scholarly journals Preliminary Planning for a Multi-institutional Database for Ultrasound Bone Segmentation

10.29007/m1ll ◽  
2019 ◽  
Author(s):  
Prashant Pandey ◽  
Hridayi Patel ◽  
Pierre Guy ◽  
Ilker Hacihalilogu ◽  
Antony J. Hodgson

Ultrasound (US) bone segmentation is a key component in many US-based computer assisted orthopaedic systems. Although numerous US bone segmentations techniques exist, there remains no direct way of comparing their performances. This is primarily due to the lack of an accessible US bone image database, and secondly due to a lack of standard vali- dation practices. To address this issue, we are beginning a multi-institutional international collaboration across multiple research centres with the aim of creating an open database for US bone segmentation consisting of several thousand US images and corresponding bone surface segmentations. Our collaboration also aims to address outstanding issues in US bone segmentation, such as determining the reliability of manual segmentations and establishing a set of evaluation metrics which should be reported in future segmentation studies. Finally, we strongly encourage interested researchers to join and contribute to this project as this will help to create a more diverse database and knowledgeable collaboration.


2020 ◽  
Vol 46 (4) ◽  
pp. 921-935 ◽  
Author(s):  
Prashant U. Pandey ◽  
Niamul Quader ◽  
Pierre Guy ◽  
Rafeef Garbi ◽  
Antony J. Hodgson


Author(s):  
P. Foroughi ◽  
E. Boctor ◽  
M. J. Swartz ◽  
R. H. Taylor ◽  
G. Fichtinger




2021 ◽  
Author(s):  
Xue Hu ◽  
Ferdinando Rodriguez y Baena

Abstract An automatic markerless knee tracking and registration algorithm has been proposed in the literature to avoid the marker insertion required by conventional computer-assisted knee surgery, resulting in a shorter and less invasive surgical workflow. However, such an algorithm considers intact femur geometry only. The bone surface modification is inevitable due to intra-operative intervention. The mismatched correspondences will degrade the reliability of registered target pose. To solve this problem, this work proposed a supervised deep neural network to automatically restore the surface of processed bone. The network was trained on a synthetic dataset that consists of real depth captures of a model leg and simulated realistic femur cutting. According to the evaluation on both synthetic data and real-time captures, the registration quality can be effectively improved by surface reconstruction. The improvement in tracking accuracy is only evident over test data, indicating the need for future enhancement of the dataset and network.



2015 ◽  
Author(s):  
Matthew Lougheed ◽  
Gabor Fichtinger ◽  
Tamas Ungi


2014 ◽  
Vol 493 ◽  
pp. 354-360
Author(s):  
Pei Yuan Lee ◽  
Jiing Yih Lai ◽  
Chung Yi Huang ◽  
Yu Sheng Hu

Abstract. The objective of this study is to present an integrated surgical simulation program on a personal computer for the preoperative planning of pelvic fractures. It first provides a visualization module to display 2D images and 3D model simultaneously. A semi-automatic bone segmentation module is then provided to separate the bony structures, enabling the manipulation of individual fractured bone and bone fragment. A bone reduction module is provided for the localization of the fractured bones. The simulation of plate and screw fixation is also presented, which provides useful information for determining the shape and size of the implants. Also, an example with real CT images are presented to demonstrate the feasibility of the proposed method.



Sign in / Sign up

Export Citation Format

Share Document