Detector development for microPET II: a 1 microl resolution PET scanner for small animal imaging.

We are currently developing a small animal positron emission tomography (PET) scanner with a design goal of 1 microlitre (1 mm3) image resolution. The detectors consist of a 12 x 12 array of 1 x 1 x 10 mm lutetium oxyorthosilicate (LSO) scintillator crystals coupled to a 64-channel photomultiplier tube (PMT) via 5 cm long optical fibre bundles. The optical fibre connection allows a high detector packing fraction despite the dead space surrounding the active region of the PMT. Optical fibre bundles made from different types of glass were tested for light transmission, and also their effects on crystal identification and energy resolution, and compared to direct coupling of the LSO arrays to the PMTs. We also investigated the effects of extramural absorber (EMA) in the fibre bundles. Based on these results, fibre bundles manufactured from F2 glass were selected. We built three pairs of prototype detectors (directly coupled LSO array, fibre bundle without EMA and fibre bundle with EMA) and measured flood histograms, energy resolution, intrinsic spatial resolution and timing resolution. The results demonstrated an intrinsic spatial resolution (FWHM) of 1.12 mm (directly coupled), 1.23 mm (fibre bundle without EMA coupling) and 1.27 mm (fibre bundle with EMA coupling) using an approximately 500 microm diameter Na-22 point source. Using a 330 microm outer diameter steel needle line source filled with F-18, spatial resolution for the detector with the EMA optical fibre bundle improved to 1.05 mm. The respective timing and energy FWHM values were 1.96 ns, 21% (directly coupled), 2.20 ns, 23% (fibre bundle without EMA) and 2.99 ns, 30% (fibre bundle with EMA). The peak-to-valley ratio in the flood histograms was better with EMA (5:1) compared to the optical fibre bundle without EMA (2.5:1), due to the decreased optical cross-talk. In comparison to the detectors used in our current generation microPET scanner, these detectors substantially improve on the spatial resolution, preserve the timing resolution and provide adequate energy resolution for a modern high-resolution animal PET tomograph.     

Machine learning approach for obstructive sleep apnea screening using brain diffusion tensor imaging

Patients with obstructive sleep apnea (OSA) show autonomic, mood, cognitive, and breathing dysfunctions that are linked to increased morbidity and mortality, which can be improved with early screening and intervention. The gold standard and other available methods for OSA diagnosis are complex, require whole-night data, and have significant wait periods that potentially delay intervention. Our aim was to examine whether using faster and less complicated machine learning models, including support vector machine (SVM) and random forest (RF), with brain diffusion tensor imaging (DTI) data can classify OSA from healthy controls. We collected two DTI series from 59 patients with OSA [age: 50.2 ± 9.9 years; body mass index (BMI): 31.5 ± 5.6 kg/m²; apnea-hypopnea index (AHI): 34.1 ± 21.2 events/h 23 female] and 96 controls (age: 51.8 ± 9.7 years; BMI: 26.2 ± 4.1 kg/m²; 51 female) using a 3.0-T magnetic resonance imaging scanner. Using DTI data, mean diffusivity maps were calculated from each series, realigned and averaged, normalised to a common space, and used to conduct cross-validation for model training and selection and to predict OSA. The RF model showed 0.73 OSA and controls classification accuracy and 0.85 area under the curve (AUC) value on the receiver-operator curve. Cross-validation showed the RF model with comparable fitting over SVM for OSA and control data (SVM; accuracy, 0.77; AUC, 0.84). The RF ML model performs similar to SVM, indicating the comparable statistical fitness to DTI data. The findings indicate that RF model has similar AUC and accuracy over SVM, and either model can be used as a faster OSA screening tool for subjects having brain DTI data.