Elsevier

Computers in Industry

Volume 61, Issue 3, April 2010, Pages 187-197
Computers in Industry

STL rapid prototyping bio-CAD model for CT medical image segmentation

https://doi.org/10.1016/j.compind.2009.09.005Get rights and content

Abstract

This paper presents a simple process to construct 3D rapid prototyping (RP) physical models for computer tomography (CT) medical images segmentation. The use of stereolithography (STL) triangular meshes as a basis for RP construction facilitates the simplification of the process of converting CT images to an RP model. This is achieved by constructing the STL triangular meshes directly from data points without having to draw the curve model first. The grey prediction algorithm is used to sort contour point data in each layer of the medical image. The contour difference detection operation is used to sequence the points for each layer. The 3D STL meshes are then constructed by this proposed layer-by-layer sequence meshes algorithm to build the STL file. Once this STL file is saved, a 3D physical model of the medical image can be fabricated by RP manufacturing, and its virtual reality model can also be presented for visualization. CT images of a human skull and femur bone were used as the case studies for the construction of the 3D solid model with medical images. The STL models generated using this new methodology were compared to commercial computer-aided design (CAD) models. The results of this comparative analysis show that this new methodology is statistically comparable to that of the CAD software. The results of this research are therefore clinically reliable in reconstructing 3D bio-CAD models for CT medical images.

Introduction

Bio-medical engineering is a technological field with great potential for future advances. This field encompasses medical treatment engineering, tissue engineering, genetic technology, and medicine engineering [1], [2]. Building sets of medical information, medical images, and bio-medical materials and applying these sets of medical data assists in the development of all aspects of bio-medical engineering. In recent years, CAD has been increasingly applied in bio-medical design. The integration of CAD and medical technology is referred to as bio-CAD [3], [4], [5], [6]. Bio-CAD includes regenerative medicine engineering, computer-aided surgery [7], structural modeling of tissue and tissue informatics [8], design of orthopedic devices and implants, design of tissue scaffolds [9], reverse engineering (RE) and 3D reconstruction [10], [11], heterogeneous tissue modeling [12], and solid freeform fabrication or bio-manufacturing [13], [14].

CT medical imaging is the key tool for viewing the internal structure of the human body, but is limited by its 2D image presentation in that it does not allow doctors to quickly diagnose illnesses and explain symptoms and treatments to patients [15]. Medical images in 3D solid models are therefore very important in the diagnosis and treatment process. All reconstructed 3D solid models can be converted to RP physical models and Virtual Reality Modeling Language (VRML) format for visualization [16], [17], [18], [19]. A number of open sources and commercial products for 3D bio-mechanical construction are available, but there still does not appear to be a simple and accurate geometric method for bio-image acquisition and analysis available [20], [21].

There are three methods that can be applied to reconstruct a 3D solid model for bio-medical imaging from its 2D CT image. The first method involves a swept blend from the contours of each layer in point data [22], [23]. The second method is via voxel to stack and construct the model by using the marching-cube algorithm [24], [25]. With the third method, contour detection in each layer is used to construct the mixed layers in the triangular STL model for RP fabrication by connecting the vertices of two parallel polygons [26], [27], [28]. Each of these methods has disadvantages. With the first method, the curves swept blend method is extremely complicated, and it cannot be applied without drawing the curve model, since the spline must first be constructed before modeling can take place [29], [30]. In the second method, the voxel stacking technique may make holes, aliases, and saw-toothed paths within the voxel connection [31], [32]. The third method suffers from drawbacks in the contour detection for each CT layer and file errors in the construction and use of STL meshes [33], [34], [35], [36], [37].

In the research project undertaken here, we proposed a simple but robust process for converting a CT medical image to a bio-CAD RP physical model, to improve upon the drawbacks of the previous method and to construct a 3D bio-medical image accurately [38]. The CT image is detected using the grey prediction algorithm and subsequent conversion of the data into point data within a layer. A layer-by-layer sequence meshes algorithm for STL file reconstruction is proposed. This new algorithm focuses on the exploration of STL and use of the mesh sequences of adjacent layers as the basis for 3D reconstruction [39], [40]. This integrated method formats STL directly from point data to convert to a 3D model. This does not require the transformation of data points to spline curves, thereby effectively simplifying the reconstruction of 3D medical images. Therefore, the primary purpose of this research was to describe the reconstruction of a 2D bio-medical image into a 3D medical prototype for visualization.

This was achieved by the four processes presented in this paper. The first process describes image processing in the grey prediction algorithm. The second process involves the reconstruction of the 3D solid model by contouring difference detection operation and layer-by-layer sequence meshes. The third process compares the STL file with the swept blend model using existing CAD software and analyzes the resulting deviation. The last process demonstrates the RP model and virtual reality (VR) displays for visualization, as shown in Fig. 1.

