Spatial Markov Kernels for Image Categorization and Annotation.

This paper presents a novel discriminative stochastic method for image categorization and annotation. We first divide the images into blocks on a regular grid and then generate visual keywords through quantizing the features of image blocks. The traditional Markov chain model is generalized to capture 2-D spatial dependence between visual keywords by defining the notion of "past" as what we have observed in a row-wise raster scan. The proposed spatial Markov chain model can be trained via maximum-likelihood estimation and then be used directly for image categorization. Since this is completely a generative method, we can further improve it through developing new discriminative learning. Hence, spatial dependence between visual keywords is incorporated into kernels in two different ways, for use with a support vector machine in a discriminative approach to the image categorization problem. Moreover, a kernel combination is used to handle rotation and multiscale issues. Experiments on several image databases demonstrate that our spatial Markov kernel method for image categorization can achieve promising results. When applied to image annotation, which can be considered as a multilabel image categorization process, our method also outperforms state-of-the-art techniques.

Three-dimensional virtual-reality surgical planning and soft-tissue prediction for orthognathic surgery.

Complex maxillofacial malformations continue to present challenges in analysis and correction beyond modern technology. The purpose of this paper is to present a virtual-reality workbench for surgeons to perform virtual orthognathic surgical planning and soft-tissue prediction in three dimensions. A resulting surgical planning system, i.e., three-dimensional virtual-reality surgical-planning and soft-tissue prediction for orthognathic surgery, consists of four major stages: computed tomography (CT) data post-processing and reconstruction, three-dimensional (3-D) color facial soft-tissue model generation, virtual surgical planning and simulation, soft-tissue-change preoperative prediction. The surgical planning and simulation are based on a 3-D CT reconstructed bone model, whereas the soft-tissue prediction is based on color texture-mapped and individualized facial soft-tissue model. Our approach is able to provide a quantitative osteotomy-simulated bone model and prediction of postoperative appearance with photorealistic quality. The prediction appearance can be visualized from any arbitrary viewing point using a low-cost personal-computer-based system. This cost-effective solution can be easily adopted in any hospital for daily use.

Computer-assisted three-dimensional surgical planing and simulation. 3D soft tissue planning and prediction.

The purpose of this paper is to report a new technique for three-dimensional facial soft-tissue-change prediction after simulated orthognathic surgical planning. A scheme for soft tissue deformation, "Computer-assisted three-dimensional virtual reality soft tissue planning and prediction for orthognathic surgery (CASP)", is presented. The surgical planning was based on three-dimensional reconstructed CT visualization. Soft tissue changes were predicted by two newly devised algorithms: Surface Normal-based Model Deformation Algorithm and Ray Projection-based Model Deformation Algorithm. A three-dimensional color facial texture-mapping technique was also used for generating the color photo-realistic facial model. As a final result, a predicted and simulated patient's color facial model can be visualized from arbitrary viewing points.

Computer-assisted three-dimensional surgical planning and simulation: 3D virtual osteotomy.

A computer-assisted three-dimensional virtual osteotomy system for orthognathic surgery (CAVOS) is presented. The virtual reality workbench is used for surgical planning. The surgeon immerses in a virtual reality environment with stereo eyewear, holds a virtual "scalpel" (3D Mouse) and operates on a "real" patient (3D visualization) to obtain pre-surgical prediction (3D bony segment movements). Virtual surgery on a computer-generated 3D head model is simulated and can be visualized from any arbitrary viewing point in a personal computer system.

Three-dimensional virtual reality surgical planning and simulation workbench for orthognathic surgery.

A new integrated computer system, the 3-dimensional (3D) virtual reality surgical planning and simulation workbench for orthognathic surgery (VRSP), is presented. Five major functions are implemented in this system: post-processing and reconstruction of computed tomographic (CT) data, transformation of 3D unique coordinate system geometry, generation of 3D color facial soft tissue models, virtual surgical planning and simulation, and presurgical prediction of soft tissue changes. The basic mensuration functions, such as linear and spatial measurements, are also included. The surgical planning and simulation are based on 3D CT reconstructions, whereas soft tissue prediction is based on an individualized, texture-mapped, color facial soft tissue model. The surgeon "enters" the virtual operatory with virtual reality equipment, "holds" a virtual scalpel, and "operates" on a virtual patient to accomplish actual surgical planning, simulation of the surgical procedure, and prediction of soft tissue changes before surgery. As a final result, a quantitative osteotomy-simulated bone model and predicted color facial model with photorealistic quality can be visualized from any arbitrary viewing point in a personal computer system. This system can be installed in any hospital for daily use.

Computer analysis of 2-d electrophoresis gels : image analysis, data base, and graphic AIDS.

Although two-dimensional (2-D) gel electrophoresis has been used for the past ten years, a large amount of potentially useful information is still generally unused, since accurate and rapid methods for extracting, managing, and analyzing the vast quantity of data that can be derived from the gels are not widely available. The difficulty is partly caused by the large number of spots as well as artifacts present in a 2-D gel and the non-uniformity in the electrophoresis process that causes distorted spot patterns and makes comparison among gels difficult. With the advent of data base management systems (DBMS) and computer graphics, 2-D gel data can be organized and manipulated with relative ease. This chapter describes work that has been done toward automating the process of extracting useful quantitative information from 2-D gels and comparing spot data derived from separate gels. The design and implementation of a data base and a data analysis system for 2-D gels are also outlined.

A preliminary study of computer recognition and identification of skeletal landmarks as a new method of cephalometric analysis.

The usefulness of results obtained with cephalometrics is limited by measurement error. Much of this is due to the subjective nature of landmark identification on X-rays. Such subjectivity is not eliminated by electronic plotting equipment - despite its high resolution. Automatic (non-subjective) landmark identification of sella and menton was carried out in this study using the CLIP4 image processing system. The results obtained on 23 radiographs were compared with those of conventional techniques.