Classification of micro-CT images using 3D characterization of bone canal patterns in human osteogenesis imperfecta

Few studies have analyzed the microstructural properties of bone in cases of Osteogenenis Imperfecta (OI), or ‘brittle bone disease’. Current approaches mainly focus on bone mineral density measurements as an indirect indicator of bone strength and quality. It has been shown that bone strength would depend not only on composition but also structural organization. This study aims to characterize 3D structure of the cortical bone in high-resolution micro CT images. A total of 40 bone fragments from 28 subjects (13 with OI and 15 healthy controls) were imaged using micro tomography using a synchrotron light source (SRµCT). Minkowski functionals - volume, surface, curvature, and Euler characteristics - describing the topological organization of the bone were computed from the images. The features were used in a machine learning task to classify between healthy and OI bone. The best classification performance (mean AUC – 0.96) was achieved with a combined 4-dimensional feature of all Minkowski functionals. Individually, the best feature performance was seen using curvature (mean AUC - 0.85), which characterizes the edges within a binary object. These results show that quantitative analysis of cortical bone microstructure, in a computer-aided diagnostics framework, can be used to distinguish between healthy and OI bone with high accuracy.

[1]  Christian Kroos,et al.  Analysis of tongue configuration in multi-speaker, multi-volume MRI data , 2000 .

[2]  A. Meyer-Base,et al.  Stability analysis of a self-organizing neural network with feedforward and feedback dynamics , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[3]  Peter Hastreiter,et al.  Bildverarbeitung für die Medizin 2003 , 2003 .

[4]  Anke Meyer-Bäse,et al.  Segmentation and classification of dynamic breast magnetic resonance image data , 2006, J. Electronic Imaging.

[5]  Axel Wismüller,et al.  Prediction of Biomechanical Properties of Trabecular Bone in MR Images With Geometric Features and Support Vector Regression , 2011, IEEE Transactions on Biomedical Engineering.

[6]  Axel Wismüller,et al.  The Exploration Machine - A Novel Method for Data Visualization , 2009, WSOM.

[7]  Mahesh B. Nagarajan,et al.  Performance of topological texture features to classify fibrotic interstitial lung disease patterns. , 2011, Medical physics.

[8]  Michel Verleysen,et al.  Recent Advances in Nonlinear Dimensionality Reduction, Manifold and Topological Learning , 2010, ESANN.

[9]  Anke Meyer-Bäse,et al.  Fully automated biomedical image segmentation by self-organized model adaptation , 2004, Neural Networks.

[10]  A. A. MacDowell,et al.  X-ray micro-tomography at the Advanced Light Source , 2012, Optics & Photonics - Optical Engineering + Applications.

[11]  Thomas Villmann,et al.  Exploratory Observation Machine (XOM) with Kullback-Leibler Divergence for Dimensionality Reduction and Visualization , 2010, ESANN.

[12]  Thomas Villmann,et al.  Neighbor embedding XOM for dimension reduction and visualization , 2011, Neurocomputing.

[13]  M. Reiser,et al.  Classification of Small Contrast Enhancing Breast Lesions in Dynamic Magnetic Resonance Imaging Using a Combination of Morphological Criteria and Dynamic Analysis Based on Unsupervised Vector-Quantization , 2008, Investigative radiology.

[14]  T. Nattkemper,et al.  Breast MRI data analysis by LLE , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[15]  John Jameson,et al.  3D micron-scale imaging of the cortical bone canal network in human osteogenesis imperfecta (OI) , 2013, Medical Imaging.

[16]  Axel Wismüller,et al.  Classification of Small Lesions in Breast MRI: Evaluating The Role of Dynamically Extracted Texture Features Through Feature Selection. , 2013, Journal of medical and biological engineering.

[17]  Axel Wismüller,et al.  A Neural Network Approach to Functional MRI Pattern Analysis — Clustering of Time-Series by Hierarchical Vector Quantization , 1998 .

[18]  Axel Wismüller.  The exploration machine: a novel method for analyzing high-dimensional data in computer-aided diagnosis , 2009, Medical Imaging.

[19]  Anke Meyer-Bäse,et al.  Model-free functional MRI analysis based on unsupervised clustering , 2004 .

[20]  M.Kleinberg Jon,et al.  Advances in Self-Organizing Maps, 7th International Workshop, WSOM 2009, St. Augustine, FL, USA, June 8-10, 2009. Proceedings , 2009, WSOM.

[21]  Klaus Mecke,et al.  Utilizing Minkowski functionals for image analysis: a marching square algorithm , 2008 .

[22]  H. Ritter,et al.  The Deformable Feature Map — Adaptive Plasticity for Function Approximation , 1998 .

[23]  Axel Wismüller,et al.  Classification of interstitial lung disease patterns with topological texture features , 2010, Medical Imaging.

[24]  Thomas Martinetz,et al.  Medical image compression using topology-preserving neural networks , 2005, Eng. Appl. Artif. Intell..

[25]  Anke Meyer-Baese,et al.  Model-free functional MRI analysis using cluster-based methods , 2003, SPIE Defense + Commercial Sensing.

[26]  J. Gram,et al.  Bone geometry, density, and microarchitecture in the distal radius and tibia in adults with osteogenesis imperfecta type I assessed by high‐resolution pQCT , 2012, Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research.

[27]  Helge J. Ritter,et al.  The deformable feature map - a novel neurocomputing algorithm for adaptive plasticity in pattern analysis , 2002, Neurocomputing.

[28]  Axel Wismüller,et al.  Classification of small lesions in dynamic breast MRI: eliminating the need for precise lesion segmentation through spatio-temporal analysis of contrast enhancement , 2012, Machine Vision and Applications.

[29]  Axel Wismüller,et al.  Volumetric characterization of human patellar cartilage matrix on phase contrast x-ray computed tomography , 2015, Medical Imaging.

[30]  Anke Meyer-Bäse,et al.  Cluster analysis of dynamic cerebral contrast-enhanced perfusion MRI time-series , 2006, IEEE Transactions on Medical Imaging.

[31]  T. Twellmann,et al.  Detection of suspicious lesions in dynamic contrast enhanced MRI data , 2004, The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[32]  Phil Hoole,et al.  A Segmentation and Analysis Method for MRI Data of the Human Vocal Tract , 2003, Bildverarbeitung für die Medizin.

[33]  Axel Wismüller.  A Computational Framework for Nonlinear Dimensionality Reduction and Clustering , 2009, WSOM.

[34]  Axel Wismüller,et al.  Segmentation with neural networks , 2000 .

[35]  M. Reiser,et al.  Cluster analysis of signal-intensity time course in dynamic breast MRI: does unsupervised vector quantization help to evaluate small mammographic lesions? , 2006, European Radiology.

[36]  T. Keaveny,et al.  Evaluation of teriparatide treatment in adults with osteogenesis imperfecta. , 2014, The Journal of clinical investigation.