A new artificial synapse, controlled entirely by light, enables in-sensor neuromorphic processing for more efficient and noise-resistant imaging systems.
Key Details
- 1Researchers engineered a synaptic device using a rare-earth-doped crystal with persistent optical afterglow.
- 2The synapse uses light for both input and state update, avoiding electrical signals and reducing energy demands.
- 3It mimics neural plasticity, showing both signal enhancement (UV facilitates) and suppression (near-infrared depresses).
- 4Integration with a silicon imaging sensor led to a neuromorphic camera prototype capable of in-sensor contrast enhancement and denoising.
- 5Neural networks using this optical synapse achieved nearly 96% accuracy in digit recognition, vs. 78% without the technique.
Why It Matters
By combining sensing, memory, and processing using optical signals, this approach points towards future imaging systems—potentially relevant to clinical imaging and radiology AI—that are faster, more power-efficient, and intrinsically robust to noise.
Source
EurekAlert
Related News
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.
•EurekAlert
AI Method Protects Sensitive Data in ECGs While Retaining Clinical Value
Researchers developed an AI model (PP-VAE) that safeguards sensitive personal data in electrocardiograms without sacrificing clinical utility.