Titulo Estágio
A Smart Blood Pressure Monitoring System for Prevention of Hypertension Episodes
Local do Estágio
CISUC-DEI
Enquadramento
This thesis is part of the ongoing funded project POWER with Altice Labs. Hypertension, known as the “silent killer”, is a leading cause of disability and death. The development of digital solutions, aiming at the promotion and enhancement of lifestyle behaviors addressing blood pressure (BP) self-management, in order to manage the disease, improve compliance and achieve healthy living is of major importance.
Objetivo
The main goal of this thesis is to research and implement a set of algorithms and models to support the remote management of hypertension patients.
Two major applications are required two perform adequate management of hypertension using remote monitoring solutions: i) algorithms for assessing the current status and forecasting the evolution of blood pressure (BP) values, enabling the early detection of hypertension episodes; ii) recommendation solutions, addressing interventions in life habits (e.g., meal intake, exercise, etc.) based on the current status and the future evolution of patient's condition.
The research team has recently developed some algorithms that forecast the BP values based on past BP values. However, since BP is influenced by life habits, additional context information is needed for the algorithms to accurately forecast the progress of BP. So, the first research goal of the student is to design and train machine learning models that leverage additional context information to improve the accuracy of the BP forecasting models.
For this objective, a dataset will be collected in a clinical study that is planned the scope of the Power project, involving the Cardiology department of CHUC – Coimbra University Hospital Center. Nevertheless, and to avoid possible constraints in data gathering, the student will prepare a dataset of simulated data to train the models upon.
The second research goal is the development of a recommendation application, related to interventions in life habits (e.g., meal intake, exercise, etc.). To address this issue, knowledge-based approach in cooperation with experts from the Centro Hospitalar e Universitário of Coimbra (CHUC) will be researched and implemented.
Most importantly, across both tasks, the system must be adaptable in order to work with different types of data available, since in the real-life scenario not all variables might be collected and available at all timepoints.
Plano de Trabalhos - Semestre 1
▪ State of the art on blood pressure and forecasting models
▪ State of the art on recommendation rules for hypertension prevention
▪ Data simulation and preparation
▪ Initial development of forecasting models with context information
Plano de Trabalhos - Semestre 2
▪ Research and development of multivariable forecasting models, adaptative for the different types of data available
▪ Research and development of the life habits intervention recommendation module
▪ Writing the thesis
Condições
-
Observações
Co-supervised by Prof. Paulo Carvalho.
Orientador
Marco Simões
msimoes@dei.uc.pt 📩