Jyh-Wen Chai

Taichung Veterans General Hospital, 臺中市, Taiwan, Taiwan

Are you Jyh-Wen Chai?

Claim your profile

Publications (21)27.69 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: A new TRIO algorithm method integrating three different algorithms is proposed to perform brain MRI segmentation in the native coordinate space, with no need of transformation to a standard coordinate space or the probability maps for segmentation. The method is a simple voxel-based algorithm, derived from multispectral remote sensing techniques, and only requires minimal operator input to depict GM, WM, and CSF tissue clusters to complete classification of a 3D high-resolution multislice-multispectral MRI data. Results showed very high accuracy and reproducibility in classification of GM, WM, and CSF in multislice-multispectral synthetic MRI data. The similarity indexes, expressing overlap between classification results and the ground truth, were 0.951, 0.962, and 0.956 for GM, WM, and CSF classifications in the image data with 3% noise level and 0% non-uniformity intensity. The method particularly allows for classification of CSF with 0.994, 0.961 and 0.996 of accuracy, sensitivity and specificity in images data with 3% noise level and 0% non-uniformity intensity, which had seldom performed well in previous studies. As for clinical MRI data, the quantitative data of brain tissue volumes aligned closely with the brain morphometrics in three different study groups of young adults, elderly volunteers, and dementia patients. The results also showed very low rates of the intra- and extra-operator variability in measurements of the absolute volumes and volume fractions of cerebral GM, WM, and CSF in three different study groups. The mean coefficients of variation of GM, WM, and CSF volume measurements were in the range of 0.03% to 0.30% of intra-operator measurements and 0.06% to 0.45% of inter-operator measurements. In conclusion, the TRIO algorithm exhibits a remarkable ability in robust classification of multislice-multispectral brain MR images, which would be potentially applicable for clinical brain volumetric analysis and explicitly promising in cross-sectional and longitudinal studies of different subject groups.
    PLoS ONE 01/2015; 10(2):e0115527. · 3.53 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A computer-aided diagnostic system for colonoscopic imaging has been developed to classify colorectal polyps by type. The modules of the proposed system include image enhancement, feature extraction, feature selection and polyp classification. Three hundred sixty-five images (214 with hyperplastic polyps and 151 with adenomatous polyps) were collected from a branch of a medical center in central Taiwan. The raw images were enhanced by the principal component transform (PCT). The features of texture analysis, spatial domain and spectral domain were extracted from the first component of the PCT. Sequential forward selection (SFS) and sequential floating forward selection (SFFS) were used to select the input feature vectors for classification. Support vector machines (SVMs) were employed to classify the colorectal polyps by type. The classification performance was measured by the Az values of the Receiver Operating Characteristic curve. For all 180 features used as input vectors, the test data set yielded Az values of 88.7%. The Az value was increased by 2.6% (from 88.7% to 91.3%) and 4.4% (from 88.7% to 93.1%) for the features selected by the SFS and the SFFS, respectively. The SFS and the SFFS reduced the dimension of the input vector by 57.2% and 73.8%, respectively. The SFFS outperformed the SFS in both the reduction of the dimension of the feature vector and the classification performance. When the colonoscopic images were visually inspected by experienced physicians, the accuracy of detecting polyps by types was around 85%. The accuracy of the SFFS with the SVM classifier reached 96%. The classification performance of the proposed system outperformed the conventional visual inspection approach. Therefore, the proposed computer-aided system could be used to improve the quality of colorectal polyp diagnosis.
    Computerized medical imaging and graphics: the official journal of the Computerized Medical Imaging Society 06/2014; · 1.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Magnetic resonance (MR) image analysis is generally performed by spatial domainbased image processing, referred to as inter-pixel image processing, which takes advantage of spatial correlation among sample pixels. Unfortunately, in many areas, several tissue substances are usually present and mixed in a single image pixel in which such an inter-pixel processing either fails or is ineffective. To resolve this dilemma, this paper develops an unconventional approach, called intra-pixel processing, which considers MR images as multispectral images where a multispectral MR image pixel is actually a pixel vector, of which each component is captured by a particular image pulse sequence used for MR image acquisition. Since the commonly used receiver operating characteristic (ROC) curves cannot directly deal with the issues arising in intra-pixel processing, a 3D ROC analysis is developed by including a parameter t as the third dimension that represents abundance fractions thresholded by τ.
