Automatic detection of scintillation light splashes using conventional and deep learning methods

Abstract

Six methods for the automatic detection of scintillation light splashes in a portable gamma camera are compared. Each imaging frame might contain any number of light splashes (including none), and the location and size of each light splash must be identified. For real-time imaging, splashes must be identified and characterised quickly and with minimal processing overhead. The techniques are compared on their ability to accurately determine the number, position, and size of light splashes, and to reconstruct the deposited energy within each splash for a simulated data set with known ground-truths. The speed of each technique and the ease of implementation are also discussed. For accuracy in blob (light splash) localisation, a Laplacian of Gaussian approach was found to provide the most accurate estimation. However, its performance greatly relies on the appropriate tuning of preprocessing parameters prior to image analysis and the number of blobs in each frame. Deep learning approaches (Faster-RCNNs) performed significantly better than traditional algorithms in terms of predicting the size of each light splash, did not require image preprocessing and were also more stable over a range of frame occupancies. Moreover, the paper fine-tuned a VGG16 based Faster-RCNN model with the simulated data set for the scintillation light splash detection, called DeepSplashSpotter (DSS).

Intraoperative Gamma Cameras: A Review of Development in the Last Decade and Future Outlook

“…Charge-sharing correction algorithms, which have already been implemented for multiple detector systems, offer the ability to recover the spectral performance lost by small-anode-pixel semiconductor detectors and allow for exceptional energy resolution and spatial resolution simultaneously [118,147,148]. The application of deep learning and neural networks to the optimisation of current dataprocessing tasks looks to advance IGCs in a range of ways, including: improved energy resolution reconstruction, sub-pixel event positioning, improved event localisation, and improved near-field coded-aperture image reconstruction [119,124,149,150].…”