Chanjira Sinthanayothin

Feasibility study on computer-aided screening for diabetic retinopathy.

PurposeTo conduct a feasibility study of computer-aided screening for diabetic retinopathy by developing a computerized program to automatically detect retinal changes from digital retinal images.MethodsThe study was carried out in three steps. Step 1 was to collect baseline retinal image data of 600 eyes of normal subjects with normal fundi and data of 300 eyes of diabetic patients with diabetic retinopathy. All data were recorded by digital fundus camera. Step 2 was to analyse all retinal images for normal and abnormal features. By this method, the automated computerized screening program was developed. The program preprocesses colour retinal images and recognizes the main retinal components (optic disc, fovea, and blood vessels) and diabetic features such as exudates, haemorrhages, and microaneurysms. All of the accumulated information is interpreted as normal, abnormal, or unknown. Step 3 was to evaluate the sensitivity and specificity of the computerized screening program by testing the program on diabetic patients and comparing the programs results with the results of screening by retinal specialists.ResultsDiabetic patients (182 patients, 336 eyes) were examined by retinal specialists; 221 eyes had a normal fundus and 115 eyes had nonproliferative diabetic retinopathy. Digital retinal images were taken of these 336 eyes and interpreted by the automated screening program. The program had a sensitivity and specificity of 74.8% and 82.7%, respectively.ConclusionsThe automated screening program was able to differentiate between the normal fundus and the diabetic retinopathy fundus. The program may be beneficial for use in screening for diabetic retinopathy. Further development of the program may provide higher sensitivity. Jpn J Ophthalmol 2006;50:361–366

Automatic retinal vessel tortuosity measurement using curvature of improved chain code

Measurement of blood vessel tortuosity is a useful capability for automatic ophthalmological diagnostic tools. Screening of Retinopathy of Prematurity (ROP), a disease of eye that affects premature infants, for example, depends crucially on automatic tortuosity evaluation. Quite a few techniques for tortuosity measurement and classification have been proposed, but they do not always match the clinical concept of tortuosity. In this paper, we propose the alternative method of automatic tortuosity measurement for retinal blood vessels that uses the curvature calculated from improved chain code algorithm taking the number of inflection point into account. The tortuosity calculated from the proposed method is independent of the segmentation of vessel tree. Our algorithm can automatically classify the image as tortuous or non-tortuous. The test results are verified against two expert ophthalmologists. For an optimal set of training parameters the prediction is as high as 100% on 18 images.

Computerized algorithm for 3D teeth segmentation

This paper presents the computerized segmentation algorithm on the 3D digital dental model for separating individual tooth using Borland C++ Builder® and OpenGL. The purpose is to apply the algorithm in the 3D orthodontic diagnosis and treatment software package. The method consists of 5 main steps as follows. First the digital dental model is produced by 3D scanner in STL format and the occlusal plane is calculated. Second the arch form of the model is approximated. Third the panoramic of the dental cast in 3D system is constructed. Fourth, the 3D-panoramic is converted to 2D system to find the cutting points for segmentation of individual tooth, and finally the 2D image is converted back to 3D system for creating the vertical plane for separating teeth into difference colors at the cutting points. The result has shown that the 3D digital dental model can be segmented for separating individual tooth, and the algorithm is ready to apply in orthodontic diagnosis and treatment software package.

Orthodontics treatment simulation by teeth segmentation and setup

This paper presents the development of three-dimensional (3D) computer aided orthodontic software using Borland C++ Builderreg and OpenGL with the purpose of creating the 3D orthodontic treatment simulation. The method consists of 4 main steps as follows: the dental cast produced by 3D scanner in STL format was transformed from un-indexed data to an indexed data and taken as input in the first step. The cast model can be displayed in various views while user can move the model to any directions as required. Second step is the 3D teeth segmentation which has 2 methods available: segmentation using landmarks and segmentation using planes. Third step is to rearrange the segmented teeth to align with standard arch form curves using translation and rotation tools developed to let the user interaction with ease. Finally, the simulation of the orthodontics treatment has been developed, which display the alignment of each tooth from the beginning to the final outcome. The result has shown that the 3D orthodontic treatment simulation can be shown as a guideline of treatment preview for the patients. It is also a training tool for orthodontists, residents and students to enhance the experience through simulation before taking actual orthodontic planning.

