SMART IMPLANTS FOR ORTHOPAEDICS SURGERY (SImOS)Peter Ryser

Outline

Knee prosthesis from our partner Symbios Introduction Current needs (video) System overview Force sensors 3-D kinematics Miniaturization ASIC design and integration Technology demonstrator Knee simulator Achievements and next steps

Knee prosthesis from our partner Symbios3

On the basis of the work undertaken over 15 years at the EPFL, a team of Vincent Leclerc (Symbios) and Prof. Pierre-Franois Leyvraz (CHUV) developed a new prosthesis with an improved design compared to the state-ofthe art SImOS should show if further improvement gives advantages for the

Introduction4

The expected lifetime of a knee prosthesis is around 15 years, less for young people. A prosthesis failure requires a revision surgery, which is more complex and traumatic than the first replacement An instrumented prosthesis could help in order to:

improve the precision of the implant positioning quickly detect complications

Introduction: on the concept of the project5

Provide useful metrics for the clinicians and prosthesis designers

Correct unbalanced wearing Provide adequate rehabilitation Model force distribution Design new generation of prosthesis

Provide 3D joint angles by fusing with skin mounted sensors

Enhance accuracy Extract kinematics metrics relying on movement limitation

Provide micro-motion of the prosthesis relative to the bone

Estimate loosening using vibrating plate-form

Current needs6

Real need to get in vivo data & relevant objective outcome metrics

Inside the knee Force, pressure pain

breaks, increased t and Pwear, repartition implant loosening, osteolysis

Mobility function

implant parts knee joint potential conflicts with surrounding anatomical structures (ligaments muscles,)

Current needs7

Real need to get in vivo data & relevant objective outcome metrics

Inside the knee Force, pressure pain

breaks, increased t and Pwear, repartition implant loosening, osteolysis

Mobility function

implant parts knee joint potential conflicts with surrounding anatomical structures (ligaments muscles,)

Current needs8

Real need to get in vivo data & relevant objective outcome metrics

Inside the knee Force, pressure pain

breaks, increased t and Pwear, repartition implant loosening, osteolysis

Mobility function

implant parts knee joint potential conflicts with surrounding anatomical structures (ligaments muscles,)

Current needs9

Real need to get in vivo data & relevant objective outcome metrics

Inside the knee Force, pressure pain

breaks, increased t and Pwear, repartition implant loosening, osteolysis

Mobility function

implant parts knee joint potential conflicts with surrounding anatomical structures (ligaments muscles,)

SImOS objectives are:10

Patient specific monitoring

Improved rehabilitation process Improved follow-up process Improved complications detection and corrective measures to take

Almost perfect quality of care for patients

Origin of pain detection Adapted physiotherapy program for increased knee mobility Sports activities, enhancement programming

System overview11

SENSORS:

Force measurementsEnergy Commands

Kinematics measurementsMagnet

Data

d

AMR sensors

SensorsForce

Electronics

Communicatio n & Power Supply

Kinematics

System overview6

ELECTRONICS:

SensorsForce

Electronics

Communicatio n & Power Supply

Kinematics

System overview6

COMMUNICATION AND POWER SUPPLY:Reader Antenna Implanted Antenna

SensorsForce

Electronics

Communicatio n & Power Supply

Kinematics

System overview6

In the final system, all the components will be inside the polyethylene part of the prosthesis. All the energy necessary will be transmitted to the electronics by the external readerSensorsForce

Electronics

Communicatio n & Power Supply

Kinematics

Force Sensors15

Finite Element Analysis:Force

Region of interest for placement of sensing elements

Force Sensors16

Packaging Sensors and Calibration: Left Right

Left

Right

3D Kinematics17

Using 2 magnets + 3 magnetic sensors Sensor placement based on simulationBest placement of sensors

Magnet 1

A cut of PE sensors

Validated towards Motion-CaptureMagnet 2

Angles estimation results18

Different estimators were applied. Different estimated angles:15 Reference Internal-External Angle Estimated Internal-External Angle 10 5

IE angle

0

-5

-10

-15 040 35 30 25 20 15 10 5 0 -5 -10

25

50 Time (Sec)

75

1002

Estimated Flexion Angle Reference Flexion Angle

1 0 -1

Estimated Abduction Angle Reference Abduction Angle

Abduction angle4200 4300 4400 4500 4600 4700 4800 4900 5000 5100 5200 Sample

Flexion angle

-2 -3 -4 -5 -6 -7 -8 4200 4300 4400 4500 4600 4700 4800 4900 5000 5100 5200 Sample

