Dataset Information

Learning to detect anatomical landmarks of the pelvis in X-rays from arbitrary views.

ABSTRACT: PURPOSE:Minimally invasive alternatives are now available for many complex surgeries. These approaches are enabled by the increasing availability of intra-operative image guidance. Yet, fluoroscopic X-rays suffer from projective transformation and thus cannot provide direct views onto anatomy. Surgeons could highly benefit from additional information, such as the anatomical landmark locations in the projections, to support intra-operative decision making. However, detecting landmarks is challenging since the viewing direction changes substantially between views leading to varying appearance of the same landmark. Therefore, and to the best of our knowledge, view-independent anatomical landmark detection has not been investigated yet. METHODS:In this work, we propose a novel approach to detect multiple anatomical landmarks in X-ray images from arbitrary viewing directions. To this end, a sequential prediction framework based on convolutional neural networks is employed to simultaneously regress all landmark locations. For training, synthetic X-rays are generated with a physically accurate forward model that allows direct application of the trained model to real X-ray images of the pelvis. View invariance is achieved via data augmentation by sampling viewing angles on a spherical segment of [Formula: see text]. RESULTS:On synthetic data, a mean prediction error of 5.6 ± 4.5 mm is achieved. Further, we demonstrate that the trained model can be directly applied to real X-rays and show that these detections define correspondences to a respective CT volume, which allows for analytic estimation of the 11 degree of freedom projective mapping. CONCLUSION:We present the first tool to detect anatomical landmarks in X-ray images independent of their viewing direction. Access to this information during surgery may benefit decision making and constitutes a first step toward global initialization of 2D/3D registration without the need of calibration. As such, the proposed concept has a strong prospect to facilitate and enhance applications and methods in the realm of image-guided surgery.

SUBMITTER: Bier B

PROVIDER: S-EPMC7297500 | biostudies-literature | 2019 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Learning to detect anatomical landmarks of the pelvis in X-rays from arbitrary views.

Bier Bastian B Goldmann Florian F Zaech Jan-Nico JN Fotouhi Javad J Hegeman Rachel R Grupp Robert R Armand Mehran M Osgood Greg G Navab Nassir N Maier Andreas A Unberath Mathias M

International journal of computer assisted radiology and surgery 20190420 9

<h4>Purpose</h4>Minimally invasive alternatives are now available for many complex surgeries. These approaches are enabled by the increasing availability of intra-operative image guidance. Yet, fluoroscopic X-rays suffer from projective transformation and thus cannot provide direct views onto anatomy. Surgeons could highly benefit from additional information, such as the anatomical landmark locations in the projections, to support intra-operative decision making. However, detecting landmarks is ...[more]

PMID: 31006106

Similar Datasets

Project description:3D free-hand ultrasound (3DFUS) is becoming increasingly popular to assist clinical gait analysis because it is cost- and time-efficient and does not expose participants to radiation. The aim of this study was to evaluate its reliability in localizing the anterior superior iliac spine (ASIS) at the pelvis and the hip joint centers (HJC). Additionally, we evaluated its accuracy to get a rough estimation of the potential to use of 3DFUS to segment bony surface. This could offer potential to register medical images to motion capture data in future. To evaluate reliability, a test-retest study was conducted in 16 lean and 19 obese individuals. The locations of the ASIS were determined by manual marker placement (MMP), an instrumented pointer technique (IPT), and with 3DFUS. The HJC location was also determined with 3DFUS. To quantify reliability, intraclass correlation coefficients (ICCs), the standard error of measurement (SEm), among other statistical parameters, were calculated for the identified locations between the test and retest. To assess accuracy, the surface of a human plastic pelvic phantom was segmented with 3DFUS in a distilled water bath in 27 trials and compared to a 3D laser scan of the pelvis. Regarding reliability, the MMP, but especially the IPT showed high reliability in lean (SEm: 2-3 mm) and reduced reliability in obese individuals (SEm: 6-15 mm). Compared to MMP and IPT, 3DFUS presented lower reliability in the lean group (SEm: 2-4 mm vs. 2-8 mm, respectively) but slightly better values in the obese group (SEm: 7-11 mm vs. 6-16 mm, respectively). Correlations between test-retest reliability and torso body fat mass (% of body mass) indicated a moderate to strong relationship for MMP and IPT but only a weak correlation for the 3DFUS approach. The water-bath experiments indicated an acceptable level of 3.5 (1.7) mm of accuracy for 3DFUS in segmenting bone surface. Despite some difficulties with single trials, our data give further rise to the idea that 3DFUS could serve as a promising tool in future to inform marker placement and hip joint center location, especially in groups with higher amount of body fat.

Project description:BackgroundBone shapes strongly influence force and moment predictions of kinematic and musculoskeletal models used in motion analysis. The precise determination of joint reference frames is essential for accurate predictions. Since clinical motion analysis typically does not include medical imaging, from which bone shapes may be obtained, scaling methods using reference subjects to create subject-specific bone geometries are widely used.Research questionThis study investigated if lower limb bone shape predictions from skin-based measurements, utilising an underlying statistical shape model (SSM) that corrects for soft tissue artefacts in digitisation, can be used to improve conventional linear scaling methods of bone geometries.MethodsSSMs created from 35 healthy adult femurs and tibiae/fibulae were used to reconstruct bone shapes by minimising the distance between anatomical landmarks on the models and those digitised in the motion laboratory or on medical images. Soft tissue artefacts were quantified from magnetic resonance images and then used to predict distances between landmarks digitised on the skin surface and bone. Reconstruction results were compared to linearly scaled models by measuring root mean squared distances to segmented surfaces, calculating differences of commonly used anatomical measures and the errors in the prediction of the hip joint centre.ResultsSSM reconstructed surface predictions from varying landmark sets from skin and bone landmarks were more accurate compared to linear scaling methods (2.60-2.95 mm vs. 3.66-3.87 mm median error; p < 0.05). No significant differences were found between SSM reconstructions from bony landmarks and SSM reconstructions from digitised landmarks obtained in the motion lab and therefore reconstructions using skin landmarks are as accurate as reconstructions from landmarks obtained from medical images.SignificanceThese results indicate that SSM reconstructions can be used to increase the accuracy in obtaining bone shapes from surface digitised experimental data acquired in motion lab environments.

Dataset Information

Learning to detect anatomical landmarks of the pelvis in X-rays from arbitrary views.

Publications

Learning to detect anatomical landmarks of the pelvis in X-rays from arbitrary views.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets