This document discusses using computational intelligence and biomedical engineering to help people with disabilities perform daily tasks through prosthetics. It describes a robotic arm modeled after a human arm with 4 degrees of freedom. Electrode signals from 3 arm muscles are processed through an artificial neural network and used to control the arm positions through inverse kinematics equations. The system was able to stimulate movement of the neuro-prosthetic arm but with less accuracy than a natural human arm.

2. Overview
 Biomedical Engineering and
Computational Intelligence
 Modeling Prosthesis
 Creating a Link
 Inverse Kinematics
 Conclusion
 Q & A???

3. Biomedical Engineering and
Computational Intelligence
 Helping people with disabilities to perform simple day to day
task.
 Decoding Arm Movements by Myoelectric Signals and
Artificial Neural Networks
 Computational Intelligence (CI) is the computing of
algorithms and learning how to use machines.
 Research in both CI and Biomedical Engineering help
provide new prosthesis that make everyday physical tasks
possible.

4. Modeling Prosthesis
 A robotic arm is developed, modeled as close to a human
arm as possible.
 The segment of the hand and arm for this particular test has
4 DOF which include flexion and extension of the forearm,
rotation of the arm, flexion and extension of the wrist and
opening and closing of the end-effector (hand).

5. Creating a Link
 The myoelectric signal processing was created through an
artificial neural network (ANN).
 The channels were connected by placing electrodes on three
different muscles ( Flexor Carpi Ulnaris, Extensor Carpi
Raidialis Longus, and the Biceps).
 The signals were transferred to a NI USB-6009 acquisition
board and sent through a Neural Network.

6. Creating a Link

7. Inverse Kinematics
 X = L1 cos(theta1) + L2 cos(theta1+theta2)
 Y = L1 sin(theta1) + L2 sin(theta1+theta2)
 Theta2 = arccos[(X^2 + Y^2 – L1^2 – L2^2)/ 2* L1^2 *L2^2]
 Theta1 = arctan(Y/X) – arctan[(L2 sin(Theta2))/ L1 * L2cos(Theta2)]

8. Conclusion
 The proposed system used some
myoelectric signal channels combined
with digital processing to stimulate the
neuro-prosthetic arm.
 The accuracy was below average when
compared to an actual human arm.

9. Questions
 ????????