IEEE MeMeA & Health by Tech - RECENTRE WP2 on Stage
Over the past few weeks, researchers from Work Package 2 (WP2) of the 4TU RECENTRE Research Program actively presented their collaborative research at two conferences.
At IEEE MeMeA 2025 in Chania, Greece, we presented our work (📄 full paper will be available on IEEEXplore soon):
- Unsupervised detection of postoperative complications. This paper explores how LSTM autoencoders can detect early signs of complications in patients recovering at home after major abdominal cancer surgery, using wearable data. Developed in collaboration with Medisch Spectrum Twente, the goal is to support early detection for timely interventions and improve patient outcomes.
- "Sensor fusion using 1D-CNNs in atrial fibrillation detection and decision support." In this paper, the potential to enhance decision support for atrial fibrillation (AF) has been explored by applying 1D convolutional neural networks (1D-CNNs) for multi-sensor data fusion, with interpretability provided through LIME, ultimately aiming to reduce clinical workload.
At the 14th Supporting Health by Technology Conference in Enschede, WP2:
- ⚙Shared insights from our unsupervised change point detection approach, which analyzes wearable vital sign data to detect subtle changes in a patient’s recovery after surgery—potentially indicating early signs of complications.
- Collaborated with WP4 on a scoping review that examined how diverse populations are represented in wearable validation studies. Conducted by Rebecca Schipper, the review emphasized the need for inclusive approaches in health technology development.
These events highlighted the collaborative efforts driving RECENTRE research, with contributions from both academic and clinical partners.
