InfoBeyond is awarded to develop AI technology to Revolutionize Atom Trap Trace Analysis

The detection and analysis of radioisotopes are integral to a wide range of scientific research and technological applications, including archaeology, hydrology, cosmochemistry, biomedicine, and nuclear monitoring.

The traditional Low-Level Counting (LLC) measures radionuclide activities by directly observing radioactive decay, making it effective for short-lived isotopes such as ⁸⁵Kr. However, it necessitates extremely controlled environments to prevent disturbances from cosmic radiation and building materials, thus resulting in reduced measurement efficiencies. The Coincidence Counting technique can detect multiple nuclear particles nearly simultaneously by connecting two or more particle counters to an electronic coincidence circuit. However, the limitations in temporal resolution can introduce false coincidences, consequently raising noise levels. Nevertheless, its selectivity levels may be compromised due to the diverse physical and chemical properties of elements and isotopes. Even the more advanced High-Resolution Mass Spectrometry (HRMS), equipped with accelerators, struggles with low efficiency and requires long isotope half-lives, making it unsuitable for monitoring isotopes like ⁸⁵Kr in nuclear activities. In 1999, the U.S. DOE Argonne National Lab developed Atomic Trap Trace Analysis (ATTA) in order to offer high sensitivity, single/multiple atom detection, and superior selectivity, emerged as an exceptional tool for the precision detection of radionuclides in diverse fields.

Awarded by U.S. Defense Threat Reduction Agency (DTRA), InfoBeyond is developing a new technology namely,A²TTA (AI-based ATTA), to automate radioisotope detection and analysis accurately and effectively.A²TTA is an AI-based deep learning architecture to provide accurate radioisotope identification and atom/isotope quantification under various complicated contexts and operational environments. It can effectively learn a spectrum of image features to precisely identify the atom’s presence via classification in a full range of count rates, and quantify the atoms over the full range of abundance levels in real-time. Further A²TTA aims to remove the performance limitations of low detection/quantification accuracies due to the spurious photon counts; additional uncertainties during the quantification procedure; incapable of handling ultra-low/high abundance samples; low efficiency due to the prolonged processing/analysis turnaround times.

A²TTA can be seamlessly integrated into a variety of systems such as heavy-duty laser equipment and high-vacuum chambers integral to semiconductor manufacturing for a variety of applications, such as medical diagnosis, earth climate, solar neutrinos, and environmental studies. For example, it can provide deep insights into areas of climate modeling and global warming research. Its capability to quantify atoms across a wide range of abundance levels in different environmental mediums, including atmosphere, oceans, and groundwater, is unparalleled. These in-depth analyses offer valuable insight into climate-related phenomena and changes, greatly contributing to our understanding of complex environmental dynamics. In the realm of hydrology, A²TTA brings about substantial advancement through its real-time intelligent analysis. It can decode the nature of water sources and flow paths, contributing significantly to water resource management, accurate drought prediction, and effective pollution monitoring with ultra-fast processing speed.