Filters








553 Hits in 2.7 sec

Intelligent Photographing Interface with On-Device Aesthetic Quality Assessment [chapter]

Kuo-Yen Lo, Keng-Hao Liu, Chu-Song Chen
2013 Lecture Notes in Computer Science  
This paper proposes a efficient method for instant photo aesthetics quality assessment that can be implemented on general portable devices. The classification performance is guaranteed to 0.89 on benchmark photo database. We also port our method onto a middle-level tablet computer to execute instantly and we find it reaches good acceptable efficiency. Moreover, an aesthetic information display to present the aesthetics evaluation results to users is introduced.
doi:10.1007/978-3-642-37484-5_43 fatcat:ppc4fgefmzdjtmkp3tfnupjkay

Selection Strategy for VM Migration Method

Yan-Ren Chen, I-Hsien Liu, Keng-Hao Chang, Chuan-Gang Liu, Jung-Shian Li
2019 Journal of Robotics, Networking and Artificial Life (JRNAL)  
I-Hsien Liu He is a research fellow in the Taiwan Information Security Center @ National Cheng Kung University (TWISC@NCKU) and department of electrical engineering, National Cheng Kung University, Taiwan  ...  Chuan-Gang Liu He is an associate professor in the department of Applied informatics and Multimedia, Chia Nan university of Pharmacy and Science.  ... 
doi:10.2991/jrnal.k.190602.003 fatcat:kphigt3y6jftdjhethx2nf7kue

Progressive Sample Processing of Band Selection for Hyperspectral Image Transmission

Keng-Hao Liu, Shih-Yu Chen, Hung-Chang Chien, Meng-Han Lu
2018 Remote Sensing  
Band selection (BS) is one of the important topics in hyperspectral image (HSI) processing. Many types of BS algorithms were proposed in the last decade. However, most of them were designed for off-line use. They can only be used with pre-collected data, and are sometimes ineffective for applications that require timeliness, such as disaster prevention or target detection. This paper proposes an online BS method that allows us obtain instant BS results in a progressive manner during HSI data
more » ... nsmission, which is carried out under band-interleaved-by-sample/pixel (BIS/BIP) format. Such a revolutionary method is called progressive sample processing of band selection (PSP-BS). In PSP-BS, BS can be done recursively pixel by pixel, so that the instantaneous BS can be achieved without waiting for all the pixels of an image. To develop a PSP-BS algorithm, we proposed PSP-OMPBS, which adopted the recursive version of a self-sparse regression BS method (OMPBS) as a native algorithm. The experiments conducted on two real hyperspectral images demonstrate that PSP-OMPBS can progressively output the BS with extremely low computing time. In addition, the convergence of BS results during transmission can be further accelerated by using a pre-defined pixel transmission sequence. Such a significant advantage not only allows BS to be done in a real-time manner for the future satellite data downlink, but also determines the BS results in advance, without waiting to receive every pixel of an image.
doi:10.3390/rs10030367 fatcat:sg2yjtasmba5fih5e7tybr6sku

Surgical management of gallbladder sarcomatoid carcinoma

Keng-Hao Liu, Ta-Sen Yeh, Tsann-Long Hwang, Yi-Yin Jan, Miin-Fu Chen
2009 World Journal of Gastroenterology  
AIM: To study the behavior as well as optimal treatment of gallbladder sarcomatoid carcinoma, we reviewed the results of treatment of gallbladder sarcomatoid carcinoma from Chang Gung Memorial Hospital. METHODS: From 1987 to 2005, six patients were diagnosed with gallbladder sarcomatoid carcinoma and treated at our institution. Tumor staging was based on 2002 revised tumor-node-metastasis (TNM) staging for gall bladder cancer from the American Joint Committee on Cancer. The clinical
more » ... , laboratory d a t a a n d p r e o p e ra t i ve w o r k u p w e r e r e v i e w e d retrospectively. RESULTS: Five patients were female and one was male. The age ranged from 51 to 66 years (median, 58 years). Surgical procedures included three curative resections, two palliative resections and one biopsy. There were two surgical complications (33.3%) and one case of surgical mortality (16.7%). The followup time ranged from 30 d to 5 mo. The median survival was 2.5 mo. The prognosis was extremely poor, even after curative resection and postoperative chemotherapy. CO N C L U S I O N: T h e p r o g n o s i s o f g a l l b l a d d e r sarcomatoid carcinoma was not dependent on TNM stage and was always dismal. The clinicopathological features were different from those of gall bladder cancer. P e e r re v i e w e r : To r u I s h i k a w a , M D , D e p a r t m e n t o f Gastroenterology, Saiseikai Niigata Second Hospital, Teraji 280-7,
doi:10.3748/wjg.15.1876 pmid:19370786 pmcid:PMC2670416 fatcat:qr5t252zovfipkw2cp4s5jqeca

