Development of a low-cost robotized 3D-prototype for automated optical microscopy diagnosis: An open-source system

Area of Science:

• Biomedical Engineering
• Microscopy Automation
• Medical Diagnostics

Background:

• Conventional optical microscopy is crucial for clinical diagnosis but faces declining expertise and potential for human error.
• Molecular techniques and rapid tests are impacting traditional microscopy use, necessitating innovative solutions.
• Microscopy automation is essential to overcome limitations of manual sample analysis, including fatigue and errors.

Purpose of the Study:

• To develop an affordable, automated system for visualizing microbiological/parasitological samples using conventional optical microscopes.
• To design a system specifically for implementation in resource-limited laboratory settings.
• To integrate automation with advanced diagnostic tools like convolutional neural networks.

Main Methods:

• A 3D-printable prototype was designed using biodegradable materials and open-source platforms (Arduino).
• The system automates microscope stage (X-Y) and auto-focus (Z) movements using servo motors.
• Smartphone integration facilitates image acquisition for diagnosis via convolutional neural networks.

Main Results:

• The automated system achieved an average auto-focus time of 27.00 ± 2.58 seconds per field of view.
• Auto-focus performance was validated with a mean average maximum Laplacian value of 11.83.
• The system successfully integrates automated microscopy with convolutional neural network-based image analysis.

Conclusions:

• The developed low-cost automated microscopy system offers a viable solution for resource-poor settings.
• Automation reduces diagnostic errors and fatigue, improving the reliability of microscopy-based diagnoses.
• This accessible technology can enhance diagnostic capabilities and provide valuable tools for laboratories globally.