This document discusses a project called KHARE that uses mirror neuron research and human telemetry for rehabilitation. KHARE collects data from patients during physical therapy exercises using technologies like Kinect. The data is analyzed to help doctors increase therapy effectiveness and predict patient recovery. It aims to decrease healing times, increase functionality, and improve rehabilitation services. A proof of concept was tested in 2017 on a controlled group in partnership with universities and healthcare organizations. The goal is to extend KHARE's telemetry capabilities and empower all humans to improve skills.
Mirror Neurons and Human Telemetry for Rehabilitation
1. Mirror Neurons
and Human Telemetry
for Rehabilitation
ALESSANDRO LOMBARDI, DIGITAL ADVISOR | ITALY
KHARE
2. THETEAM
2
The team: Microsoft
Microsoft enables health sector’s digital transformation by
helping care organizations better engage patients, empower
care teams, optimize clinical and operational effectiveness, and
transform the care continuum.
Alessandro
Lombardi
Digital Advisor
Luca
di Biagio
Solution Architect
Paolo
Mele
Sr Consultant
Pedro
Fortes
Sr Consultant
Saverio
Guardato
Consultant
Roberto
Robustelli
Delivery Manager
11. • Average decrease of healing
times and increased movement
functionality; movements quality,
from the most approximate to the
most refined ones, for the normal
daily activities through the ones
needed for working.
• Increased number of
rehabilitations services that are
distributable by a single health
structure and subsequent costs
reduction for the patients and the
healthcare system.
• Patients and helpers satisfaction;
mainly because of the rehab
treatment continuum, given by the
relationship between the patient
and the health structure that
initially took the patient,
rehabilitation is also obtained with
the right psychological
contribution.
What are the advantages
of KHARE?
17. “All our dreams can come
true, if we have the courage to
pursue them.”
- Walt Disney
18.
19. COLLECT
data from the patients while they
are executing their physical
therapy exercises
TRANSFORM
data into a
standardized format
MANAGE
data as real-time streams and
store as relational or non-
relational datasets
in the cloud
ANALYZE
and predict with machine
learning and custom
queries and algorithms
VISUALIZE
through
dashboards, reports
to easily study data
DECIDE
The doctors
will be able
to increase the
therapies effectiveness
IOT HUB
Stream
analytics
Azure
StorageEvent
HUB
Api App
Machine
Learning
20. • Windows 8.1 app (Kinect SDK)
• Event Hub to collect data from Kinect in real time
• Table storage
• Integration with Microsoft Band
• Measure angles and distances between joints
• Create complex exercises
• Measure quality
App
Features
The proof of concept, scope
21. The proof of concept, data sheet
Input Values
Average message size every
second
Number of captured Joints
Average length of rehab
session (active time)
45.000 bytes 25 30 minutes
Average data size per rehab
session
Average data size per hour Therapy length
81 MB 162 MB 20 sessions per patient
Information per year
Number of expected patients Total sessions Total ingested data
50.000 1 M 81 TB
22. 23
Moving on with INAIL
Medical Experimentation set up (2016)
DCOD (IT) department focuses on building a
Minimum Viable Product to be used in
experimentation
Rehabilitation department finalizes the new
medical protocol for AOT with Parma
Video preparation
Medical Experimentation (2017)
CRM Volterra starts experimentation
according to the new protocol on a controlled
group with the support of Parma
Technological POC
Envisioning
ForumPA divulgation
Analysis and design
Implementation and release