Reconfigurable and In-sensor Computing Pb-free Perovskite Array towards Intelligent X-ray Imaging

Guan-Hua Dun,Ken Qin,Ze-Shu Wang,Xiang-Shun Geng,Yuan-Yuan Li,Peng Wan,Yi-Chu He,Ren-Rong Liang,Dan Xie,He Tian,Yi Yang,Tian-Ling Ren
DOI: https://doi.org/10.1109/IEDM45625.2022.10019447
2022-01-01
Abstract:Conventional X-ray imaging architectures feature data redundancy and hardware consumption due to the separated sensory terminal and computing units. In-sensor computing architectures is promising to overcome such drawbacks. However, its realization in X-ray range remains elusive. We propose ion distribution induced reconfigurable mechanism, and demonstrate the first X-ray band in-sensor computing array based on Pb-free perovskite. Redistribution of Br- ion in perovskite induces the switching of PN and NP modes under electrical pooling. X-ray detection sensitivity can be switched between two stable self-power sensing modes with 4373 +/- 298 and -7804 +/- 429 mu CGy(air)(-1)cm(-2) respectively, which are superior than that of commercial a-Se detectors (20 mu CGy(air)(-1)cm(-2)). Both modes exhibit low detection limit of 48.4 nGy(air) s(-1), which is two orders lower than typical medical dose rate of 5.5 mu Gy(air) s(-1). The perovskite array sensors can integrate with thin film transistors (TFTs) with low-temperature (80 degrees C) process with good uniformity. An in-sensor computing algorithm of attention mechanism is performed on array sensors for chest X-ray images COVID-19 recognition, which enables an accuracy improvement up to 98.2%. Our results can pave the way for future intelligent X-ray imaging.
What problem does this paper attempt to address?