TRIAGE SMART DECISION COVID19

Performing prior triage in hospital emergency departments, through artificial intelligence techniques, carrying out the first screening and differentiation of COVID and non-COVID users, thus avoiding contact with other patients and healthcare professionals without any need, and subsequently classifying them in terms of level of urgency of care and treatment, thus seeking to reduce risks, care failures and waiting times, obtaining better planning and adequacy of resources, improving care and quality of care.

Project Description:

Development and research of a collaborative artificial intelligence, capable of managing covid19 emergency decision making in patients attending healthcare centres.

Problem to be solve:

To develop a system based on Artificial Intelligence for the automation of emergency health triage through Natural Language Processing and Deep Learning techniques, (t-SDCOVID-19), developing an algorithm capable of carrying out the protocols of discriminated access (covid-no covid zone), and care in hospital emergency services, automatically, without contact, and with self-learning capacity, to be able to adapt to different casuistry and thereby continuously improve the reliability of the system.

Project Resolution Objectives:

On the one hand, the project aims to offer a technology that enables intelligent basic triage to be carried out, in accordance with emergency department protocols, in an automated and humanised way, thanks to the use of artificial intelligence techniques, self-learning and a multi-language and inclusive conversational system. This system must be aligned with a new automated triage strategy to reduce the length of stay in the emergency department and minimise patient waiting times.
On the other hand, it is intended to develop a derivative mobility tool that allows citizens to make responsible declarations to COVID.

In order to achieve the project’s objective, the following specific objectives have been established:

Development of artificial intelligence algorithm/s for the automation of triage in hospital emergency services.
Development of a classification and self-learning system through Deep-Learning techniques to perfect the accuracy of the Artificial Intelligence technique as retraining is carried out.
Development of a multi-language and inclusive conversational agent that allows any patient, regardless of their origin, gender and disability, visual, cognitive or auditory, who comes to the emergency department, to communicate with him/her through a “humanised” avatar that can be found in a plug and play system. This system will be in charge of capturing, through a series of questions, the nature and state of the patient, and thus proceed to carry out the first screening and triage.
To serve as a first screening of COVID and non-COVID patients, thus reducing contact with the smallest number of healthcare workers and with other patients in unsafe areas, referring them directly to different wards, depending on whether they are at risk of COVID or not, and subsequently depending on their level of severity, as a basic care planning system.
Develop a universal system for the automatic generation of responsible declarations via mobile phone, in accordance with current and future healthcare protocols, and in accordance with cybersecurity requirements, for secure access to workplaces, establishments and events.
Develop a tool to support the basic diagnosis of patients in emergencies.
Improve the user experience in the access services to emergencies, establishments and events where the COVID19 protocol must be complied with.