even in real-world conditions, by using data from multiple sensors and personalizing the model for each user.
A deep learning approach to using wearable seismocardiography (SCG) for diagnosing aortic valve stenosis and predicting aortic hemodynamics obtained by 4D flow MRI
This study shows that wearable heart vibration sensors combined with AI can predict blood flow and diagnose aortic valve problems as accurately as advanced MRI scans, offering a cheaper and faster alternative for heart disease screening.
Cross-Domain Detection of Pulmonary Hypertension in Human and Porcine Heart Sounds
This study shows that heart sound recordings from pigs can help train AI models to detect pulmonary hypertension in humans, offering a non-invasive and accurate alternative to invasive procedures like catheterization.
Synthetic Seismocardiography Signal Generation by a Generative Adversarial Network
Researchers used AI to create realistic heart vibration signals, helping scientists train heart-monitoring systems without needing expensive patient data collection.
Introducing the Electromechanical Risk Factor Score Derived from Seismocardiography for Estimating the Likelihood of Coronary Artery Disease
This study developed a new heart vibration-based score that better detects coronary artery disease, reducing false positives compared to current methods.
End-to-end sensor fusion and classification of atrial fibrillation using deep neural networks and smartphone mechanocardiography
This study shows that smartphones can detect atrial fibrillation (AFib) using vibrations from the chest with high accuracy, offering a practical and affordable heart monitoring solution.
Heartbeat Detection in Seismocardiograms with Semantic Segmentation
This study shows that a deep learning model can accurately detect heartbeats from chest vibrations, offering a promising alternative to traditional ECG-based methods for heart monitoring.
BIOWISH: Biometric Recognition using Wearable Inertial Sensors detecting Heart Activity
This research shows how wearable sensors can use heart vibrations to identify people with high accuracy, even weeks after enrollment. It also demonstrates how these sensors can recognize activities like walking or lying down, making them useful for secure health monitoring.
Detecting Aortic Stenosis Using Seismocardiography and Gryocardiography Combined with Convolutional Neural Networks
This study shows that heart vibrations measured by wearable sensors and analyzed with AI can detect aortic stenosis with over 98% accuracy, offering a simpler alternative to traditional echocardiography.
Contactless Seismocardiography via Deep Learning Radars
This research shows how radar and AI can monitor heart vibrations without physical contact, achieving accuracy similar to clinical ultrasound for detecting key heart movements.
Accurate Detection of Dobutamine-induced Haemodynamic Changes by Kino-Cardiography: A Randomised Double-Blind Placebo-Controlled Validation Study
This study shows that a wearable device measuring body vibrations can accurately track heart function changes caused by medication, offering a new way to monitor heart health non-invasively.
Determining the Respiratory State From a Seismocardiographic Signal--A Machine Learning Approach
This study shows that chest vibrations from the heart (SCG signals) can predict breathing patterns using advanced machine learning, with neural networks being the most accurate method. This could help monitor breathing and heart health more easily and affordably.
Wearable Seismocardiography
This study shows that wearable devices can use heart vibrations and AI to diagnose aortic valve problems and predict blood flow metrics as accurately as advanced MRI scans, offering a cheaper and faster alternative for heart health monitoring.