Holographic 3D Spatial Localization for Bioassay Based on Depth-Label Encoding and End-to-End Attention Network

Area of Science:

• Optics and Photonics
• Biomedical Engineering
• Machine Learning

Background:

• Lens-free holographic imaging struggles with limited depth of field and volumetric detection.
• Current methods for 3D localization are often slow and costly.

Purpose of the Study:

• To develop a holographic bioassay platform with enhanced depth of field and volumetric detection capacity.
• To enable accurate, real-time 3D localization for improved biosensing.

Main Methods:

• Integration of depth-label encoding with an attention-based neural network for 3D spatial localization.
• Development of a dual-task framework combining bounding box detection and depth classification.
• Application of the platform for chloramphenicol detection.

Main Results:

• Achieved extended depth of field and increased microsphere density for volumetric detection.
• Enabled chloramphenicol detection over a wide dynamic range (5 pg/mL to 100 ng/mL).
• Demonstrated a 10-fold sensitivity enhancement compared to 2D hologram-based assays.

Conclusions:

• The 3D_EEA biosensor offers a scalable solution for trace-level antibiotic detection.
• The platform shows excellent performance and practical potential for food safety and environmental monitoring.
• Overcomes limitations of traditional holographic reconstruction, enabling real-time 3D localization.