Salient object detection via local saliency estimation and global homogeneity refinement

Hsin-Ho Yeh, Keng-Hao Liu, Chu-Song Chen
2014 Pattern Recognition  
This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex
more » ... rm) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights a b s t r a c t This paper presents a new hybrid approach for detecting salient objects in an image. It consists of two processes: local saliency estimation and global-homogeneity refinement. We model the salient object detection problem as a region growing and competition process by propagating the influence of foreground and background seed-patches. First, the initial local saliency of each image patch is measured by fusing local contrasts with spatial priors, thereby the seed-patches of foreground and background are constructed. Later, the global-homogeneous information is utilized to refine the saliency results by evaluating the ratio of the foreground and background likelihoods propagated from the seed-patches. Despite the idea is simple, our method can effectively achieve consistent performance for detecting object saliency. The experimental results demonstrate that our proposed method can accomplish remarkable precision and recall rates with good computational efficiency.
doi:10.1016/j.patcog.2013.11.015 fatcat:ujdn4ccf5rbsdft6rbsixel66a

The Diet-Aware Dining Table: Observing Dietary Behaviors over a Tabletop Surface [chapter]

Keng-hao Chang, Shih-yen Liu, Hao-hua Chu, Jane Yung-jen Hsu, Cheryl Chen, Tung-yun Lin, Chieh-yu Chen, Polly Huang
2006 Lecture Notes in Computer Science  
We are what we eat. Our everyday food choices affect our long-term and short-term health. In the traditional health care, professionals assess and weigh each individual's dietary intake using intensive labor at high cost. In this paper, we design and implement a diet-aware dining table that can track what and how much we eat. To enable automated food tracking, the dining table is augmented with two layers of weighing and RFID sensor surfaces. We devise a weight-RFID matching algorithm to detect
more » ... and distinguish how people eat. To validate our diet-aware dining table, we have performed experiments, including live dining scenarios (afternoon tea and Chinese-style dinner), multiple dining participants, and concurrent activities chosen randomly. Our experimental results have shown encouraging recognition accuracy, around 80%. We believe monitoring the dietary behaviors of individuals potentially contribute to dietaware healthcare.
doi:10.1007/11748625_23 fatcat:mljfz72xrbbtfo5d7eooigunxq

N3 Subclassification Incorporated into the Final Pathologic Staging of Gastric Cancer

Chun-Nan Yeh, Shang-Yu Wang, Jun-Te Hsu, Kun-Chun Chiang, Chi-Tung Cheng, Chun-Yi Tsai, Yu-Yin Liu, Chien-Hung Liao, Keng-Hao Liu, Ta-Sen Yeh
2015 Medicine  
The seventh edition of the American Joint Committee on Cancer (AJCC) TNM classification system for gastric cancer (GC) was established in 2009. We assessed the unmet medical needs of patients with the N3 classification of the seventh TNM staging system by comparing survival according to the extent of nodal involvement, with a particular focus on the cutoff points for the number of involved nodes in the N3 classification. We retrospectively reviewed 3178 patients with GC who were registered in
more » ... e GC database of the Department of General Surgery at the Chang Gung Memorial Hospital between 1994 and 2010. Among them, 884 patients undergoing curative intent resection had N3 lymph node involvement. The clinicopathological features and surgical outcomes were compared among all patients with GC and between the N3a and N3b groups. N3b might impose GC patients with poor clinical outcome. We proposed a modified staging system, based on AJCC seventh edition, accordingly. T1-3N3 might be not simply categorized into stage IIIA as seventh AJCC suggested. Taking N3a and N3b into consideration, T1-3N3 might be further categorized into stage IIIB and IIIC, respectively, as we proposed, based on survival analysis. In addition, T4bN3bM0 is as dismal as M1 disease. In our proposed staging system, good discriminations between different stages are still maintained. The N3 category should be subclassified as N3a or N3b due to the survival differences. Furthermore, T1-3N3aM0 could be categorized as stage IIIB, T1-3N3bM0 could be categorized as stage IIIC, T4aN3bM0 could be categorized as stage IIID, and T4bN3bM0 might be regarded as stage IV as we proposed. (Medicine 94(8):e575) Abbreviations: AJCC = American Joint Committee on Cancer, CA = carbohydrate antigen, CEA = carcinoembryonic antigen, CGMH = Chang Gung Memorial Hospital, CT = computed tomography, GC = gastric cancer, LN = lymph node, US = ultrasonography. Editor: Neil Merrett.
doi:10.1097/md.0000000000000575 pmid:25715257 pmcid:PMC4554155 fatcat:3vrbfiqyerb3tl4yx2afil5bcy

