Transformer-CNN hybrid network for improving PET time of flight prediction

Abstract

In positron emission tomography (PET) reconstruction, the integration of Time-of-Flight (TOF) information, known as TOF-PET, has been a major research focus. Compared to traditional reconstruction methods, the introduction of TOF enhances the signal-to-noise ratio (SNR) of images. Precision in TOF is measured by Full Width at Half Maximum (FWHM) and the offset from ground truth, referred to as coincidence time resolution (CTR) and bias. 
This study proposes a network combining Transformer and Convolutional Neural Network (CNN) to utilize TOF information from detector waveforms, using event waveform pairs as inputs. This approach integrates the global self-attention mechanism of Transformer, which focuses on temporal relationships, with the local receptive field of CNN. The combination of global and local information allows the network to assign greater weight to the rising edges of waveforms, thereby extracting valuable temporal information for precise TOF predictions. Experiments were conducted using lutetium yttrium oxyorthosilicate (LYSO) scintillators and silicon photomultiplier (SiPM) detectors. The network was trained and tested using the waveform datasets after cropping.
Compared to the constant fraction discriminator (CFD), CNN, CNN with attention, Long Short-Term Memory (LSTM) and Transformer, our network achieved an average CTR of 189 ps, reducing it by 82 ps (more than 30%), 13 ps (6.4%), 12 ps (6.0%), 16 ps (7.8%) and 9 ps (4.6%), respectively. Additionally, a reduction of 10.3, 8.7, 6.7 and 4 ps in average bias was achieved compared to CNN, CNN with attention, LSTM and Transformer.
This work demonstrates the potential of applying the Transformer for PET TOF estimation using real experimental data. Through the integration of both CNN and Transformer with local and global attention, it achieves optimal performance, thereby presenting a novel direction for future research in this field.

Holistic evaluation of a machine learning-based timing calibration for PET detectors under varying data sparsity

“…One way to approach this circumstance is the usage of machine learning, which is given as classical learning systems (Naunheim et al 2023a) or neural networks (Chen and Liu 2021), with its capabilities to transcend traditional static or linear timestamp adjustments by providing event-by-event corrections. Recent research demonstrated the capabilities of using waveform information to improve CTR (Berg and Cherry 2018, Ai et al 2021, Onishi et al 2022, Feng et al 2024. In most cases, the researchers decided to use convolutional neural networks (CNNs) due to their ability to capture local patterns and invariance against shifts given by the architecture that exploits convolutional filters.…”