    International Journal of Computational Science and Engineering 01/2013; 8(2):87 - 110.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Focal nodular hyperplasia is the second most common benign hepatic tumor. Unlike adenoma as well as the malignant neoplasms, focal nodular hyperplasia can often be managed successfully without surgery. Use of liver-specific contrast-enhanced magnetic resonance imaging allows clinicians to confirm the diagnosis noninvasively in some patients, allowing select patients to avoid surgery. We report a case of a patient who presented with the rare profile of multiple, confluent lesions that were diagnosed, using magnetic resonance imaging with gadolinium-dimeglumine, as focal nodular hyperplasia. This complicated case was managed successfully and noninvasively based on algorithm found in the recent literature that allows patients to avoid unnecessary surgery.
    Journal of the Chinese Medical Association 07/2012; 75(7):355-8. · 0.75 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To solve the problem of the basal descent movement in quantification of the regional left ventricular (LV) myocardial wall thickness (WTh) and wall thickening (%WT) in short-axis (SA) cine MRI for effectively assessing the regional wall motion of LV myocardium. LV long-axis tagged MRI and SA cine MRI were performed to calculate the longitudinal translation and circumferential WTh of LV myocardium in eight normal volunteers. The new SA end-systolic thickness (EST) data were reconstructed from the original EST data, based on the quantified longitudinal translation of LV myocardium. The mean %WT of six segments in the basal section after correction was significantly different from that before correction in both intra- and inter-operator experiments. The polar map also showed the significant improvement of the variability of regional %WT and lack of quantification of %WT in the most basal SA slices after correction. The proposed technique demonstrated an important advantage to calculate the %WT in the most basal SA myocardial tissue, which was considered difficult to be achieved using cine MRI.
    Journal of Magnetic Resonance Imaging 02/2011; 33(2):464-73. · 2.57 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The paper proposed a weighted Radial basis function kernel (WRBF) approach that can be used to detect and classify anomalies in Magnetic Resonance (MR) images. A weighted Radial basis function kernel (WRBF) approach, despite the fact that the idea of WRBF kernels can be traced back to the work (1), its application to Radial basis function (RBF) kernel is new. It includes the Support Vector Machines (SVMs) using RBF as its special case where the RBF is considered to be uniformly weighted. Methods MR data of abnormal brain data were used to evaluate the accuracy of multiple sclerosis lesions classification by using the proposed method. The data were obtained from the BrainWeb Simulated Brain Database at the McConnell Brain Imaging Centre of the Montreal Neurological Institute (MNI), McGill University. Experimental results via various MR images show that WRBF kernels provide better classification.
  • [Show abstract] [Hide abstract]
    ABSTRACT: To effectively perform quantification of brain normal tissues and pathologies simultaneously, independent component analysis (ICA) coupled with support vector machine (SVM) is investigated and evaluated for effective volumetric measurements of normal and lesion tissues using multispectral MR images. Synthetic and real MR data of normal brain and white matter lesion (WML) data were used to evaluate the accuracy and reproducibility of gray matter (GM), white matter (WM), and WML volume measurements by using the proposed ICA+SVM method to analyze three sets of MR images, T1-weighted, T2-weighted, and proton density/fluid-attenuated inversion recovery images. The Tanimoto indexes of GM/WM classification in the normal synthetic data calculated by the ICA+SVM method were 0.82/0.89 for data with 0% noise level. As for clinical MR data experiments, the ICA+SVM method clearly extracted the normal tissues and white matter hyperintensity lesions from the MR images, with low intra- and inter-operator coefficient of variations. The experiments conducted provide evidence that the ICA+SVM method has shown promise and potential in applications to classification of normal and pathological tissues in brain MRI.
    Journal of Magnetic Resonance Imaging 07/2010; 32(1):24-34. · 2.57 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper develops a volume-based technique, called Volume Sphering Analysis (VSA) which can process all acquired Magnetic Resonance (MR) image slices formed image cube using only one set of training samples obtained from an image slice. So, several significant advantages and benefits can be gained from our proposed VSA. In the past, when MR image classification is performed, each image slice requires its own specific training samples and training samples obtained from one slice are not applied to another slice. The VSA allows users to reduce computational time. In addition, it saves significant effort in selecting training samples for each of image slices. Thirdly, it is robust to all image slices compared to the traditional one-slice MR image classification which is sensitive to each image slice. Experimental results demonstrate that the VSA performs as well as does that using specifically selected training samples for individual image slices.
