1 of 10

MobiGuide: Guiding Patients Anytime Everywhere��Societal Impact with and of Technology: Health

Mor Peleg, Dept. of Information Systems, University of Haifa, Israel

Yuval Shahar, Dept. of Software and Information Systems Engineering, Ben-Gurion University, Israel

Silvana Quaglini, Dept. of Electrical, Computer and Biomedical Engineering, University of Pavia, Italy

​

CACM Europe Region Special Section Workshop

August 26, 2021

2 of 10

The need

  • Aging population with chronic disease and compromised wellbeing
  • Shortage of clinicians and hospital beds

​

The solution

  • An AI system that monitors and manages patients remotely
    • Using sensors on patients, their smartphones, and decision-support servers
  • Its successful impact has been evaluated with patients of two types, in two countries, who have used the system for up to nine months

Congratulations, you are following recommendations about blood glucose monitoring very well.

3 of 10

MobiGuide: Generic Architecture for Guideline-based DSSs: Anytime, Anywhere

DSS

PHR

Body Area Network [BAN]

EMR1

EMR2

Personalized

Projection

&callback

Data Analysis

Knowledge base of

Computer-interpretable guidelines (CIG)

(Parallel GLs)

Secure

semantically-integrated

What do patients want?

4 of 10

  • ECG data are collected and abstracted into 1’ sessions.
  • Detected AF sessions stored in PHR.

MobiGuide for Atrial Fibrillation (AF)

  • DSS monitors for patterns of two or more sessions with AF

within a period of 10’

  • Alerts to patients
  • Notifications and further recommendations to doctors

​

5 of 10

  • The system monitors compliance and metabolic control, and accordingly applies evidence-based plans regarding diet, exercise, and measurement schedule
  • Personalized, context-sensitive reminders
  • Feedback and education
  • Advice on adding carbs, seeing doctor
  • Notification and further recommendations to doctors

​

Regular

Holiday

MobiGuide for Gestational Diabetes Mellitus (GDM)

6 of 10

  • Main benefit: Patients stayed at home but felt cared for! Movie
  • Compliance to measurements was high and higher than historical cohort
    • BP significantly lower in the MobiGuide cohort
    • Trend for fewer C-sections in the MobiGuide cohort
    • Most MobiGuide patients reported an improvement in their quality of life in the EuroQoL questionnaire, although there was some deterioration in the AFEQT questionnaire's scores (most of which are not related to the MobiGuide project)

​

Benefits to Patients

7 of 10

  • Clinicians used the system outside patient visits
  • AF doctors changed a long-time diagnosis for two patients!
  • GDM Doctors started insulin earlier for two patients!
  • Questionnaire: MDs Found the system useful
    • To identify priorities, increase productivity, QoD increased patient safety
    • Effective visits due to availability of data

Benefits to Care Providers (and to Patients)

7

8 of 10

  • Possible to operationalize successfully a large-scale complicated system such as MobiGuide
    • distributed decision support
    • innovative methodology for knowledge projection
    • based on computer-interpretable clinical guidelines
    • personalized to patient preferences and contexts
  • Generalizable to different clinical domains
  • Patients managed at home has economic benefits
    • Good compliance, good clinical outcomes
    • Saving visits and hospitalization
    • Patients that require faster intervention are detected more quickly, enhancing the level of care

​

​

​

​

Summary: it worked!

9 of 10

Discussion: Special Issues

  • Legal issues
    • Who is responsible if something goes wrong?
      • Currently in the EU: The hardware manufacturer!
        • Initiatives proposed to move the burden to the software developer
        • Proposals made in the past to assess the blame according to the chain of knowledge management and application, from the medical expert, through the knowledge engineer, software developers, sensor developers, and clinicians
    • Can patient data be transported across borders?
      • Current national laws often forbid that option, making universal accessibility of personalized care very tricky
  • Care-provider (re) education
    • Exploiting optimally evidence-based decision–support systems
  • Patient (re) education
    • Getting used to implications of empowerment and accountability

10 of 10

https://capable-project.eu addresses wellbeing of cancer patients

    • Mental wellbeing, nutrition, side effect
    • Multimorbidity support
    • Non-medication exercises (mindfulness, positive psychology, exercise)
    • Psychological theories for patient engagement and health habit formation
    • Machine learning on sensor data for:
      • Detecting preliminary conditions of developing adverse events
      • Finding the right moment to propose non-medication exercise

​

    • MobiGuide Web site: https://sites.google.com/hevra.haifa.ac.il/mpeleg/research/mobiguide

Discussion: The Next Frontiers