Report an accessibility problem

Engineering | SENSIP

NSF IRES: Sensor Information Processing and Machine Learning for Wearable Devices

Collaborative between ASU SenSIP – DCU Insight Center
ASU: G. Raupp (PI), A. Spanias (Co-PI and POC), E. Forzani (Co-PI), and N. Kellam (Co-PI)
DCU: N. O’Connor (POC) S. Daniels, S. Little, N. Dunne and (Research Team at DCU)

New sensing and cloud computing capabilities in mobile platforms are creating game changing opportunities for mobile health research. A significant drawback is the lack of precision in emerging off-the-shelf mobile health products and apps. Advances in synergistic sensor and machine learning (ML) algorithm design are key to improving precision in mobile health and wearables. The aim of the IRES project is to embed students in cutting edge research that will provide integrated hardware and algorithm/software solutions for the next generation wearables and mobile health monitoring technologies.  This Interdisciplinary international program will address challenges at the overlap of sensor design, big data analytics, signal processing (SP), ML, and associated policies and standards. The IRES Track 1 program will provide opportunities for US students to collaborate internationally with researchers at the Dublin City University (DCU) Insight center. Students will be mentored by ASU and DCU faculty and will focus on solving application-driven problems involving integrated sensor devices and ML algorithms. The IRES will support 4 undergraduate and 2 graduate students for 6 weeks annually at DCU Dublin, Ireland. Intensive pre-training and post-engagement plans are provided to ensure successful outcomes. Provisions will be made (if necessary) to operate the IRES project under COVID-19 restrictions. The ASU Co-PI team brings expertise from Professors G. Raupp (sensors), E. Forzani (biomarkers), A. Spanias (ML, SP), and N. Kellam (Eng. Education). Core DCU collaborators are Professors S. Daniels (sensors), S. Little (big data, ML), N. Dune (nanosensors), and N. O’Connor (Vision, ML). The project will also engage Nobel Laureate Leland Hartwell as a thought leader for mobile health solutions. Additional faculty from both institutions will provide research mentorship and crosscutting training.

Intellectual Merit: Inexpensive sensors are required for mobile health monitoring and wearables. To improve precision, sensor design must be accompanied by corrective ML and SP algorithms. This requires integrative research that bridges the gap between sensor design and algorithm development. This IRES will focus on research and projects at the overlap of sensor device design and ML algorithm development. IRES projects will be planned in several areas including flexible sensors and ML, sensor info management and visualization, efficient deep learning, and big data analysis. Sample projects described include biomarker detection, big data processing, gait detection, and deep neural nets. Several additional projects will be defined later including COVID-19 hotspot detection, reduction of ML bias, and sweat sensors. Students will be embedded in research projects in an international setting which will positively impact their future careers. Embedding students in DCU labs will provide knowledge on EU and global research practices, standards, and policies. Students will build their own international network and will present their work beyond US borders. The program will include culture preparation at ASU and DCU. The prospect of research in ‘hot’ areas at a modern facility will enable us to recruit a strong and diverse student cohort. Workshops at ASU and DCU will train IRES participants to: a) present their innovations in global settings, b) communicate with international stakeholders, c) become culturally and geographically aware of global interactions. Industry engagement at ASU and DCU will provide feedback and annual external evaluation will assess progress and outcomes across all IRES activities.

Project sponsored by NSF OISE award  2107439.