Boston Scientific · Automation
Automated visual inspection
Automating guidewire inspection with a motorized feed mechanism, dual microscopes, and machine-learning defect detection.
- My work
- Fixture design, feed mechanism, and imaging integration
- Result
- Inspection time reduced from 60 minutes to 2 minutes
- Status
- Built and tested at Boston Scientific
Automating a manual inspection method
The existing test method required manually inspecting a long medical guidewire under a small microscope after procedures. I saw an opportunity to automate that workflow, proposed it, and designed and built a system that combined mechanical feeding, microscope imaging, and machine learning defect detection.
The design decision
My initial concept rotated the cameras around the device. Iteration led to a simpler assembly: two stationary microscopes, compliant feed wheels, and stepper-driven device motion.
| Design aspect | Initial concept | Built system |
|---|---|---|
| Imaging arrangement | Cameras rotate around the guidewire | Two stationary microscopes |
| Device handling | Relative motion provided by the cameras | Stepper-driven compliant feed wheels advance the guidewire |
Designing the inspection hardware
I designed the fixture in SolidWorks around two digital microscopes. Rubber compliant feed wheels advanced the guidewire through the imaging area; the compliant contact was chosen to feed the wire without damaging it. Arduino coordinated the hardware, while Python connected motion control and image acquisition.
Training the defect-detection model
I collected training images and trained an AWS Lookout for Vision machine learning model to identify defects. Consistent lighting and guidewire positioning helped produce repeatable images for training and inspection.
Reported test results
- 96.7%
- Inspection time reduction · 60 minutes to 2 minutes
- 96.4%
- Reported defect detection accuracy
- 508
- Images in the training dataset
The system was tested with devices containing known defects. The technical report estimates annual savings of $5,000 and 60 engineering hours.
Engineering takeaways
Simplify the mechanism before the controls
I moved from a concept with rotating cameras to two stationary microscopes and a moving guidewire. That decision reduced the moving hardware and shaped the motion-control and imaging sequence.
Design the fixture as part of the measurement system
Compliant feed wheels, consistent lighting, and repeatable guidewire positioning supported image acquisition. Mechanical variation can appear in the images, so fixture design matters to detection performance.
Measure the complete workflow
Inspection time fell from 60 minutes to 2 minutes in the reported comparison. The result came from mechanical feeding, image capture, and analysis working together. The reported 96.4% detection accuracy was measured using devices with known defects; broader performance needs separate validation.