Acquisition and Classification of Biopotentials for HMI

An increasing number of upper limb amputations occur each year due to a variety of reasons including trauma, hazardous workplace incidents and congenital defects etc. This results in people being unable to perform activities of daily living (ADLs) with ease, to feel like an outcast among society and

Project Title

Acquisition and Classification of Biopotentials for HMI

Project Area of Specialization

Biomedical Engineering

Project Summary

An increasing number of upper limb amputations occur each year due to a variety of reasons including trauma, hazardous workplace incidents and congenital defects etc. This results in people being unable to perform activities of daily living (ADLs) with ease, to feel like an outcast among society and limitations in their job opportunities as well leaving them both financially and mentally distressed and unable to survive independently. This is where rehabilitative devices such as prosthetic arms come in that are aimed towards restoring lost functionality. Looking at the Pakistani market, where most amputees belong to the lower-class laborer or middle-class category, we can see that there exists a wide gap between the product available and the demands of the consumer. For example:

• Devices available limited in functionality

• Have primitive and purely cosmetic designs with limited functionality

• Maybe highly functional but extremely expensive.

Human Machine Interfacing (HMI) is the connection of human cognition with different machines in order to achieve a variety of tasks. An increasing use for HMI is the operation of prosthetic limbs using signals acquired from the human body (biopotentials). Such an HMI system allows people who have undergone amputations to return to normalcy as myoelectric and artificial intelligence (AI) based prosthetic limbs become more common.  

Usually advanced and highly functional prosthetics made either locally or internationally utilize crude biopotentials with limited signal processing. Since biopotentials extracted from amputees tend to differ profoundly from those of healthy individuals, using predefined algorithms can result in poor control of these prosthetic limbs. 

Our aim is to design a HMI system that utilizes AI to perform classification of biopotentials extracted from the amputee, in order to achieve maximum accuracy as well as minimize erroneous control outputs, that can then be used in conjunction with any prosthetic limb hardware. We hope to do this by training an AI system extensively on personalized data.

Project Objectives

• Cost-effective design

• Extract viable biopotentials from an amputee

• High accuracy classification of biopotentials using AI to generate a system of control signals

• Accurate control of effector machines

Project Implementation Method

The project will be made using components easily available in the market so as to make the design feasible, easy to manufacture and reduce service costs overall. The HMI system is to be designed for an above-the-elbow amputee which will constitute of three elements or parts: the electrodes, the processor unit, and the output signals.

Gold plated reusable electrodes will be used as the permanent signal acquisition electrodes. These offer little resistance and are best suited for capturing complex biopotentials from above human skin. 

The processor unit will serve as the center for computation for the HMI system. It will consist of commercially available microcontrollers and peripheral attachments that will process incoming biopotentials, filter any interferences, classify different signals using AI, generate different control signals in accordance with the inputs, and output these control signals to effectors elsewhere in the device. 

The output signals will be carried through high fidelity wires to the necessary effectors as required for the prosthetic limb. These include motors, servos, switches, etc.  

The system will receive power from the prosthetic arm it is fitted into.

Benefits of the Project

• A cost-effective & affordable product

• Generalized algorithm 

• Wide range of customization 

• Finer and more accurate control of machines

• Low computing power requirements 

• Easy to implement on widely available microcontrollers  

• Highly functional to allow the user to perform activities of daily living with ease

• Community integration

• Vocational Rehabilitation

Technical Details of Final Deliverable

• Capture of biopotentials from amputation site

• Filtering and denoising of acquired potentials  

• Use of AI to classify potentials 

• Generation of viable control signals from classified potentials 

• Interfacing with effectors 

• Operating effectors using control signals 

Final Deliverable of the Project

Software System

Core Industry

Health

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People, Decent Work and Economic Growth, Reduced Inequality

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com