Xi Long, PhD

Research

Reliable learning for intelligent health monitoring

Reliable learning aims for delivering reliable systemd with machine-assisted signal processing & artificial intelligence models to realize intelligent health monitoring. Such systems should include three important capabilities - unmet needs: good performance both in accuracy and cost-effectiveness, good generalisability to handle data heterogeneity, and good explainability being able to support clinicians with explainable models and decision making. Key application areas are the following.

Infant sleep monitoring

The research aims at assessing (preterm and term) infant sleep using vital signs that are possible to be acquired through unobtrusive or contactless sensors such as camera, smart mat, and radar.

Neonatal analysis and alarm management

The research aims at analyzing neonatal health conditions/diseases, measuring vital signs, and managing in-hospital alarms in the neonatal intensive care unit.

Unobtrusive sleep assessment

The research focuses on assessment of adult sleep and sleep problems (e.g. obstructive sleep apnea) using unobtrusive signals such as heart rate (variability), respiration, body movement, and audio.

Brain activity for neurological diseases

The research focuses on detecting neurological disorders (epilepsy and Parkinson's disease) through analyzing brain activity (EEG), motor activity and/or other biosignals.

Wearable sensing & physiological analysis

The research focuses on investigating wearable sensing and physiological monitoring for blood pressure, physical activity, cardiac diseases, EMG analysis, etc.

Pregnancy health management

The research focuses on early risk stratification or screening and symptom monitoring of pregnancy complications such as hypertensive disorders, gestational diabetes, and adverse fetal outcomes.

Heart health monitoring

The research leverages new sensors and advanced signal processing / AI technologies for monitoring cardiovascular health (such as cardiac arrhythima, atrial fibrillation, heart failure).

Acute and critical care

The research aims at studying AI models and algorithms and telehealth for early warning / prediction of acute and critical conditions (such as sepsis, acute respiratory distress syndrome) in hospitals.

Dr. Long's research and innovation work has contributed to multiple commercial products or the related research such as Philips WeST wearable sensing technologies, Philips Health Watch, Philips Health Band, Philips Avent Connected baby monitor, uGrow smart baby monitor and uGrow development tracker, Philips Pregnancy+ app, Philips CareEvent management system for neonatal ICU, Philips DirectLife activity tracker, Neolook Screen2Screeen solutions (Philips spin-off), NightWatch for epilepsy seizure detection, Tencent QQ Music app, Tencent Pengyou social network, etc.

Dr. Long collaborates with many leading institutes and hospitals such as: Utrecht University (NL), TU Delft (NL), Wageningen University & Research (NL), Radboud University (NL), Donders Institute for Brain, Cognition and Behaviour (NL), Máxima MC (NL), Kempenhaege (NL), Catharina Hospital (NL), UMC Utrecht (NL), UMC Groningen (NL), SEIN (NL), Amsterdam UMC (NL), KU Leuven (BE), RWTH Aachen (DE), Imperial College London (UK), The University of Sheffield (UK), University at Buffalo (US), Fudan University (CN), Women's Hospital Zhejiang University (CN), Peking University (CN), ShanghaiTech University (CN), CAS IME (CN), CAS-SIBET, South China Normal University (CN), South China Normal University (CN), and Southeast University (CN).

Main Projects

MIRACLE (2024-2029, PPS funding: e/MTIC, paartially HTSM-TKI)
Medical Innovation and Research Advancing Clinical Learning and Excellence
MEDEIA (2022-2027, PPS funding: e/MTIC, partially HTSM-TKI)
MEDical Engineering, Innovations and Applications
PISANO (2021-2026, PPS funding: e/MTIC, partially HTSM-TKI)
Perioperative Innovations, Sleep Apnea and Newborn Opportunieis
MEDUSA (2020-2025, PPS funding: e/MTIC, partially HTSM-TKI)
MEdical Data Utilization Solutions Accelerator
STRAP (2020-2025, funding: NWO Commit2Data)
Self TRAcking for Prevention and diagnosis of heart disease
TU Diversity (2020-2022, funding: 4TU.NIRICT)
Gender and nationality composition research in student teams of Dutch technical universities
PICASSO (2019-2024, PPS funding: e/MTIC, partially HTSM-TKI)
PerInatal, CArdiovascular and Sleep medtech SOlutions
NICUSleep (2019-2022, funding: Philips-Fudan)
Acute care solution NICU sleep research | China sleep & respiratory health
PregRisk (2018-2022, funding: Philips)
Early risk stratification and prediction of pregnancy complications
BabySleep (2016-2022, funding: Philips)
Baby health and sleep quality monitoring with contactless sensing & AI techniques
ALARM (2017-2022, funding: NWO-HTSM)
Alarm-Limiting AlgoRithm-based Monitoring in the neonatal intensive care unit
BrainWave (2016-2021, NWO-OTP)
Non-convulsive status epilepticus seizure analysis and detection
OSA+ (2016-2021, funding: STW/IWT bilaterale OTP)
Multimodal signal analysis for unobtrusive characterisation of obstructive sleep apnea
ADAM (2018-2020, funding: UMCU, Philips)
Applied Data Analytics in Medicine - big data for small babies - clinical decision support
LOTUS (2017-2020, funding: Philips)
Next-generation neonatal intensive care unit
PregHealth (2017-2019, funding: EIT Digital)
Pregnancy health: monitoring of pregnancy risk and wellbeing
BNFD (2015-2018, funding: EIT Health)
Better nights, fresh days for parents
IMPULS (2014-2018, funding: Philips)
Premature neonatal sleep monitoring
WeST (2011-2015, funding: Philips)
Wearable sensing technologies for unobtrusive sleep monitoring
DirectLife (2008-2010, funding: Philips)
Physical activity monitoring through machine learning algorithms

Google Sites

Report abuse