Back to search

HELSEVEL-H-Gode og effektive helse-, omsorgs- og velferdstjenester

Design and implementation of a low cost smart system for pre­diagnosis and telecare of infectious diseases in elderly people

Awarded: NOK 1.2 mill.

The aim of the proposal of project is to build an ICT based framework to support the early diagnosis of infectious diseases. The project includes the design, development and implementation of a Hardware platform capable to support the communication of the pervasive system with the central cloud environment. The final part will be a subsystem capable to support aided clinical diagnosis in the field of infectious diseases. This subsystem will include an inference system built upon Machine Learning to improve decision support for the prevention, treatment and management of infectious illness. Apart from that, the system will be able to run in an isolated and pervasive way to avoid problems in low connected regions. Finally, there will be a centralized pattern recognition system designed to detect disease and health patterns in collected data. The main goal of this project is to provide health services and home care by using technologies of social and medical telecare, to achieve a great efficiency in the organization and management of the primary health care reducing costs. The participation of the Norwegian side of the project took place in 2018 and the first three months of 2019. It was concentrated in the development of machine learning based solutions to predict infections taking into account data available. In order to select algorithms a systematic literature review was conducted, once identified the best options a set of tests took place to select the best approach. This model is trained and exploited using physiological signals measured on the subjects, which are uploaded by nurses into the SPIDEP cloud platform. Then, the performance of different machine learning (ML) models in the diagnosis task is explored, in order to determine which technique fits better the current problem. After that the key component of the smart system for pre-diagnosis is developed. It is the responsible for providing the ?recommendations?, understood as a support tool for clinicians to determine the probability of a patient being affected by an infectious disease.

SPIDEP is a project aimed t produce a system including hardware and software. In the Norwegian side of the project, the development was mainly based on software development, and more precisely, machine learning based solutions. As indicated in the previous rapport, this has been achieved and its impact in terms of publications is remarkable and will grow in the future with a paper sent to a LEVEL 2 journal and an invitation for a special issue. Apart from the scientific result, the project was a valid instrument to attract a new researcher at Hiof working in the project, Dr. Ogul, expand the network of partners and also attract potential researchers to the institution. In the industry side, the system is still under development in its third year of the project and initial impact was notable for the two institutions that tested the prototype.

The aim of the project is to build a framework to support the early diagnosis of infectious diseases. The project includes the design, development and implementation of a HW platform capable to support the communication of the pervasive system with the central cloud environment. The final part will be a subsystem capable to support aided clinical diagnosis in the field of infectious diseases. This subsystem will include an inference system built upon Machine Learning to improve decision support for the prevention, treatment and management of infectious illness. Moreover,the system will be able to run in an isolated and pervasive way to avoid problems in low connected regions. Finally, there will be a centralized pattern recognition system designed to detect disease and health patterns in collected data. The main goal of this project is to provide health services and home care by using technologies of social and medical telecare, to achieve a great efficiency in the organization and management of the primary health care reducing costs.The expected results of this project will allow build national central databases to store and record all medical cases in one country to help developing the national statistics that are the brick and mortar for future health strategies. In addition, the recommender system will be proposed to fill in the lack of medical doctors in rural areas and to provide medical care to elderly persons living in remote areas. Integration with e­Health IT systems is another exploitation option that will be proposed to large hospitals,which offer ambulatory medical services. The contribution from Norway will be focussed on the development of the software devoted to assist in the diagnosis using the architecture and, later on, the development of the centralized pattern recognition system. Both contributions will be based on a previous formal analysis of the alternatives, election of the most convenient one and later on, development of the needed SW

Activity:

HELSEVEL-H-Gode og effektive helse-, omsorgs- og velferdstjenester