Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Para optimizar los resultados de la búsqueda, se recomienda utilizar términos en inglés.

Detecting current state of the congestive heart failure from heart sounds

Resumen

Tipo:
Oferta Tecnológica
Referencia:
TOSI20200831001
Publicado:
08/09/2020
Caducidad:
09/09/2021
Resumen:
A Slovenian public research organization has developed a method for detecting the current state of congestive heart failure (CHF) from heart sounds using a digital stethoscope. Patients have periods of feeling well and periods where they notice weakness, heavy breathing, which often requires hospital treatment. The system provides early detection of CHF worsening. Industrial partners are sought for licensing, technical cooperation or academia partners for research cooperation agreements.

Details

Tittle:
Detecting current state of the congestive heart failure from heart sounds
Summary:
A Slovenian public research organization has developed a method for detecting the current state of congestive heart failure (CHF) from heart sounds using a digital stethoscope. Patients have periods of feeling well and periods where they notice weakness, heavy breathing, which often requires hospital treatment. The system provides early detection of CHF worsening. Industrial partners are sought for licensing, technical cooperation or academia partners for research cooperation agreements.
Description:
The Slovenian public research organization has developed a method for detecting the current state of the congestive heart failure (CHF) from phonocardiogram (PCG) using a digital stethoscope.
Congestive heart failure (CHF) is a chronic progressive condition where the heart is unable to pump enough blood to meet the body´s needs. Typically, in the clinical progress there are periods of compensated phase (the patient feels well) and decompensated phase, where patients notice significant weakness, heavy breathing, and swelling in limbs and/or abdomen. Often, such deterioration requires hospital treatment with intravenous medications for successful recompensation. Early CHF worsening detection allows a physician to spare the patient from hospitalization.
The method combines classic machine-learning (ML) and end-to-end deep-learning (DL). The classical ML learns from a large body of expert-defined features and the DL learns both from the time-domain (i.e., the raw PCG signal) representation of the signal and the spectral representation of the signal. The method was evaluated on a dataset created at the local hospital and additionally on six publicly available PhysioNet datasets. There were 15 features identified that have different distributions depending on the CHF phase (compensated or decompensated). By using just two of these features, a simple and transparent decision tree classifier capable of distinguishing between the recompensated and the decompensated phases with an accuracy of 93.2%, calculated using a Leave-One-Subject-Out (LOSO) evaluation was built.
This is an entirely novel system, worldwide, to detect worsening of CHF. Although a trained physician can detect worsening using a stethoscope, the proposed method can be applied to a wearable device or a mobile phone. The computer algorithm developed allows the patient to monitor their health without (too) frequent visits at the doctor.
Additional originality comes from an innovative combination of knowledge from the fields of physics, medicine, electronics, and computer science.
The researchers are experts in the computational theory of intelligence. They have developed several practical applications in the field of intelligent information systems, intelligent web retrieval, medicine, language technologies and others. The group has over 20 years of experience in R-Y-D, natural language processing, and cognitive sciences
The partners are sought for among e-Health and electronic (mobile phones) industry as well as academia for licencing agreements to the secret know-how, technical cooperation agreements for joint development of the mobile phone or wearable application and research cooperation for further research in this field.
Advantages and Innovations:
The main advantages of the proposed solution:
· Improved CHF treatment.
· Fewer CHF episodes and therefore fewer hospitalizations.
· The approach can be applied using only a standard intelligent mobile phone.
· Currently, no adequate, patient-friendly and non-invasive technology would allow monitoring of CHF patients.
Stage of Development:
Under development/lab tested
IPs:
Secret Know-how,Copyright

Partner sought

Type and Role of Partner Sought:
The partners are sought for among e-Health and electronic (mobile phones) industry as well as academia for:
· Licencing agreements to the secret know-how.
· Technical cooperation agreements for joint development of the system.
· Research cooperation for further research in this pioneering field of telemedicine.

Client

Type and Size of Client:
R&D Institution
Already Engaged in Trans-National Cooperation:
Si
Languages Spoken:
English

Keywords

Technology Keywords:
01004001 Applications for Health
01003003 Artificial Intelligence (AI)
01003006 Computer Software