Active Computing & AI Lungs & Breathing

Intelligent Radiation Sensor Readout System

In plain English

AI plain-English summary

A single chip will process each particle of radiation as it arrives, turning a detector into an intelligent sensor that outputs data in real time. Today’s radiation detectors generate a flood of raw data that must be sent to a separate computer for analysis, which slows down imaging and consumes power. The i-RASE project embeds artificial neural networks—algorithms inspired by the brain’s wiring—directly into the sensor’s hardware. This allows the chip to classify and interpret each photon or particle on the fly, discarding noise and keeping only useful information. If the chip works as designed, it could shrink the size, cost, and energy use of radiation detectors while boosting their speed and accuracy. In medical imaging, that might mean lower radiation doses for patients and faster scans. In industrial inspection, it could enable real-time quality checks on production lines. Space instruments and environmental monitors would also benefit from a compact, low-power sensor that sends back only the data that matters. The work is primarily an engineering and fundamental science challenge—building a new class of sensor-system-on-a-chip that merges physics-based signal processing with neuromorphic computing. There is no immediate clinical or commercial product, but the underlying approach could reshape how any radiation-based system handles data.

View original technical description
The vision of i-RASE is to pioneer a new class of radiation sensor system-in-package (SIP) chips at the intersection of computer scienceand neuroscience-oriented approaches to artificial intelligence (AI) development. The i-RASE project aims to design, build, test, and implement the first on-the-fly photon-by-photon radiation detector with transformational potential for various radiation applications, such as medical imaging, industrial inspection, scientific space instrumentation, environmental monitoring, and more. The i-RASE project will develop physics-inspired artificial neural networks (ANNs) for comprehensive sensor signal processing (SP) and real-time (RT) measurement of radiation interactions. It will compact this technology into an ultimate vision for SP embedded in hardware (HW) as an "all-in-one" SIP, enabling cost- and energy-efficient detection and intelligent radiation data output with unparalleled accuracy and speed. This approach enhances measurement precision and speed by utilizing complex SP, event characterization, and on-the-fly processing of incident radiation-induced signals in near real-time. As a result, it facilitates the retrieval of comprehensive information on incident radiation, ultimately improving measurement accuracy and speed while reducing digital data output.

View the original record at the funder ↗

Related Research

Grants with similar aims, by meaning.

Intelligent Radiation Sensor Readout System​
Smart Radiation Sensor for Intelligent Nuclear Robots
Digital Single Photon Sensors to Enable a Step change in Civil Radiation Detection
Single Photon Infrared Imaging, Detection and Ranging (SPIDAR)
Single Photon Detection using AlGaAsSb Avalanche Photodiodes

Original classification

EU-Funded

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.