Enterovesical fistula caused by a bladder squamous cell carcinoma

Chun-Hsiang Ou Yang, Keng-Hao Liu, Tse-Ching Chen, Phei-Lang Chang, Ta-Sen Yeh
2009 World Journal of Gastroenterology  
Enterovesical fistulas are not uncommon in patients with inflammatory or malignant colonic disease, however, fistulas secondary to primary bladder carcinomas are extremely rare. We herein reported a patient presenting with intractable urinary tract infection due to enterovesical fistula formation caused by a squamous cell carcinoma of the urinary bladder. This patient underwent en bloc resection of the bladder dome and involved ileum, and recovered uneventfully without urinary complaint. To the
more » ... best of our knowledge, this is the first case reported in the literature. TS. Enterovesical fistula caused by a bladder squamous cell carcinoma. World J Gastroenterol 2009; 15(33): 4215-4217 Available from: URL:
doi:10.3748/wjg.15.4215 pmid:19725163 pmcid:PMC2738825 fatcat:eh2oracd7be4db5imk7qnrwlha

A Motion Deblur Method based on Multi-scale High Frequency Residual Image Learning

Keng-Hao Liu, Chia-Hung Yeh, Juh-Wei Chung, Chuan-Yu Chang
2020 IEEE Access  
Non-uniform blind deblurring of dynamic scenes has always been a challenging problem in image processing because of the diverse of blurring sources. Traditional methods based on energy minimization cannot make accurate kernel estimation. It leads to that some high frequency details cannot be fully recovered. Recently, many methods based on convolution neural networks (CNNs) have been proposed to improve the overall performance. Followed by this trend, we first propose a two-stage deblurring
more » ... le to recover the blur images of dynamic scenes based on high frequency residual image learning. The first stage performs initial deburring with the blur kernel estimated by the salient structure. The second stage calculates the difference of input image and initially deblurred image, referred to as residual image, and adopt an encoder-decoder network to refine the residual image. Finally, we can combine the refined residual image with the input blurred image to obtain the latent image. To increase deblurring performance, we further propose a coarse-to-fine framework based on the deblurring module. It performs the deblurring module many times in a multi-scale manner which can gradually restore the sharp edge details of different scales. Experiments conducted on three benchmark datasets demonstrate the proposed method achieves competitive performance of state-of-art methods. INDEX TERMS Image deblurring, dynamic blur, non-uniform blind deblurring, deep learning.
doi:10.1109/access.2020.2985220 fatcat:qmon6jfaonemlnhpme64osxspi

Incidence, Patient-Related Risk Factors, and Outcomes of Postoperative Pneumonia after Cholecystectomy: A Population-Based Cohort Study

Yun-Hui Teng, Fu-Chao Liu, Keng-Hao Liu, Jr-Rung Lin, Huang-Ping Yu, Gernot Kaiser
2021 BioMed Research International  
Y-H Teng, F-C Liu, J-R Lin, and H-P Yu are responsible for the data analysis and interpretation. F-C Liu, K-H Liu, and H-P Yu are responsible for the funding acquisition.  ...  F-C Liu, Y-H Teng, and H-P Yu conceived and designed the study. K-H Liu and H-P Yu are responsible for the data acquisition.  ... 
doi:10.1155/2021/6614885 pmid:34055988 pmcid:PMC8131140 fatcat:rgm46khrujhojfmwkmndb5ihwu

Post-gastrectomy acute pancreatitis in a patient with gastric carcinoma and pancreas divisum

