Skip to main content
European Commission logo print header

A Novel Framework for Contactless diagnosis and forecasting of Cardiovascular Diseases

Descripción del proyecto

Pruebas acústicas para evaluar el riesgo de paro cardíaco súbito

El paro cardíaco se produce cuando el corazón deja repentinamente de bombear sangre al resto del organismo. Los síntomas aparecen de la nada y hay poco tiempo para realizar pruebas. El paro cardíaco súbito es potencialmente mortal. Con todo, los avances en tecnologías como la vigilancia a distancia y sin contacto de pacientes pueden ayudar a predecir si (y cuándo) alguien sufrirá un paro cardíaco. En el proyecto ContactlessFramework, financiado por las Acciones Marie Skłodowska-Curie, se desarrollará un marco de predicción de riesgos, VCardiac, para diagnosticar e identificar las primeras etapas de cardiopatías y pronosticar afecciones cardíacas con anticipación, utilizando la voz humana. Se creará un conjunto de datos personalizado con dos mil muestras de voz humana para clasificar los episodios acústicos de los latidos cardíacos en normales, soplos, extrasístoles y artefactos, así como otros episodios acústicos de los latidos cardíacos no categorizados.

Objetivo

"In ancient times, doctors have followed unorganized practices to diagnose cardiac arrest conditions of healthcare patients. The followed medical procedures were not very organized and accurate even though cardiac patients were diagnosed using devices, such as a stethoscope. However, the latest advancements in the technologies such as contactless remote patient monitoring, AIoMT (Artificial Intelligence of Medical Things), advanced big data and cloud-based analytics and alerts have created a paradigm shift in healthcare and provided 24 x 7 connectivity. The study proposed a VCardiac (Contactless Cardiac Classification and Risk Prediction) Framework to diagnose and identify early stages of heart diseases and forecast their cardiac conditions in advance using the human voice. In this study, a customized dataset termed ""Cardiac-2000"" with 2000 human voice samples will be designed to classify acoustic heartbeat events such as normal, murmur, extra systole, artifact, and other unlabeled heartbeat acoustic events. The average duration of the recorded heartbeat acoustic events would be 10 to 12 seconds. The primary reason for designing a customized dataset ""cardiac-2000"" is to balance the total number of samples into categories such as normal and abnormal heartbeat acoustic events. For performance evaluations of the proposed VCardiac Framework, the collected heartbeat acoustic samples will be classified using LSTM-CNN, RNN, LSTM, Bi-LSTM, CNN, K-means Clustering, and SVM methodologies. Furthermore, the proposed VCardiac Framework will also assist in classifying Age and Gender-wise risks using methodologies such as Kaplan-Meier and Cox-regression survival analysis. These methodologies will also assist in identifying the probability risk and the 10-year risk score prediction. In the end, the proposed VCardiac Framework will be tested for various Signal to Noise ratio conditions for achieving better accuracy, effectiveness, and throughput."

Régimen de financiación

HORIZON-AG-UN - HORIZON Unit Grant

Coordinador

LINNEUNIVERSITETET
Aportación neta de la UEn
€ 222 727,68
Dirección
LINNAEUS UNIVERSITY
35195 Vaxjo
Suecia

Ver en el mapa

Región
Södra Sverige Småland med öarna Kronobergs län
Tipo de actividad
Higher or Secondary Education Establishments
Enlaces
Coste total
Sin datos