6–10 Jul 2025
Bratislava, Slovakia
Europe/Zurich timezone

Testing setup of IntPixel hybrid X-ray detector with on-chip and in-pixel artificial neural network

Not scheduled
20m
Bratislava, Slovakia

Bratislava, Slovakia

poster

Speaker

Mateusz Jurczak (AGH)

Description

ABSTRACT

We present a testing setup for a hybrid pixel array (HPAD) radiation detector with an on-chip and in-pixel Artificial Neural Network (ANN). The setup takes advantage of the IC design, which supports testing of it’s individual blocks, and allows for full characterization of the IC as well as facilitates training of per-pixel ANN.

The setup is dedicated to testing of an “IntPixel” detector [1], [2]. The IntPixel integrated circuit (IC) was designed at AGH University of Krakow as a matrix of 8 x 8 square-shaped pixels, each 200 µm pitch. In every pixel, there is a charge-sensitive amplifier followed by a shaper, an analog-to-digital converter, and ANN.

The presented system design emphasize focusing on device-under-test rather than embedded system design itself. It consists of NI sbRIO 9609, an embedded and commercially available controller incorporated with an FPGA and microcontroller. It is connected to the power and bias distribution board, that also hosts a small daugherboard with the IntPixel ASIC. Along with basic IO operations, presented system provides fast data analysis and high data throughout. The testing IC is packaged and uses a socket allowing for quick exchange of individual units. We present details of our platform allowing for full characterization of all the IC’s components including charge amplifier, ADC, discriminator and an ANN.

The research leading to these results has received funding from the Norway Grants 2014–2021 via the National Centre for Research and Development (research project NOR/SGS/Intelligent_XRay_Det/0196/2020-00).

REFERENCES

[1] A. Koziol et al., “Semiconductor sensor readout integrated circuit with in-pixel artificial neural network for pulse amplitude measurement", IEEE Nuclear Science Symposium, Vancouver, Canada, 2023

[2] A. Koziol et al., “Artificial neural network on-chip and in-pixel implementation towards pulse amplitude measurement,” J. Instrum., vol. 18, no. 2, 2023, doi: 10.1088/1748-0221/18/02/C02048.

Workshop topics Detector systems

Authors

Anna Kozioł (AGH University of Krakow) Mateusz Jurczak (AGH) Paulina Marchut (AGH) Piotr Maj (AGH UST)

Presentation materials