    Genetic and Evolutionary Computing, International Conference on. 01/2010;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Olfactory bulb (OB) volume has been shown to be an indicator of olfactory function. However, few studies have been done in Asia to investigate the influence of different disorders on OB volume. Data from patients with posttraumatic anosmia were collected in our department. Their olfactory thresholds were assessed by the phenyl ethyl alcohol threshold test. They were treated with a course of high-dose steroid, and followed up for at least 3 months without any olfactory improvement. Magnetic resonance imaging was subsequently used to measure patients' OB volumes. Subjects who self-reported their olfactory function was normal were also included in the control group for comparison. Fifty-four patients with posttraumatic anosmia and 30 subjects who self-reported their olfactory function was normal were enrolled in this study. The mean right OB volume was 45.2 mm3, and the mean left OB volume was 46.3 mm3 in patients with posttraumatic anosmia. The mean right OB volume was 59.7 mm3, and the mean left OB volume was 66.0 mm3 in control subjects. The OB volumes were significantly lower in patients with posttraumatic anosmia. OB volumes were significantly lower in Taiwanese patients with posttraumatic anosmia.
    American Journal of Rhinology and Allergy 11/2009; 23(6):582-4.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Subdural hematoma (SDH) of the spine following intracranial hemorrhage is extremely rare. We present a 35-year-old woman who suffered from headache and dizziness initially, and then lower back pain, lower limb weakness and paraparesis gradually developed within 1-2 weeks. Magnetic resonance imaging revealed intracranial and spinal SDH. No vascular abnormality was seen by brain and spinal angiography. Platelet count, prothrombin time, activated partial thromboplastin time, and inflammatory markers, including C-reactive protein, were normal. A diagnosis of spontaneous spinal and intracranial SDH was then confirmed surgically. Postoperative recovery was uneventful.
    Journal of the Formosan Medical Association 04/2009; 108(3):258-61. · 1.70 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Knee-related injuries including meniscal tears are common in both young athletes and the aging population, and require accurate diagnosis and surgical intervention when appropriate. With proper techniques and radiologists' experienced skills, confidence in detection of meniscal tears can be quite high. This paper develops a novel computer-aided detection (CAD) diagnostic system for automatic detection of meniscal tears in the knee. Evaluation of this CAD system using an archived database of images from 40 individuals with suspected knee injuries indicates that the sensitivity and specificity of the proposed CAD system are 83.87% and 75.19%, respectively, compared to the mean sensitivity and specificity of 77.41% and 81.39%, respectively, obtained by experienced radiologists in routine diagnosis without using the CAD. The experimental results suggest that the developed CAD system has great potential and promise in automatic detection of both simple and complex meniscal tears of the knee.
    IEEE transactions on medical imaging. 03/2009; 28(8):1308-16.
  • Journal of Cardiovascular Magnetic Resonance 01/2009; · 4.44 Impact Factor
  • Jyh-Wen Chai, San-Kan Lee
    SPIENewsroom 01/2009;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Independent component analysis (ICA) has found great promise in magnetic resonance (MR) image analysis. Unfortunately, two key issues have been overlooked and not investigated. One is the lack of MR images to be used to unmix signal sources of interest. Another is the use of random initial projection vectors by ICA, which causes inconsistent results. In order to address the first issue, this paper introduces a band-expansion process (BEP) to generate an additional new set of images from the original MR images via nonlinear functions. These newly generated images are then combined with the original MR images to provide sufficient MR images for ICA analysis. In order to resolve the second issue, a prioritized ICA (PICA) is designed to rank the ICA-generated independent components (ICs) so that MR brain tissue substances can be unmixed and separated by different ICs in a prioritized order. Finally, BEP and PICA are combined to further develop a new ICA-based approach, referred to as PICA-BEP to perform MR image analysis.
