Space technology and space research have potential applications in many areas. A new innovative project by DTU Space Professor Irfan Kuvvetli is a prime example of this.
In collaboration with a range of partners, Irfan Kuvvetli has received 36 million DKK (4.85 million euros) for the project 'Intelligent Radiation Sensor System' (i-RASE) from the EU's Horizon Europe research and innovation program. The funds will be used to develop a unique intelligent radiation sensor system based on space technology that, for example, can be used in the healthcare sector.
In essence, the i-RASE project aims to significantly improve the accuracy and speed of radiation detection and measurements while reducing the digital data output in advanced imaging sensor techniques.
"With the i-RASE project, our aim is to revolutionize radiation detection by developing a real-time, photon-by-photon detector. By employing artificial neural networks for detector signal processing, also known as ANN, we can achieve very precise and energy-efficient detection and measurement of radiation interactions," says Irfan Kuvvetli.
"This system has the potential to impact several areas, including medical imaging, industrial inspection, and environmental monitoring."
The project spans four years under the leadership of Irfan Kuvvetli. It is carried out in collaboration with international partners, bringing together leading researchers and companies to establish what Kuvvetli call "a new paradigm in radiation detection".
Fast and accurate handling of large amounts of data
The idea is to develop electronics and software for a compact sensor system that can handle enormous amounts of data in a very short time.
For example, when generating images for physicians during X-ray screenings for breast cancer detection. This approach can lead to more accurate scans in shorter time frames with less equipment than currently used. In the most advanced sensor systems today, radiation data is stored before being processed by a computer to generate an image or spectrum.
With i-RASE, the radiation signals can be processed directly in the sensor, eliminating the need to store and process large amounts of data separately, thus saving both time and computer space. This is achieved through a newly developed i-RASE chip and software based on artificial intelligence. Moreover, the new technology occupies less space than conventional systems.
"In a development project from 2019, we conceptually demonstrated the potential of a new intelligent sensor system to interpret the collected data. The underlying technology, called 'ANN-based radiation detector signal processing,' has been patented," explains Irfan Kuvvetli.
"However, the electronics involved in ANN-based radiation detector signal processing require further development to handle the large amounts of data generated, for example, during an X-ray scan in a hospital. And that is the starting point of this new project, where we aim to develop specifications for a new chip, manufacture a prototype and develop AI-based radiation detector signal processing that will carry out all the calculations underlying the radiation image formation".
Real-time data processing with artificial intelligence
The i-RASE technology enables very fast image formation, taking just a few milliseconds, compared to current processes, which can take several minutes.
"Our system distinguishes itself from existing advanced detector systems by its ability to gather and process large volumes of intense radiation data in real time. With the i-RASE system, we process the total amount of radiation and the unique characteristics of each photon," says Irfan Kuvvetli.
"This is made possible by a combination of AI and our innovative radiation detector technology, initially developed for space instrumentation, which can detect radiation from all directions and extract high-resolution spectral, temporal, and spatial information".
Partners include, in addition to DTU Space and DTU Compute, Politecnico di Milano, University of Tübingen, and two companies, Integrated Detector Electronics AS (IDEAS) and Kromek Group plc., which develop and manufacture scanning systems for hospitals, among other applications.
DTU Space and DTU Compute will employ a project manager, an electronics engineer, two postdoc researchers, and two PhD students for the project.
It's a different task to detect X-rays on Earth
Irfan Kuvvetli has been involved in detector systems for space missions for many years, aimed at exploring exploding stars in the universe by capturing and analyzing the X-ray and gamma-ray radiation emitted from these extreme events. In space exploration, scientists deal with only a few signals, meaning that highly precise sensor systems are required.
The difference between the equipment in space and what is now being developed is that in space, there are few signals from individual sources from very distant events, which must be sorted out from a lot of noise from other light sources in the universe, while in X-ray scans on Earth, huge amounts of data from the source are obtained, which must then be processed very efficiently to form a usable scan image.
Irfan Kuvvetli's idea behind the new project is precisely based on his long experience with space technology development.
He thought it could be done faster and smarter by integrating a super-sensitive detector and readout unit with very fast data processing using AI, and that such a sensor system based on space technology could be used elsewhere.
Experts at Kromek also realized this when they became acquainted with Irfan Kuvvetli's idea. This led to the first project in 2018-21 and now to the next step, which is planned to lead to the manufacture and practical use of the new technology in healthcare and other areas where scans are performed.
It has the potential to revolutionize space exploration
In principle, the project could also revolutionize the exploration of the universe.
"Some of the very large detectors we use to explore phenomena in the universe by collecting light radiation may perhaps be replaced by the very small type of sensor we are developing. Here, we can quickly gather and process data on-site in space where the instrument is located and then send the processed data directly to researchers on Earth," explains the DTU professor.
"In this way, we can save weight on space missions and overcome the limitation that only small data packets can be sent from space to Earth at a time. Today, large amounts of data are collected in space, which can only be sorted and processed on Earth. Our system can sort before they are sent, providing entirely new opportunities in space exploration".
FACTS: Image sensor technology based on i-RASE (Intelligent Radiation Sensor System).
- High-energy radiation (electromagnetic emission and charged particle radiation) from different sources contain precise and valuable information regarding their origin. However, this radiation is extremely powerful and difficult to measure and interpret.
- With the i-RASE project, groundbreaking technology will be developed to create a sensor system capable of capturing and analyzing even the most intense high-energy radiation. In space research, high-energy radiation can reveal details about stars, black holes, subatomic particles, and the universe's fundamental building blocks.
- With i-RASE, new discoveries can be made in space. The system can also be used in medical imaging, industrial inspection, airport security, and environmental monitoring applications. The sensor system is based on a specialized detector and advanced signal processing (SP) capabilities using artificial intelligence (AI). It captures and analyzes data in real-time, decoding the complex information stored in the radiation in a very short time.
- The technology allows for the retrieval of all information about incoming radiation, in principle, improving measurement accuracy and speed while reducing the digital data output for subsequent processing.
