Internationale Partnersuche

Innovation & Technologie Angebot

Cognitive assistance for intensive care units

Country of Origin: France
Reference Number: TOFR20190717002
Publication Date: 24 July 2019

Summary

A French SME has developed a platform, combining artificial intelligence and tele-expertise, providing cognitive assistance to intensive care units. It detects critical situations in advance and proposes a clinical path validated by experts. 
Research cooperation agreement is sought with hospitals to develop diagnostic algorithms for specific diseases.Commercial agreement with technical assistance is sought with partners whose medical devices will feed the diagnostic assistance algorithms.

Description

The French company offers its platform providing cognitive assistance to intensive care units by:

• Managing the quality of data, protocols and medical decisions 

• Detecting critical situations in advance and proposing a clinical path (diagnoses and prognoses) validated by experts 

• Connecting with experts

• Reducing administrative tasks. 



Combining artificial intelligence (AI) and tele-expertise, this platform extracts, integrates and validates patient data in order to automatically structure a relevant patient file for a resuscitator doctor. On the basis of these historical data and daily clinical observations, a diagnostic and prognostic engine will assist the resuscitator in his/her diagnosis and prognosis.



The main benefits are: 

•	For resuscitators, clinicians and researchers:

o	Resuscitation team tool, easy to use, adapted to its medical flow and integrating the necessary additional tools and analyses, saving time and improving the quality of diagnostics 

o	Easy and fast access to experts through the collaborative network of expertise.  

o	A continuous increase in the skills of the medical intensive care teams  

o	Full traceability of decisions and therefore a reduction in the threat of legal proceedings

o	Taking into account the most recent and relevant scientific publications through the justification and documentation of proposed diagnoses and prognoses

o	The provision of a database and tools allowing research projects oriented towards precision resuscitation 

o	Exclusive access to validated and secure cognitive assistants



•	For hospital administrators:

o	Reducing patient stay

o	The ability to manage multiple facilities with high quality support

o	Collaborative development of tools to assist in consultation and research, thus providing motivation for medical teams and sources of income.



The French company is interested in research cooperation agreement with hospitals in order to design and test the prototype as well as to develop medical researchs on other specific diseases. Indeed the first focus was to develop a tool for patients in post-anoxic coma. The next step will be infectious meningitis and stroke.



Industrial partners with medical devices are also sought for commercial agreement with technical assistance. Indeed first of all, such medical devices will feed the diagnostic assistance algorithms. Secondly, the platform could be used as a marketplace to promote innovations dedicated to intensive care units.

Advantages and Innovations

There are many advantages such as:



- Creation of the patient file: 

The data are either analysis reports in text format or raw data (from spot analyses or monitoring devices). The text reports are extracted and semantized in a central health database.



- Traceability:

The historical patient file includes all the information relevant for decision-making by the resuscitator as well as any motivated exchanges with the required experts.



- International standards:

The algorithms of the diagnostic and pronostic engines come from different sources. One of them is the decision-making trees recognized by international standards and reflecting the state of the art.



- Consultation tools are based on quality data for several reasons:

o	When creating the patient file, the algorithms will systematically identify missing data and analyze their consistency over time. Inconsistent data are then submitted to the resuscitation medical team for action.   

o	Hospital reports from intensive care physicians are systematically compared with data from reports directly from signal processors (MRI, EEG, etc.) 

o	The data extraction and integration platform is open and allows each user to capture all medical data.

Stage Of Development

Prototype available for demonstration

Stage Of Development Comment

The MVP (Minimum Viable Product) is intended to demonstrate the fundamental functionalities of the technological support application to the value proposition.



1. Construction and updating of the patient file

In a first step of treatment, the MVP provides an engine for building a generic patient neurological record based on hospital system data and medical reports. This engine automatically analyzes and extracts information from available data using syntactic and semantic analysis and artificial intelligence technologies.

The requesting physician can consult this patient file but also modify and enrich it through the interface.



2. Diagnosis of the patient file

In a second stage of treatment, the MVP provides a diagnostic engine based on the patient file and according to predefined algorithms derived from the expertise of doctors and the state of the art in medicine and research.



3. Remote expertise

In the case of an absence of diagnosis, and at the request of the requesting physician, the MVP puts the requesting physician in contact with an expert. He presents the established patient file to the expert doctor and opens a channel of communication between the two doctors.

The additional tests, information exchanged and diagnoses developed during this exchange are recorded to enrich a feedback loop that will be integrated into the diagnostic engine.

Requested partner

Two different types of partners are sought :



As considered research cooperation agreement:

Public and private hospitals are sought as partners.

Hospitals are willing to:  

- participate to the design and the testing of this prototype

- develop medical researches focusing on pathologies specific to ICUs (sepsis, coma, ...) in order to produce relevant algorithms.



As considered commercial agreement with technical assistance

The priority is given to Industrials or companies that develop/manage medical devices

•Tasks to be performed:

Partners producing medical datas specific to intensive care units and interested to integrate their datas and medical report into such cognitive assistants.

Dissemination Countries

Austria, Belgium, Brazil, Canada, Switzerland, Germany, Denmark, Estonia, Spain, Finland, France, Greece, Hungary, Ireland, Israel, India, Iceland, Italy, Japan, South Korea, Lithuania, Luxembourg, Latvia, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Russia, Sweden, Singapore, Slovenia, Slovakia, United Kingdom, USA