With advancements in technology, more data-driven solutions using artificial intelligence and machine learning are being implemented in various sectors such as healthcare, automation, manufacturing, and education.
“In the healthcare industry, the technological development in sensors and wearables has significantly contributed as an important data source for healthcare data. Nowadays consumer-grade wearables including smartwatches are equipped with a range of sensors such as glucose monitors, ECG monitors, accelerometers gyroscopes for activity monitoring, etc. The data from these sensors could be utilized to predict chronic diseases by analyzing them using machine-learning algorithms,” says MSc Aditi Site.
In her doctoral dissertation, Aditi Site seeks to address several questions related to health sensor data analytics such as the feasibility of using multi-sensor data for enhancing disease prediction accuracy, the possibility of using the sensor data to identify the disease symptom severity, developing interpretable, computationally efficient and optimized solutions for health predictions.
Machine learning for healthcare application
Aditi Site implemented the framework for evaluating multi-sensor prediction models. She also developed the methodology for evaluating the machine learning model’s complexity and severity predictions based on more observable data obtained from sensors.
“This dissertation could be used as the basis for further research and enhancements in the healthcare data analytics sector”, says Site.
Aditi Site comes from Jabalpur, India. She started her doctoral studies at Tampere University in August 2020. She completed her MSc from Vellore Institute of Technology, Chennai, and has worked for 3,5 years as a Software development Engineer in McAfee Software India Pvt. Ltd. (now Trellix), Bengaluru, India. She is currently working on her post-doctoral project.
Public defence on Friday 29 November
The doctoral dissertation of MSc Aditi Site in the field of Electronics and Communication Engineering titled Analysis of Sensor Data Using Machine Learning Algorithms for Health Applications will be publicly examined in the Faculty of Information Technology and Communication Sciences at Tampere University on 29 November 2024 at 12.00, in the Rakennustalo building, in room RG202 (address: Korkeakoulunkatu 1, Tampere). The Opponent will be Prof. Laura Ruotsalainen from University of Helsinki, Finland. The Custos will be Prof. Jari Nurmi from Tampere University, Finland.
The dissertation is available online.
The public defence can be followed via remote connection.