Analysis of a deep learning-based method for generation of SPECT projections based on a large Monte Carlo simulated dataset

Abstract

Background In recent years, a lot of effort has been put in the enhancement of medical imaging using artificial intelligence. However, limited patient data in combination with the unavailability of a ground truth often pose a challenge to a systematic validation of such methodologies. The goal of this work was to investigate a recently proposed method for an artificial intelligence-based generation of synthetic SPECT projections, for acceleration of the image acquisition process based on a large dataset of realistic SPECT simulations. Methods A database of 10,000 SPECT projection datasets of heterogeneous activity distributions of randomly placed random shapes was simulated for a clinical SPECT/CT system using the SIMIND Monte Carlo program. Synthetic projections at fixed angular increments from a set of input projections at evenly distributed angles were generated by different u-shaped convolutional neural networks (u-nets). These u-nets differed in noise realization used for the training data, number of input projections, projection angle increment, and number of training/validation datasets. Synthetic projections were generated for 500 test projection datasets for each u-net, and a quantitative analysis was performed using statistical hypothesis tests based on structural similarity index measure and normalized root-mean-squared error. Additional simulations with varying detector orbits were performed on a subset of the dataset to study the effect of the detector orbit on the performance of the methodology. For verification of the results, the u-nets were applied to Jaszczak and NEMA physical phantom data obtained on a clinical SPECT/CT system. Results No statistically significant differences were observed between u-nets trained with different noise realizations. In contrast, a statistically significant deterioration was found for training with a small subset (400 datasets) of the 10,000 simulated projection datasets in comparison with using a large subset (9500 datasets) for training. A good agreement between synthetic (i.e., u-net generated) and simulated projections before adding noise demonstrates a denoising effect. Finally, the physical phantom measurements show that our findings also apply for projections measured on a clinical SPECT/CT system. Conclusion Our study shows the large potential of u-nets for accelerating SPECT/CT imaging. In addition, our analysis numerically reveals a denoising effect when generating synthetic projections with a u-net. Clinically interesting, the methodology has proven robust against camera orbit deviations in a clinically realistic range. Lastly, we found that a small number of training samples (e.g., ~ 400 datasets) may not be sufficient for reliable generalization of the u-net.

Image-based dosimetry for [225Ac]Ac-PSMA-I&T therapy and the effect of daughter-specific pharmacokinetics

“…For instance, thicker crystals are highly desirable particularly in the context of imaging of alpha emitters and low therapeutic activities. Further, AI-based techniques as proposed by Leube et al [ 29 ] and Ryden et al [ 30 ] could assist in a further reduction of scanning time. Finally, an RBE of 5 was used for the absorbed dose calculations.…”

On the Use of Artificial Intelligence for Dosimetry of Radiopharmaceutical Therapies

“…al [33]) and no ground truth information, i. e., the real non-image derived tracer distribution, can be obtained. Dataset generated by Monte Carlo simulations bear the potential to overcome these limitations, as shown by Leube et al [34] for a synthetic dataset of 10,000 SPECT/CTs. An alternative to the omission of projections is to shorten the acquisition times per projection.…”

Pareto optimization of SPECT acquisition and reconstruction settings for 177Lu activity quantification

“…Even if noise is typically not a major source of uncertainty for MRT dosimetry, its effect will become substantial for imaging at very low SNRs [ 18 , 19 ]. There is currently a strive to reduce imaging times for clinical dosimetry [ 20 , 21 ], resulting in increased patient comfort and decreased risk of movement artefacts. The need for shorter acquisition protocols has become particularly pressing for SPECT of 177 Lu-PSMA, where it is necessary to acquire multiple bed positions to obtain adequate axial coverage, resulting in extensive acquisition times unless the time per bed-position is reduced.…”