Integration of Force Feedback into Minimally Invasive Robotic Surgery

1
Integration of Force Feedback into Minimally
Invasive Robotic Surgery

• Team MEM-16
• Philip Hufnal
• John Monaghan
• William OConnor
• Christopher Sagedy

2
Submitted To Submitted To
Dr. B.C. Chang Dr. Wei Sun
Dr. Moshe Kam Dr. Leonid Hrebien
Pramod Abichandani Chirag Jagadish Pramod Abichandani Chirag Jagadish
3
Problem Background

• Surgical Techniques
• Open
• Minimally Invasive (MIS)
• Laparoscopic
• Minimally Invasive Robotic Surgery (MIRS)
• Traditional Laparoscopic
• Instruments manipulated by hand through small
  incision
• Endoscope used to view operating space
• Laparoscopic limitations
• chopstick effect
• 2D representation on monitor

4
Minimally Invasive Robotic Surgery (MIRS)

• Robotic systems were developed to overcome the
  awkward hand-eye coordination involved in
  laparoscopic procedures 1
• Intuitive Surgicals daVinci system is most
  popular for minimally invasive robotic surgery
  (MIRS)
• Approved by the FDA in July of 2000 2
• 860 units sold and more than 70,000 surgeries
  performed each year 3, 4

5
The da Vinci Surgical System
6
5
6
Strengths and Drawbacks of MIRS

• MIRS strengths
• Geometric accuracy
• Free from tremors and fatigue
• Sterile
• Resistant to infection
• MIRS drawbacks
• Current systems lack realistic, continuous force
  feedback 7
• Training is expensive and time consuming
• Inexperienced surgeons take longer using MIRS
  than conventional methods

7
MIRS Statistics

• Taken from Clinical Efficacy - Comparison of
  Open Prostatectomy, Laparoscopic and da Vinci
  Prostatectomy 8

Open Laparoscopic da Vinci MIRS
Patients 100 50 100
Operative Time (Min.) 164 248 140
Blood Loss (mL) 900 380 lt100
Cancer Remaining 24 24 5
Complications 15 10 5
Catheter, Days 15 8 7
Hospitalization (Days) 3.5 1.3 1.2
8
Problem Statement

• Current minimally invasive robotic surgical
  systems lack realistic force feedback
• Tactile sense is one of the most important tools
  for surgeons
• Surgeons rely on tactile sense in order to
  characterize tissue and make intra-operative
  decisions 9.
• Surgeons must rely on visual cues only, which
  requires significant experience 1.
• Lack of realistic force feedback can lead to
  tissue damage or other mistakes
• Without force feedback in blunt dissection, the
  number of errors resulting in tissue damage
  increased by over a factor of 3 10
• Increased operating time and cost

9
Force Feedback

• Preliminary research shows that force feedback
  improves on the gains made by MIRS over
  conventional techniques
• Possible to tie tighter, more consistent sutures
  7
• Improves arterial dissection by reducing
  unintentional transections 12
• Reduces learning curve for surgeons with little
  or no experience using MIRS 13

10
Design Objective

• Prototypic slave-master surgical robotic system
• Sense force at the slave (surgical manipulator)
• Generate force at the master (user controls)
• Microprocessor communication between master
  controller and slave gripper allows us to
• Control gripper angle
• Reproduce gripper forces at the surgeon user
  control
• Translate the EndoWrist in two axes

11
Method of Solution
12
System Components

• Two electromechanical systems in a slave-master
  configuration
• Master includes custom user controls, force
  feedback actuator, and microprocessor
• Slave includes EndoWrist, x-y stage, and
  microprocessor
• Host computer to display important information
  and set system parameters

13
The EndoWrist

• Precision surgical manipulator developed by
  Intuitive Surgical for the da Vinci Surgical
  System
• Offers 7 degrees-of-freedom
• Complex movements made possible using four
  control knobs

14
EndoWristControl Knobs
15
Slave Interface to EndoWrist

• Three components
• Base includes four holes into which DC motors can
  be mounted
• Cylindrical adapter can be used to attach a motor
  shaft to an EndoWrist control knob or to hold
  control knob in place
• Clamp used to attach the base to the EndoWrist
• Modular Additional motors can be added if needed

16
Base of Slave Interface

• Top View

• Bottom View

17
Adapter for Slave Interface

• Top View

• Bottom View

18
Force Sensing

• Motors will be controlled using PWM controlled
  H-bridge circuits
• To sense force, a current sensor will be placed
  in series with the EndoWrists drive motor
• Does not require any physical modifications to
  the EndoWrist