The process of reconstructing the medical 3D solid model can be divided into medical image processing and reconstruction of STL. Medical image processing is achieved through grey prediction, which serves to simplify the process of image recognition. During the process of recognition, the point data obtained are disordered. Therefore, rearrangement of the point data from the contour difference detection algorithm was proposed. After rearrangement, STL formats can be processed directly from the point data. Reconstruction of STL was achieved through layer-by-layer mesh sequencing to process the point cloud from each layer. After the reconstruction of the 3D solid model was completed, deviations between the constructed model and the commercial CAD-swept spline model were compared and analyzed. Finally, 3D models were displayed through the RP physical model and VRML format.

Section snippets

Grey prediction in the CT image

Points data are converted from CT images through grey prediction modeling. The medical image is composed of pixels, and each pixel with a grey level between 0 (black) to 255 (white) corresponds to one point on the image. The function f(x, y) is for the grey value, in which (x, y) signifies the coordinate of the image. Grey prediction is often used to forecast and analyze future data from current and past data [41], [42]. The main color of the medical image is black, which denotes null. During

STL meshes reconstruction

Pre-processing of image data must take place before obtaining point data information in the reverse engineering process. After point data have been obtained, post-processing is carried out. This involves the construction of the 3D model, editing, and modeling. In this research project, STL mesh construction was separated into the following steps: (1) obtaining points cloud data of each layer; (2) sorting point data of each layer; (3) analyzing layers; (4) constructing STL mesh; and (5)

Case study and analysis

In this research, we proposed using dynamic grey prediction, contour difference calculation, and layer-by-layer sequence meshes for STL construction between layers to create 3D bio-medical images. There are three main steps involved in the reconstruction process: (1) image recognition; (2) point-cloud arrangement; and (3) mesh construction.

Conclusion

The development of bio-medical imaging is a great advancement in the medical field, because it reduces the rate of medical misdiagnosis of illnesses. Although bio-medical images improved the ability of doctors to diagnose and treat patients, it was still a laborious process to decipher pertinent information from 2D CT images. The successful transformation of 2D bio-medical images into 3D models was the main goal of this research.

In this research, we proposed grey prediction theory, contour

Chung-Shing Wang is an associate professor in the Department of Industrial Design, Tunghai University, Taiwan. He got his PhD degree from Department of Mechanical Engineering, Imperial College, London in 1991. His research interests include computer geometric modeling, CAD/CAM, rapid prototyping, reverse engineering, product development methodology, modular design and bio-medical engineering.

References (43)

  • C.M. Cheah et al.

    Integration of laser surface digitising with computer-aided design and manufacturing for developing facial prostheses—Part 1: design and fabrication of prostheses replicas

    International Journal of Prosthodontics

    (2003)
  • C.M. Cheah et al.

    Integration of laser surface digitising with computer-aided design and manufacturing for developing facial prostheses—Part 2: development of moulding techniques for casting prosthetic parts

    International Journal of Prosthodontics

    (2003)
  • T. McInereny et al.

    Topology adaptive deformable surfaces for medical image volume segmentation

    IEEE Transactions on Medical Image

    (1999)
  • C.K. Chua et al.

    Rapid prototyping assisted surgery planning

    International Journal of Advanced Manufacturing Technology

    (1998)
  • C.K. Chua et al.

    Development of a tissue engineering scaffold structure library for rapid prototyping. Part 1. Investigation and classification

    International Journal of Advanced Manufacturing Technology

    (2003)
  • K.D. Creehan et al.

    Computer-aided reverse engineering of the human musculoskeletal system

    Virtual and Physical Prototyping

    (2006)
  • S. Swann

    Integration of MRI and stereolithography to build medical models: a case study

    Rapid Prototyping Journal

    (1996)
  • Y. Yan et al.

    Biomaterial forming research using RP technology

    Rapid Prototyping Journal

    (2003)
  • B. Sanghera et al.

    Preliminary study of rapid prototype medical models

    Rapid Prototyping Journal

    (2001)
  • B. Preim et al.

    Visualization in Medicine

    (2007)
  • S. Gopakumar

    RP in medicine: a case study in cranial reconstructive surgery

    Rapid Prototyping Journal

    (2004)
  • Cited by (0)

    Chung-Shing Wang is an associate professor in the Department of Industrial Design, Tunghai University, Taiwan. He got his PhD degree from Department of Mechanical Engineering, Imperial College, London in 1991. His research interests include computer geometric modeling, CAD/CAM, rapid prototyping, reverse engineering, product development methodology, modular design and bio-medical engineering.

    Wei-Hua Andrew Wang is an associate professor in the Department of Industrial Engineering and Enterprise Information, Tunghai University, Taiwan. He got his PhD degree from the Department of Industrial and Systems Engineering, The Ohio State University USA in 1989. His research interests include CAD/CAM, Network Design, Bayesian Networks, Artificial Neural Networks and systems dynamics.

    Man-Ching Lin is pursuing her PhD degree in the Department of Industrial Engineering and Enterprise Information, Tunghai University. She received her Master degree in Industrial Engineering and Systems Management from Feng Chia University in 2002. Her research interests include systems engineering, reverse engineering, CAD/CAM and technology management.

    View full text