Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Duangrat Gansawat has a total of 50 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Quality and Usability of Mobile Fundus Camera for Schematic Eyes: A Comparative Analysis (Preprint)

    by Supaporn Tengtrisorn, Somyot Chirasatitsin, Prawit Kaeonarong, Duangrat Gansawat, Sitthichok Chaichulee, Virasakdi Chongsuvivatwong

    BACKGROUND Photographs or videos taken from a mobile phone with a +20-D lens can be an alternative to indirect ophthalmoscopy in mass screening and telemedicine. OBJECTIVE To develop a mobile fundus camera (MFC) that ensures eye screening with safety, ease of use, and high accuracy in detecting fundus lesions. METHODS From October 1st, 2021, to September 30th, 2022, we developed an MFC based on Samsung S21 and iPhone 12 mobile phones with four different types of connecting lenses. Ten ophthalmology personnel were trained to use the MFC on schematic eyes and evaluate the developed systems. The four schematic eyes had a normal fundus, diabetic retinopathy (DR), normal optic disc, and glaucomatous optic disc. The system safety was assessed based on the ISO 15004-2 Visible wavelength (400–700 nm). All users completed the Thai version of the Usability Experience Questionnaire (UEQ). The evaluation of image quality was based on the sharpness index (SI), disc-to-image ratio, and color discrimination, which included relative redness, red-green difference, and red-blue difference. RESULTS Our system satisfied the safety criteria. The mean UEQ scores for the overall MFCs were above average in the domains of attractiveness, dependability, stimulation, and novelty. However, the efficiency and perspicuity were below average. The MFC based on the Samsung S21 had a significantly better overall SI but poorer overall color discrimination power than that of the iPhone 12 (p < 0.05). The best SI was achieved using Samsung S21+Volk 28D, whereas iPhone 12+oDocs 20D provided the best color discrimination. Interparticipant differences were not statistically significant. CONCLUSIONS All the developed MFCs tested were safe and usable. Based on the SI criteria, the Samsung S21+ Volk 28D should be used for further development.

  • Open Access

    Simulation of 3D Body Shapes for Pregnant and Postpartum Women

    by Chanjira Sinthanayothin, Piyanut Xuto, Wisarut Bholsithi, Duangrat Gansawat, Nonlapas Wongwaen, Nantaporn Ratisoontorn, Parut Bunporn, Supiya Charoensiriwath

    Several studies have reported that pre-pregnant women’s body mass index (BMI) affects women’s weight gain with complications during pregnancy and the postpartum weight retention. It is important to control the BMI before, during and after pregnancy. Our objectives are to develop a technique that can compute and visualize 3D body shapes of women during pregnancy and postpartum in various gestational ages, BMI, and postpartum durations. Body changes data from 98 pregnant and 83 postpartum women were collected, tracked for six months, and analyzed to create 3D model shapes. This study allows users to simulate their 3D body shapes in real-time and online, based on weight, height, and gestational age, using multiple linear regression and morphing techniques. To evaluate the results, precision tests were performed on simulated 3D pregnant and postpartum women’s shapes. Additionally, a satisfaction test on the application was conducted on new 149 mothers. The accuracy of the simulation was tested on 75 pregnant and 74 postpartum volunteers in terms of relationships between statistical calculation, simulated 3D models and actual tape measurement of chest, waist, hip, and inseam. Our results can predict accurately the body proportions of pregnant and postpartum women.


  • Robotic Process Automation-Based Glaucoma Screening System: A Framework

    by Somying Thainimit, Panaree Chaipayom, Duangrat Gansawat, Hirohiko Kaneko


  • Simulation of three-dimensional female body shapes with proportional representation for various weights and heights

    by Chanjira Sinthanayothin, Wisarut Bholsithi, Duangrat Gansawat, Nonlapas Wongwaen, Piyanut Xuto, Nantaporn Ratisoontorn, Parut Bunporn, Supiya Charoensiriwath

    Obesity is a significant factor in health information and increases the risk of health problems. Hence, an application that can help users to monitor their body mass index (BMI) timelines is needed. The simulation of a personalized 3D body shape may encourage women to control their BMI for a healthy body and pleasant appearance. Therefore, this paper aims to develop computerized 3D models of female shapes for various weights and heights, and consists of three important parts. First, the preparation of six avatars is described. Second, the body proportions of 6767 female datasets are analyzed to find the relationship of variables in various weights and heights. Last, 3D morphing of different female shapes is developed and analyzed experimentally for appropriate morphing parameters. Accuracy tests are carried out in three ways. First, body proportions calculated using the Z-Size Ladies application, called “Z-Size calculations,” are compared with the body proportions of data obtained from 3D scanners. Second, the Z-Size calculations are compared with tape measurements. Last, the Z-Size calculations are compared with measurements of Z-Size 3D morph models. The results of accuracy tests are shown as the relationship graphs between the BMI and body proportion measurements of chest, waist, hip, and inseam. Bland–Altman plots and Pearson correlation calculation show high correlation. In conclusion, the data obtained from the Z-Size calculations, 3D Scanner, tape measurements, and Z-Size morph models’ measurements are in good agreement and are highly correlated. The simulation of 3D female shapes for different weights and heights as proposed shows good performance and high accuracy.