19
x-y Stage

• The EndoWrist and the mechanical interface
  assembly will be mounted on an x-y stage
• The stage will be positioned vertically to allow
  to allow the user to grip, lift, and pull a
  sample placed beneath the EndoWrist

20
Slave Processor

• The slave processor will
• Execute control algorithms for the EndoWrist and
  the x-y stage
• Collect and transmit force measurements to the
  master processor
• A second, less powerful processor (the stage
  controller) will drive the motors for the x-y
  stage

21
Master Grip Controller

• Two gripper levers onto which plastic finger
  splints will be placed
• Levers will rotate the shaft of a DC motor
• Optical encoder will measure shaft position
• Drive motor to provide force feedback

22
Translation Controller

• Used to control the movement of the x-y stage
• Will be integrated with the grip controller
• Possible options
• Joystick controller
• Translation sensor (e.g. optical mouse sensor)

23
Master Processor

• The master processor will
• Receive force measurements from the slave
  processor
• Transmit position data to the slave processor
• Execute a position-force control algorithm to
  create a realistic sense of touch
• Transmit data to the host computer via USB

24
Host Computer

• Connected to the master processor via USB
• A graphical user interface (GUI) will display
  information about system operation to the user
• May include real time graphs of force and
  position
• GUI will also be used to set system parameters
  (e.g. scaling ratios)

25
(No Transcript)
26
Deliverables

• The design should meet the following criteria
• The user will command grip and 2D translational
  motion of the EndoWrist
• The user will be able to sense an object in the
  grasp of the EndoWrist

27
Alternative Solutions
28
Alternative Solutions

• Two key components to any force feedback solution
  
• Instrumentation Design - Placement of sensors
• Sensing Method - Measure force
• The key components are interdependent certain
  methods of sensing force require specific
  placements for sensors
• Based on the selection of the key components,
  its possible to choose a control algorithm
• Two common force feedback control algorithms are
• Force-Position
• Position-Error

29
Alternative Solutions (Contd)

• For force feedback, the force sensing elements
  placed at the slave, on the laparoscopic
  instrument
• Laparoscopic instrument is inserted into the body
  through an insertion which is 3-12mm in diameter
• Particular sensors may have limitations on
  feasible locations

30
Placement of Sensors

1. At the joint actuation unit
2. Shaft portion outside abdominal wall
3. Shaft portion inside abdominal wall
4. End effector's articulated joints

14
31
Methods of Sensing Force

1. Displacement
2. Current
3. Pressure
4. Resistive
5. Capacitive
6. Piezoelectric
7. Vibration
8. Optical

32
Sensor and Placement Selections

• Current-based sensing
• Exploits the relationship between shaft torque
  and armature current, which is linear in the
  region of interest
• Placed at the joint actuation unit
• Four motor driver knobs to control gripper
  orientation and actuation

33
Advantages of Selections

• Minimizes redesign of da Vinci
• Allows the EndoWrist to be used without redesign
• Known to be comparatively simple approach
• Calibration and experimentation can be used in
  lieu of system modeling to determine erroneous
  forces
• No need for sterilizability and biocompatibility
• Lower recurring cost to end user than internally
  placed sensor - no need to be cleaned or replaced
  after each operation

34
Disadvantages of Selections

• Measurements may be inaccurate due to
  approximations which are based on torque-current
  curve
• Force signal may have magnitude and phase
  distortions
• Mechanical linkages introduce error due to
• Friction
• Backlash
• Inertia
• Mechanical moments

35
Final Analysis of Key Component Selections

• Inexpensive
• Low complexity
• Adaptable to existing robotic surgery systems
• Alternative methods may require redesign of the
  laparoscopic instrument
• Feasible using available senior design resources

36
Design Constraints

• Cost and time
• Limited funding
• Must be completed within next two terms
• Complexity limited by funding and duration
• FDA regulations specify that the slave surgical
  manipulator must be biocompatible and
  sterilizable 15
• Constraint is met by placing sensors outside the
  body

37
Design Constraints (Contd)

• Geometry
• Ability of instrument to be inserted through
  trocar
• Induced Master Motion
• Stability is compromised by allowing force
  feedback signal to influence position
• Gains are limited to values that ensure system
  stability

Trocar
Skin
1
38
Project Management
39
Gantt Chart
40
Task Tree
41
Industry Budget
a. Includes cost of H-bridge, current sensor, and
quadrature decoder b. Includes cost of Pro-E and
SolidWorks c. Cost of Labor4 junior engineers,
55K salary, 15 hrs/week, 36 weeks d. Overhead
50 of total Cost of Labor
42
Economic Analysis