Point-cloud-to-point-cloud technique on tool calibration for dental implant surgical path tracking

Dental implant is one of the most popular methods of tooth root replacement used in prosthetic dentistry. Computerize navigation system on a pre-surgical plan is offered to minimize potential risk of damage to critical anatomic structures of patients. Dental tool tip calibrating is basically an important procedure of intraoperative surgery to determine the relation between the hand-piece tool tip and hand-pieces markers. With the transferring coordinates from preoperative CT data to reality, this parameter is a part of components in typical registration problem. It is a part of navigation system which will be developed for further integration. A high accuracy is required, and this relation is arranged by point-cloud-to-point-cloud rigid transformations and singular value decomposition (SVD) for minimizing rigid registration errors. In earlier studies, commercial surgical navigation systems from, such as, BrainLAB and Materialize, have flexibility problem on tool tip calibration. Their systems either require a special tool tip calibration device or are unable to change the different tool. The proposed procedure is to use the pointing device or hand-piece to touch on the pivot and the transformation matrix. This matrix is calculated every time when it moves to the new position while the tool tip stays at the same point. The experiment acquired on the information of tracking device, image acquisition and image processing algorithms. The key success is that point-to-point-cloud requires only 3 post images of tool to be able to converge to the minimum errors 0.77%, and the obtained result is correct in using the tool holder to track the path simulation line displayed in graphic animation.

Toward robot-assisted dental surgery: Path generation and navigation system using optical tracking approach

The main aim of this paper is to develop dental implant surgical navigation system based on homogenous transformation algorithms. This work is a partial section of robot-assisted surgical development. The previous works are presented in numerous basic research. They are methodology design on tool tip calibration, optical marker recognition, and pose determination using neural networks. This paper concerns with tracking path generation system based on fundamental of optical tracking. The intraoperative system is the principal focus area of this study. The homogenous transformation has been calculated in term of kinematics equation among marker relationship. The stereo camera is utilized to retrieve 3D position of different pattern styles of markers. The beneath marker recognition algorithm using rotation-invariant neural network and physical method is performed to identify markers. The fundamental relationship among markers are computed to obtain the orientation and translation between the guided path and the instruments tool tip. The experiment has been demonstrated and performed under prototype model. The method is to work on procedure step by step. They begin with patient information input and continuously perform on marker recognition, tool tip calibrations and marker digitization. The path tracking is executed to observe the accuracy of the system. The result shows that the system can be performed to track path based on beforehand planning.

Digital Dental Model Analysis

This paper is to introduce dental model analysis software with facilitating functions to allow direct dental measurement along with simulation after dental alignment and extraction. Model analysis is one of the orthodontic analyses to assess dental alignment and dental space whether the teeth are at the appropriated positions and space for orthodontic treatment or dental extraction is in need. There are several model analysis researches done by local researchers done by tedious and time-consuming manual means while there are a few cases model analyses by using dental software but most using imported and expensive orthodontic software. There are very few local researches on the software to perform dental model analysis but none of them could perform direct dental measurement and dental simulation after dental alignment and extraction. This model analysis software could perform dental width and dental arches measurements which results have been applied for Bolton analysis, which calculate the dental ratios. User could perform dental model simulation after selecting individual teeth appeared on the dental simulation image. Further developments of this model analysis program will allow the measurement of dental curve widths and curve heights along with other measurements relating to dental model analyses.

Wavelet transformation for dental X-ray radiographs segmentation technique

Teeth segmentation of dental X-ray radiographs is the important step in teeth recognition. A segmentation to partition teeth from other areas in dental X-ray images. This paper proposes wavelet transform (WT) to segment dental X-ray images. Wavelet transform are small waves located in different time that analyze according to scale. Indeed, they are obtained using scaling and translation of a scaling function and wavelet function. The result of segmentation is a teeth feature marking each pixel. Such as, edges detection from image segmented.

ZBaby: Android application for pregnancy due date, fetus development simulation and weight gain during pregnancy

For awareness of becoming parents, this paper proposes the development of Z-Baby as an android application to calculate the pregnancy due date, fetus development simulation, weight gain during pregnancy and the amount of calorie in needs. Pregnancy due date is calculated based on user defined data which is the last menstruation. Next, user can view the simulation of 3D fetus development with description week by week using OpenGL ES. Then the weight gain during pregnancy is calculated from the BMI based on the specification by IOM 2009 while allowing a user to manage her weight gain as well. Updated weight data were plotted dynamically using Android Plot API. Daily calorie requirement can be calculated during pregnancy by Harris-Benedict Equation to maintain the appropriated weight for healthy baby without overeating. Other available functions are recording photos & data in monthly basis, locating moms position on Google map to find the nearest hospital located on the map. For accuracy testing, Z-Baby has been compared to other software packages which have the same functions. Z-Baby in now available for free in Google Play.

Vessel extraction in retinal images using multilevel line detection

This paper presents an algorithm to segment the blood vessels in retinal images. The contribution of this work is that we exhibit how a line detection can be applied in a multi-scale framework to fulfill the sophisticated task of vessel segmentation. The key idea of our algorithm is to systematically combine the outputs of Gaussian based line detection that is applied sequentially to the input image at several scales from coarse to fine levels. This brings down to the strategy that the evidences of larger size vessels are used to steer the process of addition of smaller vessels. Our proposed method is evaluated on the public DRIVE database and shows results with false positive rate.