    IEEE Transactions on Biomedical Engineering 07/2008; · 2.23 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Primary nasal natural killer (NK)/T-cell lymphoma is the most common cellular subtype seen in nasal lymphomas. It is rare in the Western population but occurs more frequently in Asia, South America, and Mexico. The purpose of this study was to describe the computed tomography (CT) and magnetic resonance (MR) imaging findings of primary nasal NK/T-cell lymphoma. During the period between January 1990 and June 2006, the CT (n=24) and MR (n=6) images of 24 patients with biopsy-proved nasal NK/T-cell lymphoma were reviewed retrospectively. Both CT and MR images were evaluated for site and extent of disease and for pattern of involvement of adjacent areas. The most common symptoms at presentation were nasal obstruction, nasal discharge, and epistaxis. There was involvement of the unilateral nasal cavity in 16, bilateral nasal cavity including nasal septum in 5 and nasal choana in 3. Sites of extension outside the nasal cavity included tumor extension into paranasal sinuses (n=15), nasopharynx (n=5), nasal labial fold (n=3), oropharynx (n=2), infratemporal fossa (n=2), other subcutaneous soft tissue of the face (n=2) and anterior cranial fossa base (n=1). Bony destruction was demonstrated in 18 cases, involving the sinus bony wall (n=15), nasal turbinate (n=10), lamina papyracea (n=6), orbital floor (n=3), and hard palate (n=2). Regional lymphadenopathy was also detected in 3 patients with nasal NK/T-cell lymphoma. The CT and MR appearances of nasal NK/T-cell lymphoma are nonspecific, and the diagnosis requires histologic confirmation. However, the differential diagnosis of nasal NK/T-cell lymphoma should be included if the images present soft tissue of the nasal cavity with bony erosion or destruction; involvement of the orbital cavity, nasopharynx and infratemporal fossa; and subcutaneous or nasolabial fold soft tissue infiltration, especially in Asian populations.
    Journal of the Chinese Medical Association 06/2007; 70(5):207-12. · 0.89 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A simple gradient-echo arterial spin tagging (GREAST) technique allows for quick assessment of regional tissue perfusion without the need for exogenous contrast agent. The purpose of this prospective study was to validate GREAST imaging in characterizing the regional perfusion status of focal brain lesions by comparing with relative cerebral blood volume (rCBV) maps obtained by using echo-planar imaging (EPI)-based dynamic susceptibility contrast MR imaging. Thirty-two patients whose nonenhanced brain MR images showed 34 focal brain lesions during routine examination were selected to immediately undergo GREAST and dynamic susceptibility contrast MR imaging to evaluate regional perfusion of the lesions. The Pearson correlation coefficient was used to test the relative quantification of local perfusion with the two imaging methods. Qualitative perfusion measurements agreed in 23 (79%) of 29 lesions for which GREAST and dynamic susceptibility contrast MR imaging were successful. On rCBV maps, six focal lesions with local hemorrhage were underestimated. In three patients with metal surgical implants, lesions could not be measured because of susceptibility artifacts and distortion on EPIs. After these lesions were excluded, the Pearson correlation coefficient between relative quantitative perfusion measurements on GREAST images versus rCBV maps was about 0.90 (p value = 0.000). The success rate of GREAST imaging was 94% (30 of 32 patients), higher than that of dynamic susceptibility contrast MR imaging (72%, or 23 of 32 patients). GREAST imaging was comparable to rCBV mapping for the relative quantification of regional perfusion of focal brain lesions. This technique may be useful in routine MR examination for characterizing the regional perfusion of brain focal lesions.
    The neuroradiology journal. 04/2007; 20(2):149-58.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Lhermitte-Duclos disease (LDD) is a rare benign lesion of uncertain pathogenesis characterized by distortion of the normal cerebellar laminar cytoarchitecture. We report a 22-year-old man admitted for injury sustained in a traffic accident with the incidental finding of a cerebellar mass. Magnetic resonance imaging (MRI) revealed a mass lesion within the right cerebellar hemisphere. The final diagnosis of LDD was made by obtaining a surgical specimen and identifying the characteristic appearance of the lesion by MRI study. The images showed the typical striated pattern of hyperintensity on T2-weighted images and corresponding hypointensity on T1-weighted images, as well as the typical absence of enhancement following gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA) administration. In addition, no disturbance of water diffusion on diffusion-weighted MRI, and associations of decreases in the N-acetylaspartate/creatine (NAA/Cr) and NAA/Choline (Cho) ratios with near normal values of Cho/Cr, as well as an obvious lactate peak gave supplemental information for diagnosis.