• Dr. B.C. Chang provided EndoWrist and 2 DC motors
  with encoders for free
• Manufacturing provided at Olympic Tools expense
• Freescale Semiconductor provided 2
  microcontroller development boards for free by
  sponsorship agreement
• Engineering software available
• Not considering labor and overhead costs

43
Out-of-Pocket Costs
Description Cost Quantity Total Cost
Translation (x-y) Stage 1,274.00 1 1,274.00
Digital Motor Controller Board 400.00 2 800.00
Total Total Total 2,074.00
44
Societal and Environmental Impact

• Implementation of force feedback will allow more
  delicate surgeries
• For example, Mitral Valve Repair (MVR) is
  significantly less dangerous when MIRS is used
  16

2001 STS Natl Database Sternotomy MVR da Vinci Trial
Patients 893 112
Mortality 2.2 0
Major Complications 12.1 9.8
Neurological Complications 2.4 0
Hospitalization (Days) 8.5 4.7
45
Societal and Environmental Impact (Contd)

• Improvements to MIRS should further reduce
• Complications
• Supplemental surgeries to correct complications
• Hospital stay times
• Operation time
• Cost of surgery may be decreased
• Implementation of force feedback in surgical
  robotics should have a minimal impact on the
  environment
• Reduction in number of operations and operation
  time will minimize the amount of electricity used
  in the operating room

46
Conclusion

• Statistics indicate that use of MIRS will
  continue to grow
• Prototype of slave-master surgical robotic system
  with grip force feedback
• Current-based force sensing
• Custom designed user controls, EndoWrist
  interface
• Surgeon controls grip and 2D translation of
  EndoWrist
• Beneficial impact on the evolution of medical
  robotics

47
Questions?
48
References

• 1 K. Seibold, U.and Bernhard and G. Hirzinger.
  Prototypic force feedback instrument for
  minimally invasive robotic surgery. In V.
  Bozovic, editor, Medical Robotics, page 526.
  I-Tech Education and Publishing, Vienna, Austria,
  2008.
• 2 M. Meadows. Robots lend a helping hand to
  surgeons. FDA Consum, 36(3)105, 2002.
• 3 www.intuitivesurgical.com/corporate/newsroom/m
  ediakit/da_Vinci_Surgical_System_FAQ.pdf
• 4 www.pacrimrobotics.com/forms/orangecounty_reg.
  pdf
• 5 http//www.childrenshospital.org/clinicalservi
  ces/Site1860/Images/robotics(E).jpg
• 6 http//www.healthaffairs.uci.edu/urology/prost
  ate/daVinci.html
• 7 http//www.pubmedcentral.nih.gov/articlerender
  .fcgi?artid1283126
• 8 Menon M. Robotic radical retropubic
  prostatectomy. BJU Int. 2003 Feb91(3)175-180.
• 9 G. Tholey. A teleoperativehaptic feedback
  framework for computer-aided minimally invasive
  surgery. PhD thesis, Drexel University, 2007.

49
References (Contd)

• 10 C.R. Wagner, N. Stylopoulos, and R.D. Howe.
  The role of force feedback in surgery analysis
  of blunt dissection. In Haptic Interfaces for
  Virtual Environment and Teleoperator Systems,
  2002. HAPTICS 2002. Proceedings. 10th Symposium
  on, pages 68-74, 2002.
• 11 http//www.ingentaconnect.com/content/klu/115
  48/2008/00000003/F0020003/00000228
• 12 Deml, B. Minimally Invasive Surgery
  Empirical Comparison of Manual and Robot Assisted
  Force Feedback Surgery Proceedings of
  EuroHaptics 2004, Munich Germany, June 5-7, 2004.
• 13 Reiley, C. Effects of visual force feedback
  on robot-assisted surgical task performance The
  Journal of Thoracic and Cardiovascular Surgery,
  January 2008.
• 14 P. Puangmali, K. Althoefer, L.D.
  Seneviratne, D. Murphy, and P. Dasgupta.
  State-of-the-Art in Force and Tactile Sensing for
  Minimally Invasive Surgery. Sensors Journal,
  IEEE, 8(4)371381, 2008.
• 15 Allison M. Okamura. Haptic feedback in
  robot-assisted minimally invasive surgery.
  Submitted to Current Opinion in Urology, August
  2008.
• 16 http//www.sts.org/2003webcast/shows/tatooles
  /tatooles.html