Laser doppler blood flow imaging using a CMOS imaging sensor with on-chip signal processing

Area of Science:

• Biomedical Engineering
• Microsystems and Nanoengineering
• Optical Imaging

Background:

• Conventional laser Doppler blood flow (LDBF) imaging often requires bulky off-chip electronics.
• Space constraints in sensor design necessitate integrated signal processing for efficient LDBF applications.
• Optimizing signal-to-noise ratio and silicon area is crucial for compact LDBF sensor development.

Purpose of the Study:

• To design and test the first fully integrated 2D CMOS imaging sensor with on-chip signal processing for LDBF imaging.
• To develop a space-efficient pixel-level analog signal processing design tailored for LDBF signals.
• To overcome the data bottleneck between detector and processing electronics in LDBF systems.

Main Methods:

• Designed a 64x64 pixel 2D CMOS imaging sensor with integrated analog signal processing.
• Implemented pixel-level signal normalization and AC amplification with anti-aliasing filters.
• Utilized low-resource digital processing for on-chip computation and validated with simulations.

Main Results:

• The custom sensor achieved balanced optimization for signal-to-noise ratio and silicon area.
• On-chip processing enabled efficient analog-to-digital conversion and overcame data bottlenecks.
• The sensor demonstrated good agreement with simulations and successfully detected blood flow signals in vivo.

Conclusions:

• The integrated 2D CMOS sensor is capable of real-time LDBF imaging with enhanced efficiency.
• On-chip signal processing significantly improves the practicality of LDBF imaging systems.
• The sensor effectively detects blood flow variations in tissue, showing potential for clinical applications.