I-Ming Kuo, Frank Wang, Keng-Hao Liu, Yi-Yin Jan
2009 World Journal of Gastroenterology  
Gastrectomy is commonly performed for both benign and malignant lesions. Although the incidence of post-gastrectomy acute pancreatitis (PGAP) is low compared to other well-recognized post-operative complications, it has been reported to be associated with a high mortality rate. In this article, we describe a 70-year-old man with asymptomatic pancreatic divisum who underwent palliative subtotal gastrectomy for an advanced gastric cancer with liver metastasis. His postoperative course was
more » ... ted by acute pancreatitis and intra-abdominal sepsis. The patient eventually succumbed to multiple organ failure despite surgical debridement and drainage, together with aggressive antibiotic therapy and nutritional support. For patients with pancreas divisum or dominant duct of Santorini who fail to follow the normal post-operative course after gastrectomy, clinicians should be alert to the possibility of PGAP as one of the potential diagnoses. Early detection and aggressive treatment of PGAP might improve the prognosis. Post-gastrectomy acute pancreatitis in a patient with gastric carcinoma and pancreas divisum. World J Gastroenterol 2009; 15(36): 4596-4600 Available from: URL
doi:10.3748/wjg.15.4596 pmid:19777622 pmcid:PMC2752008 fatcat:3uxplogpibgtxhkqhq2mvyj3cq

Hybrid Deep Learning Models with Sparse Enhancement Technique for Detection of Newly Grown Tree Leaves

Shih-Yu Chen, Chinsu Lin, Guan-Jie Li, Yu-Chun Hsu, Keng-Hao Liu
2021 Sensors  
The life cycle of leaves, from sprout to senescence, is the phenomenon of regular changes such as budding, branching, leaf spreading, flowering, fruiting, leaf fall, and dormancy due to seasonal climate changes. It is the effect of temperature and moisture in the life cycle on physiological changes, so the detection of newly grown leaves (NGL) is helpful for the estimation of tree growth and even climate change. This study focused on the detection of NGL based on deep learning convolutional
more » ... al network (CNN) models with sparse enhancement (SE). As the NGL areas found in forest images have similar sparse characteristics, we used a sparse image to enhance the signal of the NGL. The difference between the NGL and the background could be further improved. We then proposed hybrid CNN models that combined U-net and SegNet features to perform image segmentation. As the NGL in the image were relatively small and tiny targets, in terms of data characteristics, they also belonged to the problem of imbalanced data. Therefore, this paper further proposed 3-Layer SegNet, 3-Layer U-SegNet, 2-Layer U-SegNet, and 2-Layer Conv-U-SegNet architectures to reduce the pooling degree of traditional semantic segmentation models, and used a loss function to increase the weight of the NGL. According to the experimental results, our proposed algorithms were indeed helpful for the image segmentation of NGL and could achieve better kappa results by 0.743.
doi:10.3390/s21062077 pmid:33809537 fatcat:ad6n4d4jyzdjply47oxjwdtgve

Clinicopathologic features and responses to radiotherapy of myeloid sarcoma

Wan-Yu Chen, Chun-Wei Wang, Chin-Hao Chang, Heng-Hsiu Liu, Keng-Hsueh Lan, Jih-Luh Tang, Hwei-Fang Tien, Sung-Hsin Kuo, Ann-Lii Cheng
2013 Radiation Oncology  
To evaluate clinicopathological features, radiotherapeutic parameters, and their associations with responses to radiotherapy (RT) in patients with myeloid sarcoma (MS). Methods: We reviewed 20 patients receiving RT for MS lesions (in 43 RT courses) and analyzed the patients' clinicopathologic features and radiotherapeutic parameters, and their associations with complete responses (CR) to RT using Fisher's exact test and univariate logistic regression analysis. Generalized Estimating Equation
more » ... used to analyze all 43 irradiated lesions and account for the correlations in RT responses among lesions from the same patient. Results: We found that the underlying hematological diseases of the evaluated patients were acute myeloid leukemia (AML) in 14 patients (70%), chronic myeloid leukemia in 4 patients (20%), myelodysplastic syndrome with AML transformation in one patient (5%), and de novo MS in one patient (5%). Most patients (55%) received RT for MS at the time of relapse following bone marrow transplantation (BMT). The most common cytogenetic abnormality was t(8;21)(q22;q22). The median RT dose of 20 Gy (range 6-35 Gy), administered in 1.5-3.5 Gy fractions, provided a 63% CR rate. RT dose, sex, cytogenetics, and bone marrow status at the time of RT had no significant effect on CR. Younger age (<50 y, P = 0.06), BMT prior to RT (P = 0.05), and underlying AML (P = 0.05) were marginally associated with higher CR to RT. Conclusions: Our results indicate that a modest RT dose (20-30 Gy) achieves good local control of MS. Age, previous BMT, and underlying hematologic disease can affect RT response.
doi:10.1186/1748-717x-8-245 pmid:24148102 pmcid:PMC4016483 fatcat:5kskot7sjzf2njaqq3zqqrm4lq

