RT Conference Proceedings
T1 Human Activity Recognition From Sensorised Patient´s Data in Healthcare: A Streaming Deep Learning-Based Approach.
A1 Hurtado-Requena, Sandro José
A1 García-Nieto, José Manuel
A1 Popov, Anton
A1 Navas-Delgado, Ismael
K1 Aprendizaje automático (Inteligencia artificial)
K1 Salud
K1 Ejercicio físico
AB Physical inactivity is one of the main risk factors for mortality,and its relationship with the main chronic diseases has experiencedintensive medical research. A well-known method for assessing people’sactivity is the use of accelerometers implanted in wearables and mobilephones. However, a series of main critical issues arise in the healthcarecontext related to the limited amount of available labeled data to builda classification model. Moreover, the discrimination ability of activitiesis often challenging to capture since the variety of movement patternsin a particular group of patients (e.g. obesity or geriatric patients) islimited over time. Consequently, the proposed work presents a novel approachfor Human Activity Recognition (HAR) in healthcare to avoidthis problem. This proposal is based on semi-supervised classificationwith Encoder-Decoder Convolutional Neural Networks (CNNs) using acombination strategy of public labeled and private unlabeled raw sensordata. In this sense, the model will be able to take advantage of the largeamount of unlabeled data available by extracting relevant characteristicsin these data, which will increase the knowledge in the innermost layers.Hence, the trained model can generalize well when used in real-world usecases. Additionally, real-time patient monitoring is provided by ApacheSpark streaming processing with sliding windows. For testing purposes,a real-world case study is conducted with a group of overweight patientsin the healthcare system of Andalusia (Spain), classifying close to 30TBs of accelerometer sensor-based data.
PB SISTEDES
YR 2023
FD 2023
LK https://hdl.handle.net/10630/27669
UL https://hdl.handle.net/10630/27669
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 21 ene 2026