Automatic Faults Detection in Surgical Instruments

While manufacturing a surgical instrument, it is inspected repetitively by humans, in order to provide defect free surgical instruments. Large number of human workers are required for the inspection and separation of faulty instruments, which is expensive and time-consuming process. Being a

Project Title

Automatic Faults Detection in Surgical Instruments

Project Area of Specialization

Artificial Intelligence

Project Summary

While manufacturing a surgical instrument, it is inspected repetitively by humans, in order to provide defect free surgical instruments. Large number of human workers are required for the inspection and separation of faulty instruments, which is expensive and time-consuming process.

Being a part of surgical instruments manufacturing country, it is a responsibility to eliminate old techniques and start to acquire new technology to speed up the progress of the country in the field which it already has good reputation. So, our project’s focus is to automate the process of instruments inspection to eliminate human labor.

We proposed a computer-assisted machine vision approach for automated inspection of surgical instruments that uses image of an instrument to detect surface faults (cracks or pits) and separate faulty and fine instruments.  This project includes two sections:

1. Software Section
2. Hardware Section

Software section uses MATLAB for image acquisition followed by faults detection algorithm i.e. morphological operations, image segmentation, features extraction and classification based on K-Nearest Neighbor (K-NN) algorithm while the hardware section consists of conveyor belt, controller, light sources and 2 HD Webcams connected with a computer system. Moreover, a starwheel flipper for automatically flipping the instrument to access backside view of it.

Our project is an automated visual inspection system that detects surface faults based on geometric and shape features. This system can separate faulty and fine instruments in two different storage bins. This is a one-time investment approach, more accurate and efficient as compared to human inspection.

Project Objectives

The objective of this project is to present a model to make the inspection process fast and cheaper by effecient use of technology. By the hardware/software co-design, we’ll design such a system that meets the need of manufacturing industry. To accomplish this, we sub-divided it into smaller objectives which includes:

Automate the inspection process

The major aim is to develop reliable and automatic system as an alternate solution to human inspection as per requirement of manufacturing industry. So, we are developing a computer-assisted machine vision system for automated inspection of surgical instruments. This will reduce the burden of human labor and production cost.

Development of fast and accurate algorithm

There was need for the computationally inexpensive algorithm which also provides the better accuracy. In this regard, different algorithm-based approaches were developed and tested for execution time and accuracy. So, after features extraction from instrument's image, we will classify it using K-NN algorithm as it provides good accuracy at lower computation expense.

Project Implementation Method

The main challenges in Computer vision based smart inspection of Surgical Instruments are as:

• Low cost hardware development
• Development of fast and accurate algorithm

Low cost hardware development

The low-cost hardware development directed us to use a conveyor belt for carrying instruments. Once an instrument is placed on it, conveyor belt will move it to specific position for real time image acquisition. IR sensors have been used to detect instrument reached below cameras. The instrument is then flipped with the help of a “starwheel” object flipper to access instrument’s backside view for image acquisition.  Later, a separating plate is used to put the faulty and fine instruments into the respective storage bins. The following is how our hardware model works:

 

Development of fast and accurate algorithm

We are developing Faults detection algorithm using image processing and computer vision toolboxes in MATLAB. The overall algorithm for the operation of system is as followed,

After converting surgical instruments image into features, K-Nearest Neighbor (K-NN) algorithm is used to classify the instrument as faulty or fine classes. As per decision, our project’s hardware uses a separating plate that can put the faulty and fine instruments into the respective storage bins.

Benefits of the Project

The major benefits which the project will demonstrate are at the inspection stage while manufacturing surgical instruments are as mentioned below:

1. Automated inspection is faster as compared to tradition methods (human eye inspection). This will obviously cause increase in production rate.
2. It will reduce human labor.
3. Reduce instruments’ cost.
4. It will provide better quality of instruments because it eliminates human errors. 

Technical Details of Final Deliverable

Project delivered would be an integrated hardware/software co-design working model (i.e. computer-assisted machine vision system) which has ability to

1. Detect Cracks and pits on surface of instrument
2. Separate the faulty and fine instruments
3. Count number of faulty and fine instruments

The software designed contains a MATLAB code for surgical scissors and forceps, but the code designed is flexible and can be modified easily depending on the shape of surgical instrument.  Depending upon the geometric and shape complexities, we cannot design a single code for all different types of instruments. Some of the major hardware components are as follows:

Conveyor Belt

As per industrial requirement, we have prepared a slider bed belt conveyor for carrying surgical instruments. Following are specs of conveyor:

• Controlled by MATLAB through Arduino
• Frame is made of aluminium
• Driven by 12V DC flat gear motor
• Length, Width and Height are 5' , 1’3’’, 2’ respectively
• Accommodate a starwheel instrument flipper

16 Mega pixel HD webcam

The system includes 2 High definition 16 Mega pixel camera providing real-time high-resolution picture for surgical instrument @ 30fps. Images are directly saved to MATLAB temporary variables for further processing.

Some other components are

• Servomotors for controlling the position of separating plate
• Infrared sensor for detecting the availability of an instrument
• LED Lights

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Manufacturing

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Decent Work and Economic Growth, Industry, Innovation and Infrastructure

Required Resources

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