To Örebro University

PURPOSE: To develop, and evaluate the performance of, a deep learning based 3D convolutional neural network (CNN) artificial intelligence (AI) algorithm aimed at finding needles in ultrasound images used in prostate brachytherapy.

METHODS: Transrectal ultrasound (TRUS) image volumes from 1102 treatments were used to create a clinical ground truth (CGT) including 24422 individual needles that had been manually digitised by medical physicists during brachytherapy procedures. A 3D CNN U-net with 128x128x128 TRUS image volumes as input was trained using 17215 needle examples. Predictions of voxels constituting a needle were combined to yield a 3D linear function describing the localisation of each needle in a TRUS volume. Manual and AI digitisations were compared in terms of the root-mean-square distance (RMSD) along each needle, expressed as median and interquartile range (IQR). The method was evaluated on a dataset including 7207 needle examples. A subgroup of the evaluation data set (n=188) was created, where the needles were digitised once more by a medical physicist (G1) trained in brachytherapy. The digitisation procedure was timed.

RESULTS: The RMSD between the AI and CGT was 0.55 (IQR: 0.35-0.86) mm. In the smaller subset, the RMSD between AI and CGT was similar (0.52 [IQR: 0.33-0.79] mm) but significantly smaller (p<0.001) than the difference of 0.75 (IQR: 0.49-1.20) mm between AI and G1. The difference between CGT and G1 was 0.80 (IQR: 0.48-1.18) mm, implying that the AI performed as well as the CGT in relation to G1. The mean time needed for human digitisation was 10 min 11 sec, while the time needed for the AI was negligible.

CONCLUSIONS: A 3D CNN can be trained to identify needles in TRUS images. The performance of the network was similar to that of a medical physicist trained in brachytherapy. Incorporating a CNN for needle identification can shorten brachytherapy treatment procedures substantially.