The H2Hcare project aimed to develop a digital assistance platform that integrates Internet of Things (IoT) devices, machine learning and a digital assistant to support elderly people with heart failure in their transition from hospital discharge to home care. The platform provides a direct link between patients and doctors through which doctors can answer patients’ questions. The digital assistant gives patients personalized advices on the treatment they should follow and reminds them when to take their medication. Patient monitoring is done with IoT devices that collect relevant data on the patient’s health status as well as data on the daily activities performed. The monitored data is analyzed to detect if there is a deterioration in the state of health that requires an intervention by the doctor.
Due to its potential to improve transitional care processes, the project has a strong social and economic impact. The platform will ensure the coordinated transition from hospital to home, will increase the patient’s trust in the medical system and will reduce the number of hospitalizations. At the same time, it will make the work of the medical staff/caregivers easier and more efficient. From an economic perspective, the platform will contribute to reducing healthcare costs and to an efficient allocation of health resources. The scientific and technological results obtained in the project will contribute to the development of the eHealth care field by offering innovative solutions based on IoT and automatic learning algorithms that aim to improve the health of the elderly and the system of home medical care services.
H2HCare paper submitted to 2023 IEEE 19th International Conference on Intelligent Computer Communication and Processing (ICCP 2023), to be held October 26-28, 2023 in Cluj-Napoca, Romania:
Title: A Machine Learning based Platform for Remote Management of Heart Failure Patients
Abstract: Heart failure is a growing concern due to its high incidence nowadays, also representing a major cause of morbidity and mortality worldwide. In this paper we propose a web-based platform that incorporates both the clinical data prediction aspect and the continuous monitoring of the heart health. We implement multiple machine learning models that can support the doctors in the process of classification between a healthy and unhealthy situation. The platform benefits from an ETL (Extract, Transform, Load) sub-system that processes biometric data from smart wearables and displays it in customizable dashboards for a more illustrative visualization. The prediction service integrates three Machine Learning (ML) techniques, namely Logistic Regression, Naïve Bayes Classifier, and a custom Artificial Neural Network responsible for making classifications on the monitored data. The results illustrate that the proposed solution’s usage for performing remote monitoring and heart health assessment is feasible, obtaining promising accuracies with the aid of a public heart failure dataset (best accuracy of 88.5%).
The H2HCare final trials in Switzerland have been closed. The consortium is preparing the final project deliverable with evaluation feedback and suggestions from end-users.
As an early feedback, the system has been well received by all users, and they all provided very interesting suggestions. Seniors were motivated to test the devices and got particularly reactive if something was unusual. The feedback and suggestions resulting from the trial will be used to fine tune and refine the final prototype of the H2HCare platform.
The Hospitals University of Geneva (HUG) are currently conducting a 3-month Long Field trial of the H2HCare system in Geneva. Testing began in mid-March, and installations in all SHF homes were completed in the first week of April. 5 KOMP components from the HUG and all devices connected to the H2Hcare system were installed in 5 SHF homes, where they were trained and given explanations on how to use the system. 2 informal caregivers took part of the project.
Early results are encouraging, with seniors generally finding the system attractive and wanting to try it out, particularly the fitbit for tracking their physical activity. They are in favor of digital health solutions. We only encountered a few minor connectivity bugs with the various devices at the start of installation, but these were quickly resolved.
Tellu’s Remote Patient Monitoring service has been integrated with No Isolation’s Komp service. As TelluCare provides remote health care and Komp provides social interaction, the two services complement each other. The care plan which is stored in TelluCare is synchronized with the Komp calendar, so that the schedule of medical measurements is shown on the Komp screen. Tellu’s gateway software is running on the Komp device, transferring measurements from medical devices connected with Bluetooth to the Komp and into TelluCare, where it is available for the machine learning in the project. TelluCare also sends back an acknowledgement message to the Komp screen, so that the user gets feedback. These integrations are being tested in the final Swiss trial.
May, 24-26, 2023, Cluj-Napoca: TUC has participated to the Innovation Days event, the biggest technology event in Transilvania. TUC has managed a stand where the ongoing research projects (AAL and H2020) where represented through live presentations and demos.
At the event major companies, universities, municipalities and policy makers from Romania have participated totalling more than 200 participants.
The Technical University of Cluj-Napoca organized a new edition of the “Night of the Museums” event in Cluj-Napoca, between May 12 – 13, 2023, at The Center for Urban Culture – Casino, where a unique technology exhibition took place: ExpoTech@UTCN, a foray into the universe of innovation.
During this exhibition, the participants were able to see the most diverse exhibits areas of technology, from solar and formula 1 cars, installations, robots and type applications brain control, to construction innovations.
H2HCare was represented with live technology demos towards interested stakeholders.
TUC has presented the H2HCare project in the national event Aimed (Applications of Artificial Intelligence in Medicine) workshop, Bucharest, Romania https://aiexcellence.upb.ro/workshop_aiam2023/#descriere. The presentation with the title “Ambient Intelligence for Personalized Healthcare” has been done by Prof. Ionut Anghel from TUC team.
Many stakeholders have participated to the workshop including government representatives, eHealth companies, hospitals, municipalities, etc. The feedback was positive especially from the hospital administrators and doctors that acknowledge the utility of H2HCare solution.
H2HCare has been present to the EHiN 2022 event held on 08.11.2022, main e-health conference in Norway.
Karin Sygna, product manager for Digital home monitoring in Tellu presented the inovative solutions developed by the company toghether with the main research projects use-cases they are participating to.
Chifu, V.R.; Cioara, T.; Pop, C.B.; Anghel, I.; Demjen, D.; Salomie, I. Identification of Daily Living Recurrent Behavioral Patterns Using Genetic Algorithms for Elderly Care. Appl. Sci.2022, 12, 11030. https://doi.org/10.3390/app122111030
Abstract: A person’s routine is a sequence of activities of daily living patterns recurrently performed. Sticking daily routines is a great tool to support the care of persons with dementia, and older adults in general, who are living in their homes, and also being useful for caregivers. As state-of-the-art tools based on self-reporting are subjective and rely on a person’s memory, new tools are needed for objectively detecting such routines from the monitored data coming from wearables or smart home sensors. In this paper, we propose a solution for detecting the daily routines of a person by extracting the sequences of recurrent activities and their duration from the monitored data. A genetic algorithm is defined to extract activity patterns featuring small differences that relate to the day-to-day contextual variations that occur in a person’s daily routine. The quality of the solutions is evaluated with a probabilistic-based fitness function, while a tournament-based strategy is employed for the dynamic selection of mutation and crossover operators applied for generating the offspring. The time variability of activities of daily living is addressed using the dispersion of the values of duration of that activity around the average value. The results are showing an accuracy above 80% in detecting the routines, while the optimal values of population size and the number of generations for fitness function evolution and convergence are determined using multiple linear regression analysis.