Portable deep learning singlet microscope.

Area of Science:

• Optics and Photonics
• Computational Imaging
• Microscopy Technology

Background:

• Traditional microscopes often require multiple lenses, increasing complexity and cost.
• Developing portable and affordable microscopy solutions remains a significant challenge in scientific research.

Purpose of the Study:

• To propose a novel monochromatic/color singlet microscopy method.
• To leverage a single aspheric lens combined with deep learning for enhanced imaging.
• To demonstrate the feasibility of a lightweight and cost-effective microscope design.

Main Methods:

• Designed a singlet aspheric lens functioning as an approximate linear signal system.
• Integrated the lens with deep learning computational imaging technology.
• Developed a proof-of-concept singlet microscope weighing only 400g.

Main Results:

• The singlet aspheric lens exhibited nearly coincident modulation-transfer-function curves across the field-of-view.
• Experimental results with a USAF-1951 target and biological samples validated the system's performance.
• The singlet microscope demonstrated performance comparable to a commercial 4X/NA0.1 objective microscope.

Conclusions:

• The proposed singlet microscopy method offers a promising approach for developing cost-effective and powerful imaging systems.
• This technique has the potential to guide the design of future portable and high-performance microscopes.
• The combination of simplified optics and computational imaging opens new avenues in microscopy.