    Journal of the Chinese Medical Association 08/2006; 69(7):338-42. · 0.89 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To introduce a simple gradient-echo arterial spin tagging (GREAST) technique available for most commercial magnetic resonance (MR) systems, for a quick evaluation of tissue perfusion. The GREAST technique uses a combination of a short TR spoiled gradient-echo (SPGR) sequence with a selective presaturation radio frequency (RF) pulse that allows acquiring each tagged and control image within 10-20 seconds. The phantom and human studies were performed for evaluating the feasibility in measurement of local perfusion and the efficacy in alleviation of the asymmetric magnetization transfer (MT) and slice profile effects. Results show a good linear relationship between the signal attenuation caused by the presaturation pulse and flow rates in the phantom experiment and effective alleviation of the asymmetric MT and slice profile effects for various orientations of imaging slices. Human studies showed good perfusion results in brain imaging. Perfusion imaging on the liver and kidney were also conducted. The results could be significantly improved by effectively lessening motion-related artifacts. The GREAST technique is simple, easy to use, and applicable to examine local perfusion of the brain and other organs in the trunk.
    Journal of Magnetic Resonance Imaging 08/2002; 16(1):51-9. · 2.79 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: By using on-resonance binomial pulse saturation of the immobile component in tissues and by monitoring the time development of the longitudinal decay of the free water magnetization, it has been shown that it is possible to estimate physical model parameters not directly measurable by conventional methods. Since pulsed saturation is easy and safe to implement in a clinical MRI machine, it should be possible to estimate these parameters in vivo by the same means. In this article, we report the results of such estimates of the parameters of the principal components (gray and white matter) in brains of two normal volunteers.
    Magnetic Resonance in Medicine 06/1996; 36(1):147 - 152. · 3.40 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To develop a new magnetic resonance cholangiopancreatography technique for patients who cannot hold their breath or breathe regularly and fail to be successfully examined with conventional magnetic resonance cholangiopancreatography methods. Within a one-year period, 15 patients including 6 children and 9 adults were studied. Magnetic resonance cholangiopancreatography was performed in a 1.5 Tesla GE MR scanner with capability of single-shot fast spin echo sequence. In all patients except for the children, magnetic resonance cholangiopancreatography was attempted with the breathhold technique at first. However, this failed in some due either to the patients being too old or too weak to hold their breath long enough for single-shot fast spin echo sequence, which usually took about 20-30 seconds for the complete scanning. These 15 cases were then scanned slice-by-slice in operator's control mode by monitoring patient's respiratory pattern from the TV monitor. Scanning was started near the end of patient's expiration. This technique was performed after we had carefully decided the baseline of each patient's respiration cycle. Each slice was scanned at an interval depending on the patient's respiratory frequency. It was acquired at an interval of two respiratory cycles for each sequential slice when the patient's respiration frequency was around 10-15 cycles per minute, at three cycles when respiration frequency was around 15-20, at four cycles when the frequency was around 20-25, and at an interval of 5 cycles when it was above 25. The acquired source images were then reconstructed for a 3D image. Magnetic resonance cholangiopancreatography of good image quality was obtained in all of these fifteen patients. Each set of images took about 2-3 minutes. No marked artifact was found. The reconstructed 3D image also afforded satisfactory quality for evaluation of both normal ductal anatomy and lesions of the biliary-pancreatic system. The axial single-shot fast spin echo sequence images of the liver and pancreas were also successfully obtained with this method. Magnetic resonance cholangiopancreatography findings in these 15 patients included type I choledochal cyst (n = 4), dilatation of the intrahepatic bile ducts due to mass compression (n = 3, one Klatskin tumor and two hilar masses), pancreatic carcinoma (n = 1), acute cholecystitis without biliary tract dilatation (n = 1), acute pancreatitis with mild biliary dilatation and non-visible pancreatic duct (n = 1), dilatation of biliary tract without definite lesions (n = 2), common hepatic duct obstruction (n = 1), and normal biliary and pancreatic duct without dilatation or lesions (n = 2). Free breathing magnetic resonance cholangiopancreatography technique is very useful for patients in whom conventional methods cannot be successfully undertaken. It affords informative images that are comparable to other magnetic resonance methods.
    Hepato-gastroenterology 49(45):593-6. · 0.91 Impact Factor

Publication Stats

61 Citations
27.69 Total Impact Points


  • 1996–2014
    • Taichung Veterans General Hospital
      • Department of Radiology
      臺中市, Taiwan, Taiwan
  • 2009
    • University of Maryland, Baltimore
      Baltimore, Maryland, United States
  • 2008
    • National Chung Hsing University
      • Department of Electrical Engineering
      Taichung, Taiwan, Taiwan
  • 2007
    • National Yang Ming University
      • School of Medicine
      T’ai-pei, Taipei, Taiwan