Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Imaging Biological Samples with Optical Microscopy

Imaging Biological Samples with Optical Microscopy

Optical microscopy uses optic principles to provide detailed images of samples. Antonie van Leeuwenhoek designed the first compound optical microscope in the 17th century to visualize blood cells, bacteria, and yeast cells. In 1830, Joseph Jackson Lister created an essentially modern light microscope. The 20th century saw the development of microscopes with enhanced magnification and resolution.
In optical microscopy, the specimen to be viewed is placed on a glass slide and clipped on the stage...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Optical metasurfaces for general vision processing on the edge.

Nature·2026

Same author

An all-optical signal processor enabling terabit-per-second real-time equalization.

Science (New York, N.Y.)·2026

Same author

Photonic Mixture-of-Experts for scalable multi-task on-chip optical neural networks.

Nature communications·2026

Same author

Programmable Three-dimensional Photonic Neural Network Chip.

Nature communications·2026

Same author

All-Optical Diffractive Operators for Rapid, Computer-Free Morphological Transformations.

Nanophotonics (Berlin, Germany)·2026

Same author

Author Correction: Programmable 200 GOPS Hopfield-inspired photonic Ising machine.

Nature·2026

Same journal

Kat5 deficiency in alveolar type II cells licenses STAT6-driven glycolytic reprogramming and pulmonary fibrosis.

Nature communications·2026

Same journal

Continuous nonthermal slab gap formed by progressive tearing beneath Northeast Asia.

Nature communications·2026

Same journal

Zeolitic isolated protonic acid sites-mediated NH<sub>3</sub> storage for robust NO<sub>x</sub> removal.

Nature communications·2026

Same journal

Coaxially nested component with asymmetric fiber resonant cavity and separation membrane for gaseous and dissolved gases detection.

Nature communications·2026

Same journal

Near-unity charge readout signal in a nonlinear resonator without matching the sensor dissipation.

Nature communications·2026

Same journal

Prokaryotic Schlafen proteins cleave tRNAs during type III CRISPR immunity.

Nature communications·2026

See all related articles

Search research articles

Related Experiment Video

Updated: May 5, 2026

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Published on: March 2, 2011

Multimodal deep learning using on-chip diffractive optics with in situ training capability.

Junwei Cheng¹, Chaoran Huang², Jialong Zhang¹

¹Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan, 430074, China.

Nature Communications

|July 23, 2024

Summary

This summary is machine-generated.

This study introduces a trainable diffractive optical neural network (TDONN) chip for multimodal deep learning. The novel photonic chip efficiently processes diverse data types, achieving high accuracy in vision, audio, and touch classification.

More Related Videos

Lensless Fluorescent Microscopy on a Chip

Lensless Fluorescent Microscopy on a Chip

Published on: August 17, 2011

Lensfree On-chip Tomographic Microscopy Employing Multi-angle Illumination and Pixel Super-resolution

Lensfree On-chip Tomographic Microscopy Employing Multi-angle Illumination and Pixel Super-resolution

Published on: August 16, 2012

Related Experiment Videos

Last Updated: May 5, 2026

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Patterned Photostimulation with Digital Micromirror Devices to Investigate Dendritic Integration Across Branch Points

Published on: March 2, 2011

Lensless Fluorescent Microscopy on a Chip

Lensless Fluorescent Microscopy on a Chip

Published on: August 17, 2011

Lensfree On-chip Tomographic Microscopy Employing Multi-angle Illumination and Pixel Super-resolution

Lensfree On-chip Tomographic Microscopy Employing Multi-angle Illumination and Pixel Super-resolution

Published on: August 16, 2012

Area of Science:

Photonics
Artificial Intelligence
Neuromorphic Computing

Background:

Multimodal deep learning is crucial for artificial intelligence generated content (AIGC), but existing photonic processors struggle with diverse data types due to training limitations.
Current optical neuromorphic processors are often limited to single data modalities (e.g., vision or audio).

Purpose of the Study:

To propose and demonstrate a trainable diffractive optical neural network (TDONN) chip capable of handling multimodal data.
To overcome the limitations of single-modality processing in photonic deep learning.

Main Methods:

Development of a TDONN chip featuring on-chip diffractive optics with numerous tunable elements.
Implementation of a customized stochastic gradient descent algorithm and a drop-out mechanism for in situ training and fast optical convergence.
Single forward propagation for inference, minimizing optical-electrical conversion.

Main Results:

The TDONN chip demonstrates high performance with a potential throughput of 217.6 tera-operations per second (TOPS), computing density of 447.7 TOPS/mm², energy efficiency of 7.28 TOPS/W, and low optical latency of 30.2 ps.
Successfully achieved 85.7% accuracy in a four-class classification task across vision, audio, and touch modalities.
The chip integrates input, five hidden, and output layers for efficient deep learning operations.

Conclusions:

The developed TDONN chip offers a novel approach for multimodal deep learning using integrated photonic processors.
This technology presents a potential solution for low-power artificial intelligence (AI) large models.
Opens new avenues for advanced AI applications leveraging the capabilities of photonic technology.