TransCare technologies

Tellu, Karde and TUC will provide the technologies in the TransCare.

TransCare system overview

TUC

TUC will develop a machine learning (ML)-based post discharge analytics that analyses the data monitored on daily life activity and vital signs to determine insights that can be used by medical professionals to address the problems that require proactive intervention.

The service uses a Transformer deep-learning architecture for predicting heart rate during physical activities like running, walking, swimming or cycling, for a specific patient. The ML model uses monitored data captured through the Tellu Dialogg application: (i) intraday Fitbit timeseries data, (ii) daily summary information and (iii) data from other sensors or devices integrated into the remote monitoring platform (e.g. smart blood pressure meter, etc.). The model outputs inferences (e.g. predictions of heart rate during specific activities) allowing to compare the actual and values with the personalized predictions. Historical data sets with heart rate and activities data points are used to initially train the model. By focusing on the most relevant parts of the data, the model can learn both short-term and long-term trends.

Tellu

Tellu provides the IoT monitoring infrastructure, as well as the management and user interfaces for health personnel, for the TransCare system. This is built on the TelluCare platform and its Remote Patient Monitoring service

The TelluCare platform provides a solid and secure basis for medical and care services running in the cloud, and is used by municipalities and hospitals. It includes a state-of-the-art identity broker for authentication, which can be configured to forward authentication requests to any other standards-compliant identity broker. Together with fully encrypted data storage and role-based access control, it fulfils all security requirements for the very strict e-health domain. The system has an extensive web interface for healthcare and other professionals, with views tailored to the many different roles and services. Tellu also maintains integrations with Electronic Patient Journal systems used by customers. In TransCare, Tellu has been working on integrating Remote Patient Monitoring with the other TelluCare services of Remote Supervision and Response, to improve transitional care pathways. These services include smarthome sensors, camera surveillance and safety alarms with call centre.

The figure above shows the core features of the monitoring infrastructure as it has now been implemented in TransCare. On the right side, health personnel manage patients in TelluCare, with care plans that schedule measurement and questionnaire tasks. On the left side, the patient is using the Dialogg mobile app and sensor devices. The patient is prompted to make the measurements in the app based on the scheduled tasks in the care plan, and to transfer data collected by an activity tracker. Dialogg connects to devices over Bluetooth, to transfer the measurements made. They are sent to two places in parallel – to TelluCare where the measurements are stored and can be viewed in the traditional web interface, and to the ML Analytics where they are stored and processed. We are now working on showing the data collected in ML Analytics as graphs in TelluCare, making this new insight available to the health personnel

Karde

Below is a sketch of the Memas architecture. The support persons configure the app for the user by the Adminweb service. The user in transitional care uses the app.

Karde provides, via the life mastering assistant Memas, the communication platform between the patient and the carers (formal and informal).

The version of Memas used in the TransCare project is called Memas Rehabilitation. It is intended for people who are undergoing rehabilitation at home after a hospital stay. In such cases, the users, if capeable, can be their own helper by configuring the app with the necessary information.

Memas Rehabilitation has services for:

Calendar
Contacts
Information about the disease
Recommendations for activity and exercise
Recommendations about diet
Wellness reports
Link to the Dialogg app in TelluCare

An authoring tools has been provided which supports entering descriptions of diagnosises, activity and diet recommendations in additin to constructing questionnaires for wellness reports.

Calendar can be important for keeping track of activities during rehabilitation, and to some extent also other appointments and activities. This can be: Take your medication, (some times several times a day every day in the form of recurring appointments), joint training (which is quite common during rehabilitation), etc.

Contacts contains a list of close contacts (family, friends, therapists) with the possibility of making phone or video calls. The contacts are entered on Adminweb.

Information about the disease. Memas contains a wide range of disease descriptions, including comorbidity, where e.g. the user is in rehabilitation after COPD, heart failure and diabetes. Which disease is selected by the Adminweb user, who can often be the app user himself.

Activity and training. Depending on the disease, the app will contain training advice. This is automatically configured when the disease is selected in Adminweb.

Wellness reports gives the user the opportunity to report how he/she is feeling. This can be useful feedback for helpers/support persons/therapists. It may provide a basis for introducing corrective measures so that the user is not hospitalized again.

Link to the Dialogg app. This gives a seamless transitition to the Dialogg app.

Below is a picture of the Memas Rehabilitation app.