An Efficient Parallel Machine Learning-based Blockchain Framework

Chun-Wei Tsai,Yi-Ping Chen,Tzu-Chieh Tang,Yu-Chen Luo 한국통신학회 2021 ICT Express Vol.7 No.3

The unlimited possibilities of machine learning have been shown in several successful reports and applications. However, how to make sure that the searched results of a machine learning system are not tampered by anyone and how to prevent the other users in the same network environment from easily getting our private data are two critical research issues when we immerse into powerful machine learning-based systems or applications. This situation is just like other modern information systems that confront security and privacy issues. The development of blockchain provides us an alternative way to address these two issues. That is why some recent studies have attempted to develop machine learning systems with blockchain technologies or to apply machine learning methods to blockchain systems. To show what the combination of blockchain and machine learning is capable of doing, in this paper, we proposed a parallel framework to find out suitable hyperparameters of deep learning in a blockchain environment by using a metaheuristic algorithm. The proposed framework also takes into account the issue of communication cost, by limiting the number of information exchanges between miners and blockchain.

Increased Readmission Risk and Healthcare Cost for Delirium Patients without Immediate Hospitalization in the Emergency Department

I Chun Ma,Kao Chin Chen,Wei Tseng Chen,Hsin Chun Tsai,Chien-Chou Su,Ru-Band Lu,Po See Chen,Wei Hung Chang,Yen Kuang Yang 대한정신약물학회 2018 CLINICAL PSYCHOPHARMACOLOGY AND NEUROSCIENCE Vol.16 No.4

Objective: Hospitalization of patients with delirium after visiting the emergency department (ED) is often required. However, the readmission risk after discharge from the ED should also be considered. This study aimed to explore whether (i) immediate hospitalization influences the readmission risk of patients with delirium; (ii) the readmission risk is affected by various risk factors; and (iii) the healthcare cost differs between groups within 28 days of the first ED visit. Methods: Using the National Health Insurance Research Database, the data of 2,780 subjects presenting with delirium at an ED visit from 2000 to 2008 were examined. The readmission risks of the groups of patients (i.e., patients who were and were not admitted within 24 hours of an ED visit) within 28 days were compared, and the effects of the severities of different comorbidities (using Charlson’s comorbidity index, CCI), age, gender, diagnosis and differences in medical healthcare cost were analyzed. Results: Patients without immediate hospitalization had a higher risk of readmission within 3, 7, 14, or 28 days of discharge from the ED, especially subjects with more severe comorbidities (CCI≥3) or older patients (≥65 years). Subjects with more severe comorbidities or older subjects who were not admitted immediately also incurred a greater healthcare cost for re-hospitalization within the 28-day follow-up period. Conclusion: Patients with delirium with a higher CCI or of a greater age should be carefully considered for immediate hospitalization from ED for further examination in order to reduce the risk of re-hospitalization and cost of healthcare.

Enhanced long-range personal identification based on multimodal information of human features

Tsai, Hsin-Chun,Chen, Bo-Wei,Wang, Jhing-Fa,Paul, Anand Springer-Verlag 2014 MULTIMEDIA TOOLS AND APPLICATIONS Vol.73 No.1

Dynamic stability analysis of laminated composite plates in thermal environments

Chun-Sheng Chen,Wei-Ren Chen,Ting-Chiang Tsai,Ching-Long Wei,이재홍,전찬기 국제구조공학회 2013 Steel and Composite Structures, An International J Vol.15 No.1

This paper studies the dynamic instability of laminated composite plates under thermal and arbitrary in-plane periodic loads using first-order shear deformation plate theory. The governing partial differential equations of motion are established by a perturbation technique. Then, the Galerkin method is applied to reduce the partial differential equations to ordinary differential equations. Based on Bolotin's method, the system equations of Mathieu-type are formulated and used to determine dynamic instability regions of laminated plates in the thermal environment. The effects of temperature, layer number, modulus ratio and load parameters on the dynamic instability of laminated plates are investigated. The results reveal that static and dynamic load, layer number, modulus ratio and uniform temperature rise have a significant influence on the thermal dynamic behavior of laminated plates.

