Deep learning for instrumented ultrasonic tracking:from synthetic training data to in vivo application

Instrumented ultrasonic tracking is used to improve needle localisation during ultrasound guidance of minimally-invasive percutaneous procedures. Here, it is implemented with transmitted ultrasound pulses from a clinical ultrasound imaging probe that are detected by a fibre-optic hydrophone integrated into a needle. The detected transmissions are then reconstructed to form the tracking image. Two challenges are considered with the current implementation of ultrasonic tracking. First, tracking transmissions are interleaved with the acquisition of B-mode images and thus, the effective B-mode frame rate is reduced. Second, it is challenging to achieve an accurate localisation of the needle tip when the signal-to-noise ratio is low. To address these challenges, we present a framework based on a convolutional neural network (CNN) to maintain spatial resolution with fewer tracking transmissions and to enhance signal quality. A major component of the framework included the generation of realistic synthetic training data. The trained network was applied to unseen synthetic data and experimental in vivo tracking data. The performance of needle localisation was investigated when reconstruction was performed with fewer (up to eight-fold) tracking transmissions. CNN-based processing of conventional reconstructions showed that the axial and lateral spatial resolution could be improved even with an eight-fold reduction in tracking transmissions. The framework presented in this study will significantly improve the performance of ultrasonic tracking, leading to faster image acquisition rates and increased localisation accuracy.