BACKGROUND:Deep learning (DL) algorithms have shown promise in identifying and quantifying lesions in PET/CT. However, the accuracy and generalizability of these algorithms relies on large, diverse datasets which are time and labor intensive to curate. Modern PET/CT scanners may acquire data in list mode, allowing for multiple reconstructions of the same datasets with different parameters and imaging times. These reconstructions may provide a wide range of image characteristics to increase the size and diversity of datasets. Training algorithms with shorter imaging times and higher noise properties requires that lesions remain detectable. The purpose of this study is to model and predict the contrast-to-noise ratio (CNR) for shorter imaging times based on CNR from longer duration, lower noise images for 68Ga DOTATATE PET hepatic lesions and identify a threshold above which lesions remain detectable.METHODS:68Ga DOTATATE subjects (n=20) with hepatic lesions were divided into two subgroups. The "Model" group (n=4 subjects; n=9 lesions; n=36 datapoints) was used to identify the relationship between CNR and imaging time. The "Test" group (n=16 subjects; n=44 lesions; n=176 datapoints) was used to evaluate the prediction provided by the model.RESULTS:CNR plotted as a function of imaging time for a subset of identified subjects was very well fit with a quadratic model. For the remaining subjects, the measured CNR showed a very high linear correlation with the predicted CNR for these lesions (R2 > 0.97) for all imaging durations. From the model, a threshold CNR=6.9 at 5-minutes predicted CNR > 5 at 2-minutes. Visual inspection of lesions in 2-minute images with CNR above the threshold in 5-minute images were assessed and rated as a 4 or 5 (probably positive or definitely positive) confirming 100% lesion detectability on the shorter 2-minute PET images.CONCLUSIONS:CNR for shorter DOTATATE PET imaging times may be accurately predicted using list mode reconstructions of longer acquisitions. A threshold CNR may be applied to longer duration images to ensure lesion detectability of shorter duration reconstructions. This method can aid in the selection of lesions to include in novel data augmentation techniques for deep learning.

The 90Y PET images used for post-treatment dosimetry in SIRT of liver cancer are impacted by respiratory motion. Liver motion due to respiration will result in blurred images and can lead to inaccuracies of the dosimetric measures. Due to the low count statistics in 90Y PET data, the data-driven gating method does not work well, making the deviceless respiratory motion correction method unusable. The purpose of this study was to develop an alternative data-driven gating method to derive triggers directly from the list-mode data in order to produce respiratory motion compensated 90Y PET images and to assess the impact of respiratory motion on 90Y dosimetry. The steps for performing dosimetry have been adapted to improve dosimetry on the motion corrected PET images. Two different motion correction methods called Q.Static and Q.Rigid were investigated and evaluated on both NCAT phantom and patient data. The corrected images were compared to uncorrected respiratory motion PET images. Results demonstrate that both correction techniques lead to an improvement in dosimetry calculation for the liver tumor. Doses delivered to tumor were increased by 18.3% and 37.4% for Q.Static and Q.Rigid respectively for NCAT phantom; tumor doses were increased by 7.12 ± 12.04% and 8.76 ± 12.97% for Q.Static and Q.Rigid respectively, for patient data.

Deep neural networks have recently achieved impressive performance of automated tumor/lesion quantification with positron emission tomography (PET) imaging. However, deep learning usually requires a large amount of diverse training data, which is difficult for some applications such as neuroendocrine tumor (NET) image quantification, because of low incidence of the disease and expensive annotation of PET data. In addition, current deep lesion detection models often suffer from performance degradation when applied to PET images acquired with different scanners or protocols. In this paper, we propose a novel single-source domain generalization method, which learns with human annotation-free, list mode-synthesized PET images, for hepatic lesion identification in real-world clinical PET data. We first design a specific data augmentation module to generate out-of-domain images from the synthesized data, and incorporate it into a deep neural network for cross domain-consistent feature encoding. Then, we introduce a novel patch-based gradient reversal mechanism and explicitly encourage the network to learn domain-invariant features. We evaluate the proposed method on multiple cross-scanner $$^{68}$$ Ga-DOTATATE PET liver NET image datasets. The experiments show that our method significantly improves lesion detection performance compared with the baseline and outperforms recent state-of-the-art domain generalization approaches.

We developed a deep learning–based super-resolution model for prostate MRI. 2D T2-weighted turbo spin echo (T2w-TSE) images are the core anatomical sequences in a multiparametric MRI (mpMRI) protocol. These images have coarse through-plane resolution, are non-isotropic, and have long acquisition times (approximately 10–15 min). The model we developed aims to preserve high-frequency details that are normally lost after 3D reconstruction. We propose a novel framework for generating isotropic volumes using generative adversarial networks (GAN) from anisotropic 2D T2w-TSE and single-shot fast spin echo (ssFSE) images. The CycleGAN model used in this study allows the unpaired dataset mapping to reconstruct super-resolution (SR) volumes. Fivefold cross-validation was performed. The improvements from patch-to-volume reconstruction (PVR) to SR are 80.17%, 63.77%, and 186% for perceptual index (PI), RMSE, and SSIM, respectively; the improvements from slice-to-volume reconstruction (SVR) to SR are 72.41%, 17.44%, and 7.5% for PI, RMSE, and SSIM, respectively. Five ssFSE cases were used to test for generalizability; the perceptual quality of SR images surpasses the in-plane ssFSE images by 37.5%, with 3.26% improvement in SSIM and a higher RMSE by 7.92%. SR images were quantitatively assessed with radiologist Likert scores. Our isotropic SR volumes are able to reproduce high-frequency detail, maintaining comparable image quality to in-plane TSE images in all planes without sacrificing perceptual accuracy. The SR reconstruction networks were also successfully applied to the ssFSE images, demonstrating that high-quality isotropic volume achieved from ultra-fast acquisition is feasible.

Motivation: Using MR-priors in PET reconstruction has been performed in numerous studies. While MR-priors provide more SNR and better resolution ti PET-images, without considering motion there may be misalignment between PET and MR images which leads to crosstalk artifacts. Another concern is mismatch between MR and PET images which can potentially affect the final PET image. Goal(s): To compare the two most-widely used PET recon with MR-priors methods: the Bowsher's method and MRgBSREM. Approach: We created an MR-series with severe line artifacts and used in as MR-priors in both methods. Results: We have shown that MRgBSREM handles these mismatches better than the Bowsher’s method. Impact: MRgBSREM is more robust to mismatches between PET and MR. This is achieved by adding a PET-seed image to identify similar-voxels which avoids the situation in which voxels with significantly different PET-uptake would be considered similar based only on MR-images.