Dear Colleagues,

Preventive and occupational medicine requires the acquisition of physiological, psychological, physical, and prior medical data to generate a patient-specific model that can be used to detect anomalies. This Special Issue is looking for papers that deal with a decentralized multi-sensor-fusion approach based on flexible mobile data-collection platforms that can be used to create a preventive health-management system. We are looking for novel algorithms and hardware systems that can connect, acquire, and synchronize various sensors attached to a person’s body and then securely transmit the fused data to a cloud server. The data used by such systems may include different mobile data sources from a remote data collection system, including directly coupled wireless sensor devices, indirectly connected devices from vendor-specific cloud solutions, and prior medical knowledge. Once received at the could server, the fused data time series is then analyzed using multivariate machine learning algorithms to detect abnormal conditions that can then be transformed into a human-understandable form that users or clinicians can quickly understand. Pertinent topics include the following:

• Temporal synchronization techniques of various sensors using local body networks
• Low-level sensor fusion algorithms to compensate measurement artifacts created by patient’s activities such as motion wonders in ECG measurements and temporal coincidence algorithms to reduce false alarms in fall detection
• Multivariate temporal algorithms to detect and classify abnormal conditions
• Novel techniques to train patient-specific models from sensor data
• Novel algorithms to transform the classification of sensor data into human-understandable explanations
• Novel techniques for secure transmission and processing of sensor data, such as:
  • Use of homomorphic encryption techniques to process data
  • Scalable private networks capable of dealing with a large population

Prof. Dr. Pierre Boulanger
Guest Editor