TUCUVI Voice and Artificial Intelligence to reduce the impact of chronic diseases

Información del proyecto

TUCUVI

Identificador del acuerdo de subvención: 190184468

DOI

10.3030/190184468

Proyecto cerrado

Fecha de la firma de la CE 24 Enero 2023

Fecha de inicio 1 Febrero 2023

Fecha de finalización 31 Julio 2025

Financiado con arreglo a

• The European Innovation Council (EIC)

Coste total

€ 3 634 848,00

Aportación de la UE

€ 2 500 000,00

Coordinado por

TUCUVI CARE SOCIEDAD LIMITADA
Spain

Periodic Reporting for period 2 - TUCUVI (TUCUVI Voice and Artificial Intelligence to reduce the impact of chronic diseases)

Período documentado: 2024-02-01 hasta 2025-07-31

Resumen del contexto y de los objetivos generales del proyecto

TUCUVI Care, S.L. is a HealthTech company that helps healthcare organisations to automate outpatient communication using state-of-the-art voice conversational AI. It has a clinically validated and customizable portfolio of care processes automated, designed to improve the quality of care and increase efficiency in the healthcare system while delivering real value to healthcare providers, payors, and patients, with the highest patient engagement in the market.

Tucuvi Health Manager is an AI-platform to automate medical phone consultations. LOLA, our medical assistant is able to phone call patients (and also to receive phone calls from patients) and maintain natural, empathic conversations for any pathology and any care process. At the beginning of this project, the product was available in Spanish and some initial testing was being done in English and Portuguese and LOLA had experienced automating different care pathways. One year after starting the project, we have scaled our operations to 3 languages and drafted a promising strategy to become language agnostic thanks to our newly developed AI algorithms and the synthetic data generator. Also, LOLA is actively working in 25 care pathways and is able to operate in new care pathways in a faster way thanks to the development of new features such as Tucuvi Templates and Tucuvi DIY. Finally, we’ve also increased our interoperability capabilities thanks to the development of Tucuvi FHIR API, increase our robustness and performance thanks to the development of new AI models such as the ASR redundancy model and NLP Suggestion Tool, and augment the value we provide to healthcare professionals with the new Tucuvi Medical Dashboard.

Trabajo realizado desde el comienzo del proyecto hasta el final del período abarcado por el informe y los principales resultados hasta la fecha

The project development in this first reporting period has generally followed the anticipated schedule, while experiencing some unexpected bottlenecks (mainly, patient recruitment data), the other tasks have even been running ahead of schedule.

Below you can find an overview of the project status by WP:
- WP1: In progress
- WP2: In progress
- WP3: In progress
- WP4: In progress
- WP5: In progress
- WP6: In progress

WP1: As it is detailed later in task 1.1 there has been a delay that has affected WP 1, mainly due to the difficulties encountered in data collection. However, it has worked to find an alternative path to achieve the initially proposed objectives. In general, no significant changes will occur in this WP, reaching the proposed objectives next year.

WP2: the WP progresses satisfactorily and without relevant deviations, generating the committed deliverable.

WP3: The WP has started as planned. The bulk of its execution will take place in the second year of the project.

WP4: the pre-commercialization plan and communication activities are progressing satisfactorily in accordance with the initial project planning.

WP5: there have been no relevant deviations in the way the project management is approached.

WP6: TUCUVI is working in collaboration with external experts on the ethical issues associated with the project, as contemplated in the signing of the Grant Agreement.

Avances que van más allá del estado de la técnica e impacto potencial esperado (incluida la repercusión socioeconómica y las implicaciones sociales más amplias del proyecto hasta la fecha)

The Dialogue System allows the virtual assistant to speak in a naturally human way. A proprietary ASR to substitute third parties will allow TUCUVI a better adaptation to phone-call channel, patient
vocabulary and speech due to the possibility of training the system with the data obtained in every conversation.