A retrospective analysis of 20-year data of the surgical management of ulcerative colitis patients in Taiwan: a study of Taiwan Society of Inflammatory Bowel Disease

( Chun-chi Lin ),( Shu-chen Wei ),( Been-ren Lin ),( Wen-sy-tsai ),( Jinn-shiun Chen ),( Tzu-chi Hsu ),( Wei-chen Lin ),( Tien-yu Huang ),( Te-hsin Chao ),( Hung-hsin Lin ),( Jau-min Wong ),( Jen-kou 대한장연구학회 2016 Intestinal Research Vol.14 No.3

Background/Aims: With the recent progress in medical treatment, surgery still plays a necessary and important role in treating ulcerative colitis (UC) patients. In this study, we analyzed the surgical results and outcomes of UC in Taiwan in the recent 20 years, via a multi-center study through the collaboration of Taiwan Society of IBD. Methods: A retrospective analysis of surgery data of UC patients from January 1, 1995, through December 31, 2014, in 6 Taiwan major medical centers was conducted. The patients’ demographic data, indications for surgery, and outcome details were recorded and analyzed. Results: The data of 87 UC patients who received surgical treatment were recorded. The median post-operative follow-up duration was 51.1 months and ranged from 0.4 to 300 months. The mean age at UC diagnosis was 45.3±16.0 years and that at operation was 48.5±15.2 years. The 3 leading indications for surgical intervention were uncontrolled bleeding (16.1%), perforation (13.8%), and intractability (12.6%). In total, 27.6% of surgeries were performed in an emergency setting. Total or subtotal colectomy with rectal preservation (41.4%) was the most common operation. There were 6 mortalities, all due to sepsis. Emergency operation and low pre-operative albumin level were significantly associated with poor survival (P =0.013 and 0.034, respectively). Conclusions: In the past 20 years, there was no significant change in the indications for surgery in UC patients. Emergency surgeries and low pre-operative albumin level were associated with poor survival. Therefore, an optimal timing of elective surgery for people with poorly controlled UC is paramount.

Application of Data Quality Indicator of Carbon Footprint and Water Footprint

Tsai Chi Kuo,Miao-Ling Huang,Chia Wei Hsu,Chiuhsiang Joe Lin,Chih-Chun Hsieh,Chih-Hsing Chu 한국정밀공학회 2015 International Journal of Precision Engineering and Vol.2 No.1

To mitigate the impact of global warming on the environment, many governments, non-profit organizations, and enterprises have formulated relevant standards to assist enterprises in promoting carbon management, and to encourage the design and popularization of low-carbon products. These standards include: PAS 2050, ISO/TS 14067 and so on. Under these initiatives, enterprises try to disclose the carbon and water footprints of the products based on the life cycle. Some enterprises argue and debate that there exists uncertainty of the footprints calculation since the data collection is not systematically. To solve this problem, in this research, the calculation of carbon footprint and water footprint are evaluated with the data quality indicator (DQI) management system. The collected data is evaluated based on the footprint calculation methods. Also the pedigree matrix is constructed as an aide to solve the data uncertainty that included reliability, completeness, times, geography, and technologies differences. Through the DQI, the carbon footprint and water footprint are not only calculated simultaneously, but also correctly. The results could be the reference for products’ environmental improvement.

Clinical Features of Patients with Esophageal and Second Primary Cancers

Tsai, Huang-Wen,Chang, Chih-Chun,Sun, Jen-Tang,Liou, Ching-Biau,Lin, Hsiu-Chen,Lin, I-Hsin,Yu, Yun-Chieh,Weng, Wei-Ling,Leong, Ka-I,Yen, Tzung-Hai,Wu, Jiann-Ming Asian Pacific Journal of Cancer Prevention 2014 Asian Pacific journal of cancer prevention Vol.15 No.22

Background: The prevalence of esophageal cancer (EC) with second primary cancers (SPC) is increasing worldwide. This study was aimed to understand the clinical features of EC patients with SPC in the Taiwanese population. Materials and Methods: Clinical and laboratory data for 180 EC patients with or without SPC were collected between January 2009 and December 2013. Information on treatment approaches, location of SPCs and ABO blood type were also collected and stratified. Results: The most common SPC in EC patients was hypopharyngeal cancer, followed by laryngeal cancer and hepatocellular carcinoma in our study. Malignancies of colon, prostate and lung were also found. There was a significant higher portion of blood type A in the EC patients with SPC compared with those without (42.4% vs 19.5%, P=0.006). Conclusions: The frequency and SPC site distribution and blood type A should be considered in clinical evaluation of EC patients with a high risk of developing SPC in the Taiwanese population.

