Initial investigations of scatter cross-talk in simultaneous biplane high-speed 1000 frames per second neuro-angiography using Monte Carlo simulations.

The Aries photon counting detector (PCD) by Direct Conversion Inc. can image up to 1000 frames per second and is used to track contrast bolus in neuro-vasculature for hemodynamic calculations. For 3D tracking, synchronized biplane imaging with 1 ms acquisition times is used such that both imaging planes are exposed simultaneously. This leads to cross-scattered radiation being detected and a degradation of image quality compared to single-plane imaging. In this study, we utilize Monte Carlo (MC) methods to quantify the increase in scatter due to cross-talk without the use of a radiographic grid. EGSnrc biplane simulations were performed with the Zubal anthropomorphic head phantom. The total scatter plus primary and cross-scatter was calculated in the imaging planes for two orthogonal AP and lateral beams with a field size consistent with the 7.5×5 cm Aries detector, while the primary was determined with a 1×1 mm beam. The forward scatter was then determined from the difference between total and primary. The scatter is seen to increase by 4%-56% for AP projections and 48%-71% for lateral projections depending on detector orientation during simultaneous exposure. Scatter degradation from cross-talk can be reduced using an anti-scatter grid as well as the energy thresholding capabilities of the Aries PCD.

Real-time Detection of Patient Head Position and Cephalometric Landmarks from Neuro-Interventional Procedure Images Using Machine Learning for Patient Eye-Lens Dose Prediction.

A deep learning (DL) model has been developed to estimate patient-lens dose in real-time for given exposure and geometric conditions during fluoroscopically-guided neuro-interventional procedures. Parameters input into the DL model for dose prediction include the patient head shift from isocenter and cephalometric landmark locations as a surrogate for head size. Machine learning (ML) models were investigated to automatically detect these parameters from the in-procedure fluoroscopic image. Fluoroscopic images of a Kyoto Kagaku anthropomorphic head phantom were taken at various known X (transverse) and Y (longitudinal) shifts, as well as different magnification modes, to create an image database. For each image, anatomical landmark coordinate locations were obtained manually using ImageJ and are used as ground-truth labels for training. This database was then used to train the two separate ML models. One ML model predicts the patient head shift in both the X and Y directions and the other model predicts the coordinates of the anatomical landmarks. From the coordinates, the distance between these anatomical landmarks is calculated, and input into the DL dose-prediction model. Model performance was evaluated using mean absolute error (MAE) and mean absolute percentage error (MAPE) for the head-shift and landmark-coordinate models, respectively. The goal is to implement these two separate models into the Dose Tracking System (DTS) developed by our group. This would allow the DTS to automatically detect the patient head size and position for eye-lens dose prediction and eliminate the need for manual input by the clinical staff.

Estimating Compton scatter distributions with a regressional neural network for use in a real-time staff dose management system for fluoroscopic procedures.

Staff-dose management in fluoroscopic procedures is a continuing concern due to insufficient awareness of radiation dose levels. To maintain dose as low as reasonably achievable (ALARA), we have developed a software system capable of monitoring the procedure room scattered radiation and the dose to staff members in real-time during fluoroscopic procedures. The scattered-radiation display system (SDS) acquires imaging-system signal inputs to update technique and geometric parameters used to provide a color-coded mapping of room scatter. We have calculated a discrete look-up-table (LUT) of scatter distributions using Monte-Carlo (MC) software and developed an interpolation technique for the multiple parameters known to alter the spatial shape of the distribution. However, the file size for the LUT's can be large (~2GB), leading to long SDS installation times in the clinic. Instead, this work investigated the speed and accuracy of a regressional neural network (RNN) that we developed for predicting the scatter distribution from imaging-system inputs without the need for the LUT and interpolation. This method greatly reduces installation time while maintaining real-time performance. Results using error maps derived from the structural similarity index indicate high visual accuracy of predicted matrices when compared to the MC-calculated distributions. Dose error is also acceptable with a matrix element-averaged percent error of 31%. This dose-monitoring system for staff members can lead to improved radiation safety due to immediate visual feedback of high-dose regions in the room during the procedure as well as enhanced reporting of individual doses post-procedure.

Using a convolutional neural network for human recognition in a staff dose management software for fluoroscopic interventional procedures.

Staff dose management is a continuing concern in fluoroscopically-guided interventional (FGI) procedures. Being unaware of radiation scatter levels can lead to unnecessarily high stochastic and deterministic risks due to the effects of absorbed dose by staff members. Our group has developed a scattered-radiation display system (SDS) capable of monitoring system parameters in real-time using a controller-area network (CAN) bus interface and displaying a color-coded mapping of the Compton-scatter distribution. This system additionally uses a time-of-flight depth sensing camera to track staff member positional information for dose rate updates. The current work capitalizes on our body tracking methodology to facilitate individualized dose recording via human recognition using 16-bit grayscale depth maps acquired using a Microsoft Kinect V2. Background features are removed from the images using a depth threshold technique and connected component analysis, which results in a body silhouette binary mask. The masks are then fed into a convolutional neural network (CNN) for identification of unique body shape features. The CNN was trained using 144 binary masks for each of four individuals (total of 576 images). Initial results indicate high-fidelity prediction (97.3% testing accuracy) from the CNN irrespective of obstructing objects (face masks and lead aprons). Body tracking is still maintained when protective attire is introduced, albeit with a slight increase in positional data error. Dose reports are then able to be produced which contain cumulative dose to each staff member at the eye lens level, waist level, and collar level. Individualized cumulative dose reporting through the use of a CNN in addition to real-time feedback in the clinic will lead to improved radiation dose management.