Statistical Shape Model: comparison between ICP and CPD algorithms on medical applications

Rocco Furferi,Lapo Governi,Yary Volpe,Carlo Robilotta,Antonio Marzola

doi:10.1007/s12008-020-00725-1

Rocco Furferi, Lapo Governi + Show 3 more

Open Access

https://doi.org/10.1007/s12008-020-00725-1

Copy DOI

Abstract

The increasing availability of 3D anatomical models obtained from diagnostic images exploiting Reverse Engineering techniques allows the application of statistical analysis in the quantitative investigation of anatomical shapes variability. Statistical Shape Models are a well-established method for representing such variability, especially for complex forms like the anatomical ones. Not by chance, these models are widely used for medical applications, such as guiding segmentation of the diagnostic image and virtual reconstruction of incomplete anatomic region. The application of a statistical analysis on a set of shapes representing the same anatomical region essentially requires that shapes must be in correspondence, i.e. constituted by the same number of points in corresponding position. This work aims to compare two established algorithms, namely a modified version of the Iterative Closest Point and the non-rigid version of the Coherent Point Drift, to solve the correspondences’ problem in the construction of a Statistical Shape Model of the human cranium. The comparison is carried out on the models using the standard evaluation criteria: Generalization, Specificity and Compactness. The modified version of the Iterative Closest Point delivers a better Statistical Shape Model in terms of Generalization and Specificity, but not for Compactness, than the Coherent Point Drift-based model.

Highlights

Reverse Engineering (RE) techniques are typically related to manufacturing industry applications, but, in the last few years, they are proving their effectiveness in nontraditional fields, as biology and anatomy
Since the values of Compactness are very similar and the value of G and S are lower for the Iterative Closest Point (ICP)-based Statistical Shape Models (SSMs) for all c, it’s possible to conclude that, in the case of human neurocrania, the modified ICP version provide a better SSM if compared to Coherent Point Drift (CPD) version
The modified version of ICP algorithm is preferred to the CPD as it required less time to register the template to each sample, to detect the correspondences among all the samples of the Training Set (TS) and to build the SSM

Summary

Introduction

Reverse Engineering (RE) techniques are typically related to manufacturing industry applications, but, in the last few years, they are proving their effectiveness in nontraditional fields, as biology and anatomy. Computer-Aided Technologies (CAx) software packages provide advanced tools to properly handle the so obtained 3D model allowing more effective preoperative simulation, complex-surgery planning, quantitative evaluation of asymmetry or dysmorphism and the design of the patient-specific devices. To this aim, the increased availability of real and interactive 3D anatomical models, allows the application of statistical analysis in the quantitative investigation and modelling of anatomical shapes variability. These variations are represented by the principal components φi (main modes of variation) and their respective variance values λi , defined by applying the Principal Component Analysis (PCA) on the TS

Objectives

Results

Conclusion