High throughput software-based gradient tree boosting positioning for PET systems.

The supervised machine learning technique Gradient Tree Boosting (GTB) has shown good accuracy for position estimation of gamma interaction in PET crystals for bench-top experiments while its computational requirements can easily be adjusted. Transitioning to preclinical and clinical applications requires near real-time processing in the scale of full PET systems. In this work, a high throughput GTB-based singles positioning C++ implementation is proposed and a series of optimizations are evaluated regarding their effect on the achievable processing throughput. Moreover, the crucial feature and parameter selection for GTB is investigated for the segmented detectors of the Hyperion IIDPET insert with two main models and a range of GTB hyperparameters. The proposed framework achieves singles positioning throughputs of more than 9.5 GB/s for smaller models and of 240 MB/s for more complex models on a recent Intel Skylake server. Detailed throughput analysis reveals the key performance limiting factors, and an empirical throughput model is derived to guide the GTB model selection process and scanner design decisions. The throughput model allows for throughput estimations with a mean absolute error (MAE) of 175.78 MB/s.

Evaluation of the radiofrequency performance of a wide-bore 1.5 T positron emission tomography/magnetic resonance imaging body coil for radiotherapy planning.

BACKGROUND AND PURPOSE: The restricted bore diameter of current simultaneous positron emission tomography/magnetic resonance imaging (PET/MRI) systems can be an impediment to achieving similar patient positioning during PET/MRI planning and radiotherapy. Our goal was to evaluate the B1 transmit (B1 +) uniformity, B1 + efficiency, and specific absorption rate (SAR) of a novel radiofrequency (RF) body coil design, in which RF shielded PET detectors were integrated with the specific aim of enabling a wide-bore PET/MRI system. MATERIALS AND METHODS: We designed and constructed a wide-bore PET/MRI RF body coil to be integrated with a clinical MRI system. To increase its inner bore diameter, the PET detectors were positioned between the conductors and the RF shield of the RF body coil. Simulations and experiments with phantoms and human volunteers were performed to compare the B1 + uniformity, B1 + efficiency, and SAR between our design and the clinical body coil. RESULTS: In the simulations, our design achieved nearly the same B1 + field uniformity as the clinical body coil and an almost identical SAR distribution. The uniformity findings were confirmed by the physical experiments. The B1 + efficiency was 38% lower compared to the clinical body coil. CONCLUSIONS: To achieve wide-bore PET/MRI, it is possible to integrate shielding for PET detectors between the body coil conductors and the RF shield without compromising MRI performance. Reduced B1 + efficiency may be compensated by adding a second RF amplifier. This finding may facilitate the application of simultaneous whole-body PET/MRI in radiotherapy planning.

Characterization methods for comprehensive evaluations of shielding materials used in an MRI.

PURPOSE: In order to integrate electronic devices into a magnetic resonance imaging (MRI) scanner, shielding of the electronics with respect to the radio frequency (RF) transmit and receive system of the MRI scanner is required. Furthermore, MRI uses time-varying low-frequency magnetic fields for spatial encoding, i.e., the gradient magnetic fields. Time-varying magnetic fields induce eddy currents in all conductive elements. The eddy currents result in opposing magnetic fields, which can cause distortions of the magnetic resonance (MR) image. As shielding of lower frequencies is not feasible in this respect, an ideal shielding element should be transparent for gradient magnetic fields while providing a high RF shielding effectiveness. Furthermore, it should offer a low susceptibility to prevent distortion of the main magnetic field of the MRI. In this work, we characterize the aforesaid shielding parameters of different shielding samples. METHODS: We developed a nuclear magnetic resonance (NMR) probe to measure the magnetic fields to quantify the field distortions time-resolvedly. The relative distortion was introduced as a proportionality constant relating the eddy-current-inducing field changes and the field distortions. The relative distortion was measured in the frequency range from 0 to 10 kHz for all shielding samples using the NMR probe. We characterized the shielding effectiveness of the samples in the frequency range from 1 to 150 MHz using a network analyzer. We conducted all measurements with three different materials, two carbon fiber composites and copper, each in various thicknesses. RESULTS: The relative distortion of the magnetic fields induced by the carbon fiber composites samples was at least a factor of seven lower than the copper sample. A linear dependency on the sample thickness was measured for the main field distortion, the relative distortion and the shielding effectiveness. The relative distortion was roughly independent of the gradient frequency contrary to the shielding effectiveness, highly depending on the RF frequency. CONCLUSIONS: We presented a very sensitive method to characterize the distortion of the main field distortion and the gradient transparency using an NMR probe. We analyzed different shielding materials regarding the main field distortion, the gradient transparency, and the shielding effectiveness. From the tested materials, we identified a carbon fiber composite with the lowest distortion on the MRI.

Low-dose single acquisition rest (99m)Tc/stress (201)Tl myocardial perfusion SPECT protocol: phantom studies and clinical validation.

PURPOSE: We developed and tested a single acquisition rest (99m)Tc-sestamibi/stress (201)Tl dual isotope protocol (SDI) with the intention of improving the clinical workflow and patient comfort of myocardial perfusion single photon emission computed tomography (SPECT). METHODS: The technical feasibility of SDI was evaluated by a series of anthropomorphic phantom studies on a standard SPECT camera. The attenuation map was created by a moving transmission line source. Iterative reconstruction including attenuation correction, resolution recovery and Monte Carlo simulation of scatter was used for simultaneous reconstruction of dual tracer distribution. For clinical evaluation, patient studies were compared to stress (99m)Tc and rest (99m)Tc reference images acquired in a 2-day protocol. Clinical follow-up examinations like coronary angiography (CAG) and fractional flow reserve (FFR) were included in the assessment if available. RESULTS: Phantom studies demonstrated the technical feasibility of SDI. Artificial lesions inserted in the phantom mimicking ischaemia could be clearly identified. In 51/53 patients, the image quality was adequate for clinical evaluation. For the remaining two obese patients with body mass index > 32 the injected (201)Tl dose of 74 MBq was insufficient for clinical assessment. In answer to this the (201)Tl dose was adapted for obese patients in the rest of the study. In 31 patients, SDI and (99m)Tc reference images resulted in equivalent clinical assessment. Significant differences were found in 20 patients. In 18 of these 20 patients additional examinations were available. In 15 patients the diagnosis based on the SDI images was confirmed by the results of CAG or FFR. In these patients the SDI images were more accurate than the (99m)Tc reference study. In three patients minor ischaemic lesions were detected by SDI but were not confirmed by CAG. In one of these cases this was probably caused by pronounced apical thinning. For two patients no relevant clinical follow-up information was available for evaluation. CONCLUSION: The proposed SDI protocol has the potential to improve clinical workflow and patient comfort and suggests improved accuracy as demonstrated in the clinical feasibility study.