Micro-motion analysis19

To see if there is a fracture or micro-motion in bone prosthesis contact, we have done some vibration analyses. module Sensor(placed on the prosthesis placed in knee) Fixation (3DoF) Vibrator Plastic bone

Measurements on a real leg

Analog Front-end ASIC 10 input channels (two force sensors and up to four 2D magneto-resistors) High gain (4096), low-noise amplification chain. 14-bit Sigma-delta modulator ADC with integrated decimation filter On-chip references 4-wire SPI interface Power consumption: 1.5 mW @ Vdd = 1.8 Volt

21

The Remote Powering System LevelPC interface Modulator

Downlink & Power L1 L2

Demodulator

External Reader AntennaRegulator Remote Electronics

*Implanted Antenna

MCU

Power Amplifier

Rectifier

Voltage Doubler Demodulator Battery

High V Regulator

Sensors

Modulator

Uplink

External

Implanted

Gate driver of the power amplifierKnee implant Loaded rectifier inputs and unregulated rectifier and voltage doubler outputs Reader coil

The external controller v1

Implanted coil

* Image source: http://healthtopics.hcf.com.au/TotalKneeReplacement.aspx0

The Remote Powering & Communication Blocks of the ASICCs

Vrec Cres Rectifier Regulator

1.8VSensor Interface

Storage CapacitorSPI

Meets the high power consumption requirements(~20mW) Application specific antenna and remote powering blocks for high remote powering efficiency design for flexible remote powering

Vrec Cd1 Voltage Doubler

Vd

High V Regulator

2.8VSensors

Cd2

Vd 1.8V

Voltage Regulation Two frequency band Blocks

Cdoubler

Demodulator

Modulator

High data rate (up to 200 kbits/sec) Air core coil as the remote powering antenna JTAG interface to configure the C

data en en data

Internal communication blocks

SPI Block

Regulation & Communication

SPI

Data Communication Small implanted area with the Blocks storage capacitor

Technology demonstrator23

Simultaneous acquisition of a force sensor and two channels of a 2D analog magneto-resistors (AMR)

24

Bionix kinematic knee joint simulatorBenefits

It provides the flexibility, precision, and control capabilities to replicate the complex dynamic forces and motions associated with knee movements. It simulates the relative motions of the femur, tibia, and patella, and measures the contact area and pressure distribution on the condyles. It can be used with natural tissues.

Integrated electronics25

Fully integrated structure based on 2 ASIC and a commercial low-power MCU Remote supply & readout of up to 10 analog sensors

LMIS4

RFiC

ESPL AB

LMAM

LPM2

Analog Front-end ASIC Up to 6 magnetoresistors or Hall sensors Up to 4 gauge sensors 14-bit ADC Comm & Power Low-noise 1000x ASIC gain 1.8 Volt from inductor coil 128 kbit/s uplink 1 kbit/s downlink High efficiency: 85%

Integrated implant26

The sensors & readout electronics is implemented on a flex PCB, molded into the polyethylene insert of the prosthesis.

1- PE insert 2- Flex PCB 3- Coil for comm & power 4- Force sensors 5- Magnetoresistors (2D) 6- Analog Front-end ASIC 7- 16-bit MCU, TI MSP430 8- Comm & Power ASIC

27

Achievements and next milestonesAchievements Development of an improved large-scale demonstrator Definition of micro-fabrication process and materials for force sensors array fabrication Tape-out of final sensor electronics ASIC Set-up of the commercial knee simulator Next milestones Development of the implanted transceiver System miniaturization

28

Thanks to all the team members

Peter Ryser Kamiar Aminian

EPFL/STI/IMT/LPM2 EPFL/STI/IBI/LMAM

Catherine Dehollain EPFL/STI/GR-SCI-STI/

Pierre Andr Farine EPFL/STI/IMT/ESPLABPhilippe Renaud Vincent Leclercq Scientists: PhD students: EPFL/STI/IMT/LMIS4 CHUV/DAL/OTR Symbios Orthopdie SA Arash Arami, Matteo Simoncini, Oguz Atasoy, Brigitte Jolles-Haeberli

Arnaud Bertsch, Eric Meurville, Steve Tanner Willyan Hasenkamp, Shafqat Ali

Thanks for your attention29