  • Automatic Symmetrizing Mannequin Construction from 3D Body Scanner

    by Nonlapas Wongwaen, Duangrat Gansawat, Wisarut Bholsithi, Supiya Charoensiriwath, Apatha Peethong, Parut Bunporn, Chanjira Sinthanayothin

    In this paper, the mannequin for each individual person is constructed using information obtained from 3D body scanner. The collected measurements data are standardized for SizeThailand body shape sizing. The created mannequin represents each size of Thai women body shapes. The problems are that these collected 3D data of these sizes are asymmetry and have un-pleasant rough surface. In practical application, asymmetry body adjustment is solved manually but time-consuming and cumber-some. Hence, we explore the simple and automatically methods that are able to solve these problems and easily implemented. Our methods consist of 6 steps as follows: data rearrangement, symmetric plane construction, the template side selection, symmetric torso creation, torso trimming and mannequin surface smoothing. The 30 women samples data were tested by measuring chests, waists and hips of created mannequins comparing with the corresponding original 3D model. Processing time for each sample is approximately 894 milliseconds. The results show that the created mannequins have the accuracy of 98–99% with the SD of 0.5–1.1 for measurements of the chests, waists and hips. In overall, the sizes of 30 created mannequins are compared with their original 3D scanned sizes categorized based on SizeThailand size chart and the accuracy is calculated to be 90%. There are errors found in 3 of 30 mannequins as they are ± 1 size from their original sizes.

  • RGB-D classification using feature selection on the combination of different features

    by Wasif Khan, Ekachai Phaisangittisagul, Luqman Ali, Duangrat Gansawat, Itsuo Kumazawa

    To improve the classification performance in object recognition, many different features have been created from various techniques. However, these features may be redundant or irrelevant to implement the classification model. So, feature selection is considered one of the important tools that can be employed to select a subset of relevant features for specific model. In this paper, we applied feature selection method to form a new subset of features obtained from different feature extraction algorithms so that we can improve the object's classification performance. In the experiments, we demonstrated the use of filter method and wrapper methods for RGB-D object dataset and the results were promising.

  • Combining features for RGB-D object recognition

    by Wasif Khan, Ekachai Phaisangittisagul, Luqman Ali, Duangrat Gansawat, Itsuo Kumazawa

    Object category and instance recognition have received much attention in this era of modern technologies. Advanced image sensing technologies provide high resolution color and depth synchronized videos such as RGB-D (Kinect style) camera. At present, various features extraction schemes are introduced to improve classification performance. Extracting useful features from both color and depth images have recently gained much attention in this research area. In this paper, we proposed an approach to create new features using a feature combination of various feature extraction techniques: color autrocorrelogram, wavelet moments, local binary pattern (LBP) and principal component analysis (PCA) for RGB-D data. The experiments on benchmark dataset shows that the new features obtained by the proposed method using k-nearest neighbor (k-NN) classifier provide promising classification results.


  • Computer-aided analysis for breast cancer detection in thermography

    by Hossein Ghobadi, Somying Thainimit, Duangrat Gansawat, Nobuhiko Sugino

    The purpose of this study is to develop a texture-based algorithm for breast cancer classification in Thermography and performance evaluation for the purposed method. The aim of screening a disease is early detection in order to decrease the rate of mortality. CAD systems can be reckonable diagnostic tools for this importance and Infra-red imaging is expressed as a portable, non-invasive, non-contact and radiation-free diagnostic tool among them.


  • Open Access

    Interactive Three-Dimensional Model Reconstruction of Ultrasonography

    by Kritpanu Sopabutr, Paramate Horkaew, Duangrat Gansawat

    Interactive Three-Dimensional Model Reconstruction of Ultrasonography

  • Prediction of shape diameter undergoing coil embolization of saccular intracranial aneurysm treatment using a hybrid decision support system

    by Wiwat Owasirikul, Jaturon Tantivatana, Duangrat Gansawat, Supatana Auethavekiat

    The purpose of the study was to design a hybrid decision support system (HDSS) that could simulate the embolized coil selection pattern of the radiologists in aneurysms treatment. As the longest available length of the coils should be used in most cases, therefore only the shape diameter (SD) selection was modeled and varied. Ninety-eight aneurysms successfully treated by a radiologist with coil embolization were divided into two groups (86 for training and 12 randomly selected for validating). Eight aneurysms treated by another radiologist were also used to cross validate the proposed HDSS. The HDSS was developed using the classification and the linear regression methods (LRM). The dome and the width of an aneurysm were used as the system inputs. The system outputs were the SDs of the first three coils indexed according to the insertion order. The HDSS that consisted of Bagging classification and LRM achieved the highest accuracy for all cases. The errors were within 1 mm for the SD selection of the first two coils. For the third coil, the SD selection within 1 mm bound had 80 % accuracy. The experimental results indicated the feasibility of using the HDSS as the guidance for selecting the SDs of the first two coils. The selection of the third coil required more training data for the rarely used SD. Moreover, the cross validation with another radiologist showed the feasibility of using the proposed HDSS as the guidance, however further validation with more data is recommended.