Cosmas Mwikirize1, John L Nosher2, Ilker Hacihaliloglu3,2. 1. Department of Biomedical Engineering, Rutgers University, Piscataway, NJ, 08854, USA. cosmas.mwikirize@rutgers.edu. 2. Department of Radiology, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, 08901, USA. 3. Department of Biomedical Engineering, Rutgers University, Piscataway, NJ, 08854, USA.
Abstract
PURPOSE: We propose a framework for automatic and accurate detection of steeply inserted needles in 2D ultrasound data using convolution neural networks. We demonstrate its application in needle trajectory estimation and tip localization. METHODS: Our approach consists of a unified network, comprising a fully convolutional network (FCN) and a fast region-based convolutional neural network (R-CNN). The FCN proposes candidate regions, which are then fed to a fast R-CNN for finer needle detection. We leverage a transfer learning paradigm, where the network weights are initialized by training with non-medical images, and fine-tuned with ex vivo ultrasound scans collected during insertion of a 17G epidural needle into freshly excised porcine and bovine tissue at depth settings up to 9 cm and [Formula: see text]-[Formula: see text] insertion angles. Needle detection results are used to accurately estimate needle trajectory from intensity invariant needle features and perform needle tip localization from an intensity search along the needle trajectory. RESULTS: Our needle detection model was trained and validated on 2500 ex vivo ultrasound scans. The detection system has a frame rate of 25 fps on a GPU and achieves 99.6% precision, 99.78% recall rate and an [Formula: see text] score of 0.99. Validation for needle localization was performed on 400 scans collected using a different imaging platform, over a bovine/porcine lumbosacral spine phantom. Shaft localization error of [Formula: see text], tip localization error of [Formula: see text] mm, and a total processing time of 0.58 s were achieved. CONCLUSION: The proposed method is fully automatic and provides robust needle localization results in challenging scanning conditions. The accurate and robust results coupled with real-time detection and sub-second total processing make the proposed method promising in applications for needle detection and localization during challenging minimally invasive ultrasound-guided procedures.
PURPOSE: We propose a framework for automatic and accurate detection of steeply inserted needles in 2D ultrasound data using convolution neural networks. We demonstrate its application in needle trajectory estimation and tip localization. METHODS: Our approach consists of a unified network, comprising a fully convolutional network (FCN) and a fast region-based convolutional neural network (R-CNN). The FCN proposes candidate regions, which are then fed to a fast R-CNN for finer needle detection. We leverage a transfer learning paradigm, where the network weights are initialized by training with non-medical images, and fine-tuned with ex vivo ultrasound scans collected during insertion of a 17G epidural needle into freshly excised porcine and bovine tissue at depth settings up to 9 cm and [Formula: see text]-[Formula: see text] insertion angles. Needle detection results are used to accurately estimate needle trajectory from intensity invariant needle features and perform needle tip localization from an intensity search along the needle trajectory. RESULTS: Our needle detection model was trained and validated on 2500 ex vivo ultrasound scans. The detection system has a frame rate of 25 fps on a GPU and achieves 99.6% precision, 99.78% recall rate and an [Formula: see text] score of 0.99. Validation for needle localization was performed on 400 scans collected using a different imaging platform, over a bovine/porcine lumbosacral spine phantom. Shaft localization error of [Formula: see text], tip localization error of [Formula: see text] mm, and a total processing time of 0.58 s were achieved. CONCLUSION: The proposed method is fully automatic and provides robust needle localization results in challenging scanning conditions. The accurate and robust results coupled with real-time detection and sub-second total processing make the proposed method promising in applications for needle detection and localization during challenging minimally invasive ultrasound-guided procedures.
Authors: M Barva; M Uhercik; J M Mari; J Kybic; J R Duhamel; H Liebgott; V Hlavac; C Cachard Journal: IEEE Trans Ultrason Ferroelectr Freq Control Date: 2008-07 Impact factor: 2.725
Authors: Geert Litjens; Thijs Kooi; Babak Ehteshami Bejnordi; Arnaud Arindra Adiyoso Setio; Francesco Ciompi; Mohsen Ghafoorian; Jeroen A W M van der Laak; Bram van Ginneken; Clara I Sánchez Journal: Med Image Anal Date: 2017-07-26 Impact factor: 8.545
Authors: John Moore; Colin Clarke; Daniel Bainbridge; Chris Wedlake; Andrew Wiles; Danielle Pace; Terry Peters Journal: Med Image Comput Comput Assist Interv Date: 2009
Authors: E Checcucci; D Amparore; G Volpi; F Piramide; S De Cillis; A Piana; P Alessio; P Verri; S Piscitello; B Carbonaro; J Meziere; D Zamengo; A Tsaturyan; G Cacciamani; Juan Gomez Rivas; S De Luca; M Manfredi; C Fiori; E Liatsikos; F Porpiglia Journal: World J Urol Date: 2021-09-01 Impact factor: 3.661
Authors: Valentina Bellini; Emanuele Rafano Carnà; Michele Russo; Fabiola Di Vincenzo; Matteo Berghenti; Marco Baciarello; Elena Bignami Journal: Ann Transl Med Date: 2022-05