Selective reoperation after primary resection as a feasible and safe treatment strategy for recurrent pancreatic cancer

Shih-Chun Chang, Chih-Po Hsu, Chun-Yi Tsai, Yu-Yin Liu, Keng-Hao Liu, Jun-Te Hsu, Ta-Sen Yeh, Chun-Nan Yeh, Tsann-Long Hwang
2016 Medicine  
Local recurrence frequently occurs in patients with pancreatic cancer after intended curative resections. However, no treatment strategies have been established for isolated local recurrence. Several series have demonstrated a survival benefit for reoperation in selected pancreatic recurrence cases. This study compares the difference in overall survival (OS) between surgery and nonsurgery groups in recurrent pancreatic cancer. All patients from 1990 to 2014 with recurrent pancreatic cancer who
more » ... nderwent curative resections were investigated and retrospectively reviewed. Clinicopathological features and OS were compared. A total of 332 patients were recruited in this series. The majority had histologically pancreatic adenocarcinoma (289 patients, 87.0%). Fourteen of 332 patients (4.2%) with recurrent pancreatic cancer received subsequent resection. Most of these patients underwent curative surgery (R0 resection, 13 patients, 92.9%), and only 1 patient (7.1%) had microscopic residual tumor (R1 resection). Disease-free survival (DFS), OS, and postrecurrence survival (PRS) were all significantly longer in the surgery group (DFS 10.6 vs 6.1 months, P = 0.044; OS 57.8 vs 14.0 months, P < 0.001; PRS 14.1 vs 6.0 months, P < 0.001). The median survival times were comparable in patients with recurrent pancreatic adenocarcinoma who received surgery and those who did not (DFS 10.6 vs 6.1 months, P = 0.226; OS 23.7 vs 14.0 months, P = 0.074; PRS 8.9 vs 5.8 months, P = 0.183). However, the OS and PRS were superior in the patients who did not display adenocarcinoma histologically but underwent operation for recurrence (OS 97.2 vs 16.9 months, P = 0.016; PRS 65.7 vs 6.9 months, P = 0.010). Notably, DFS levels were similar (16.0 vs 7.0 months, P = 0.265). Surgery can feasibly and safely provide survival benefits in selective recurrent pancreatic cancer. In patients who are histologically negative for adenocarcinoma, survival is prolonged when the operation is performed with R0 resection. Patients with isolated recurrent pancreatic adenocarcinoma need multidisciplinary therapy. In addition to operation, chemoradiotherapy and intraoperative radiotherapy may also be considered; their roles should be further investigated. Abbreviations: DFS = disease-free survival, IORT = intraoperative radiotherapy, OS = overall survival, PRS = postrecurrence-free survival.
doi:10.1097/md.0000000000004191 pmid:27472688 pmcid:PMC5265825 fatcat:3qubbsh42fgwdbsexfew6bwlr4

Microencapsulation improves inhibitory effects of transplanted olfactory ensheathing cells on pain after sciatic nerve injury

Zeng-xu Liu, Hao Zhao, Bao-lin Yang, Qing Yu, Wen-jun Zhang, Keng Yuan, Hui-hong Zeng, Gao-chun Zhu, De-ming Liu, Qing Li
2015 Neural Regeneration Research  
Therefore, OEC transplantation not only exerts a therapeutic effect on nerve damage, but also diminishes immunological rejection (Liu et al., 2013; Valente et al., 2013; Cao et al., 2014) .  ...  suspension increased the expression of neurofilament 200, but suppressed that of glial fibrillary acidic protein in the injured spinal cord, and contributed to the recovery of spinal cord function (Liu  ... 
doi:10.4103/1673-5374.162769 pmid:26487865 pmcid:PMC4590250 fatcat:mwclelw2vrdijio3dkpm2k5g7q
« Previous Showing results 1 — 15 out of 553 results