Miosotis Clinical Decision Support
Clinical Decision Support
Miosotis Clinical Decision Support platform is defined as using modern artificial intelligence technologies to support clinical decision-making, with the aim of reducing clinical error and consequently, increasing efficiency.
Today, it makes no sense to think of health IT without including the use of artificial intelligence technologies, with the hope of improving diagnostic and therapeutic accuracy and consequently, improving medical care for citizens.
The platform to be developed for this specific area of the Miosotis Program has the general objective of automating the collection, recording and processing of clinical information and clinical decision support.
In addition, the aim is to provide scientific proof of the platform's technological capacity, to support its placement on the market.
In clinical practice, it will be applied to the telemonitoring of biometric parameters in patients with chronic illnesses, both in the outpatient setting and in RSEs (Residential Structures for the Elderly) and to the processing (analysis and validation) of the data collected.
It will also be used in clinical decision support, with emphasis on screening patients in an out-of-hospital environment and in clinical decision support (diagnosis and therapy), in a hospital or out-of-hospital environment, in an emergency room, ward or clinic, in public institutions (hospitals or health centers) or private institutions (hospitals, clinics or practices).
It will also be used to provide an "online symptoms checker", which will allow citizens to get advice according to the symptoms they are experiencing.
As this is an artificial intelligence platform applied to health, it is important that it guarantees high operational reliability.
Therefore, we have defined that the algorithm to be developed will ensure that the F-Score is >98%.
The process of collecting information on biometric parameters will be automated from the sensors used.
It is necessary to ensure that the data collected and its clinical processing are reliable. In this regard, we will use Machine Learning and Prompt Engineering technics.
The decision-making process and support for the automation of clinical decision-making will be managed using natural language interpretation, LLM (Large Language Models) and Prompt Engineering tools.
The main results expected are:
-
Automation in the collection and recording of biometric parameters;
-
Greater efficiency in clinical decision-making;
-
Automated and remote interpretation of complementary diagnostic tests;
-
Support for triage in in peripheral locations;
-
Support for follow-up consultations;
-
Avoidance of unnecessary patient travel;
-
Greater standardization of clinical practice;
-
Significant improvement in clinical practice;