Phase transition and energy transfer of lead-free Cs2SnCl6 perovskite nanocrystals by controlling the precursors and doping manganese ions

Tsai-Wei Lin,Chaochin Su,Chun Che Lin 한국정보디스플레이학회 2019 Journal of information display Vol.20 No.4

Perovskite quantum dots (QDs), such as all-inorganic CsPbX3 (X = Cl, Br, and I), are novel fluorescent semiconductor nanocrystals (NCs) that have attracted tremendous attention due to their excellent optical properties and great applications (e.g. display backlights, light-emitting diodes, and photodetectors). The instability and toxicity of lead-based perovskite QDs, however, are intrinsic defects that obstruct their application and commercialization. Poison is released from the lead of the unstable CsPbX3 NCs, which are generally ascribed to the labile surface, ionic character, and metastable structure. In this work, lead-free Cs2SnCl6 perovskite NCs are successfully synthesized via hot injection. Particularly, by controlling the different precursor ratios, phase transition (CsCl to Cs2SnCl6) was clearly observed from X-ray diffraction (XRD) measurements. The Cs2SnCl6 NCs exhibited a highly efficient deep-blue emission at 425 nm, with a 55nm Stokes shift and an 84nm full width at half maximum (FWHM). After doping Mn ions, the preferred formation of CsSnCl3:Mn2+ with double-wavelength emission was demonstrated based on the XRD and photoluminescence spectra. The study showed that doping synthesis should be widely used in lead-free perovskite NCs as an important strategy for next-generation solid-state lighting.

Feasibility of using Machine to Learn & Analyze Brainwave Signal for Establishing Emotional System

( Wei-tsang Huang ),( Xiang-guan Deng ),( Chun-yu Tsai ),( Chyung Ay ) 한국농업기계학회 2018 한국농업기계학회 학술발표논문집 Vol.23 No.1

「Emotion」 is a kind of psychological experience with delight, anger, sorrow, happiness and fear. This kind of experience is a reflection of ordinary people toward the attitude of objective matters. However, it is not an easy thing to see the emotion of inner heart from the appearance and change in eye or different kind of limb expression. This study is to concretize and digitize the various emotional changes of abstract conception through the brainwave analysis, and become a real-time emotional scale. As classified form the analysis and measured data, it is used as the reference standard for assessing emotion. The experiment is implemented by selecting three (3) different emotional categories of films respectively, including three (3) emotions i.e. scared, happy and inspiring emotions and editing as the visual samples, and let the subjects wearing the electroencephalograph (EEG) instrument of Emotiv EPOC with 14 channels on their heads, and use the Python program for writing to perform the measurement together with the EGG measuring system, after transforming the measured brainwave through the fast Fourier, record the energy change of each band, access the experiment data, find out the change of each band at the regular mood swings, derive the emotional conversion formula for quantification, classify the size of emotional changes, and then make the real-time scale of emotion to show the results. In addition, when the different machine learns and analyzes the test samples, it is found when analyze the T7 channel of brainwave detecting point, the success rate of highest test sample is approximately 58%, the success rate of other channels is over 50%, the more suitable operation mode will be found out in the future, and allow the emotional recognition technology can be used in the life.

Design and Performance Modeling of an Efficient Remote Collaboration System

Chun-Yi Tsai,Wei-Lung Huang 보안공학연구지원센터 2015 International Journal of Grid and Distributed Comp Vol.8 No.4

Remote Desktop Protocol (RDP) provides remote login and desktop control capabilities that enable a client to completely control and access a remote server. The protocol is implemented by Microsoft Corporation based on ITU-T T.120 family protocols. The major advantage distinguishing the RDP from other remote desktop schemes, such as the frame-buffer approach, is that the protocol is based on preferably sending graphic device interface (GDI) information from a server, instead of full bitmap images. This paper proposes a remote collaboration system based, designed and implemented on RDP architecture. The proposed system consisting of a cross control scheme with bus re-encryption architecture provides the full capabilities of all remote collaborations, such as remote control, application access and sharing among all RDP clients. In addition, it provides the RDP packets recording and replaying scheme to for replaying all of the detailed tracks through which a multiparty conference traveled. The performance modeling and evaluation show the practicality of the system and that it offers a scalable model for modern cloud services.