Molecular characterization of virulence markers, antimicrobial resistance properties and integron gene cassettes in Aeromonas spp. isolated from popular ornamental fish species in Korea

Sabrina, Hossain 충북대학교 2019 국내박사

수중에 존재하는 조건성 병원균인 에로모나스균(Aeromonas spp.)은 관상용 어류에 있어서 여러 질병과 관련된 세균으로 알려져 있다. 본 연구에서는 국내에서 사육되는 금붕어, 제브라 피쉬와 구피 3종의 관상어류로부터 분리 된 총 160 개의 에로모나스균을 대상으로 이들 균주의 병원성 및 항생제 내성 인자에 대하여 분자 생물학적 특성에 관해 연구하였다. 금붕어로부터 분리된 65균주를 gyrB 시퀀싱을 이용하여 A. hydrophila (n = 30), A. veronii (n = 32) 그리고 A. caviae (n = 3)을 동정하였으며, 이들 균종은 계통 발생적으로 별개의 군집을 형성하고 있음을 확인하였다. 또한 표현형 병원성 인자인 DNase, lipase, gelatinase, caseinase, biofilm 및 β-hemolysis가 54 (83.08 %), 47 (72.31 %), 33 (50.08 %), 32 (49.23 %), 29 (44.62 %) 및 28 (43.08 %)의 균주로부터 검출되었다. 병원성 유전자는 분리 균주로부터 다음 비율로 중합효소연쇄 반응 (PCR) 분석에 의해 검출되었다; aer (87.69%), ahyB (86.15%), fla (83.08%), ascV (67.69%), ser (67.69%), act (66.15%), ast (61.54%), alt (56.92%), gcaT (53.85%), 및 lip (49.23%). 항생제 감수성 검사에서 아목시실린 (100.00%), 날리딕신산 (100.00%), 암피실린 (98.46%), 테트라사이클린 (92.31%), 리팜피신 (86.15%), 세팔로틴 (61.54%)이 대부분의 분리 균주에 있어서 저항성이 있었으며, 그 밖의 분리 균주는 여러 항균제에 저항성을 나타내었으며 (항생제 ≤3등급에 내성), 다중 항균제내성지수 (MAR)는 0.22 ~ 0.56이었다. 또한 분리 균주에서 항생제 내성 유전자인qnrS (73.85%), qnrB (26.15%), aac(6’)-Ib-cr (4.60%), tetA (16.92%), tetE (21.54%), aac(6’)-Ib (29.23%) 및 aphAI-IAB (7.69%)이 발견되었다. IntI 유전자는 64.62 %의 균주에서 발견되었으며, 4개의 class 1 integron gene cassette profile (불완전 dfrA1, catB3-aadA1, dfrA1-orfC 및 qacE2-orfD)이 확인되었다. 제브라 피쉬에서는 A. veronii biovar veronii (n=26), A. veronii biovar sobria (n=3), Aeromonas hydrophila (n=8), A. caviae (n=3), A. enteropelogenes (n=2) 및 A. dhakensis (n=1)의 43균주가 분리, 동정되었으며, 이들 균종은 계통 발생적으로 별개의 군집을 형성하고 있음을 확인하였다. Lipase, biofilm, DNase, gelatinase, caseinase 및 β-hemolysis가 34 (79.07%), 33 (74.74%), 30 (69.77%), 25 (58.14%), 22 (51.18%) 및 21 (48.84%) 균주로부터 검출되었다. 병원성 유전자는 각 균주로부터 다음과 같은 비율로 검출되었다; aer (86.05%), hlyA (83.72%), gcaT (83.72%), lip (72.09%), act (67.44%), fla (65.12%), ascV (58.14%), ast (55.81%), ser (41.86%), ahyB (39.53%) 및 alt (25.58%). 항생제 감수성 검사에서는 아목시실린 (100.00%), 날리딕신 산 (100.00%), 옥시테트라사이클린 (100.00%), 암피실린 (93.02%), 테트라사이클린 (74.42%), 리팜피신 (67.44%) 및 이미페넴 (65.15%)이 대부분의 분리 균주에서 저항성을 나타내었으며, 모든 분리 균주는 한 개 이상의 항균제에 대해 내성을 나타내었다 (항생제 ≤ 3 등급에 내성). 그리고 다중 항균제 내성지수 (MAR)는 0.22 ~ 0.50이었다. 또한 이들 균주로부터 다음과 같은 항생제 내성 유전자tetA (67.44%), qnrS (53.49%), tetB (30.23%), tetE (30.23%), qnrB (23.26%) 및 aac(6’)-Ib-cr (4.65%)가 검출되었다. Class 1 integrase (IntI1) 유전자가 46.51 %의 균주에서 발견되었으며, 두 가지 유형의 class 1 integron gene cassette profile (qacG-aadA6-qacG 및 drfA1)이 확인되었다. 구피에서 분리된 52균주는 각각 A. veronii (n=34), A. dhakensis (n=10), A. hydrophila (n=3), A. caviae (n=3) and A. enteropelogenes (n=2)로 동정되었다. 표현형 병원성 유전자인 β-hemolysis, biofilm, caseinase, DNase, gelatinase 및 lipase가 28 (53.85%), 33 (63.46%), 28 (53.85%), 42 (80.77%), 37 (71.15%) 및 42 (80.77%)의 비율로 분리균주에서 검출되었다. 또한 병원성 유전자도 다음 비율로 분리균주로부터 검출되었다; act (84.61%), hly (80.77%), aer (73.08%), lip (73.08%), gcaT (73.08%), ascV (53.83%), ahyB (53.83%) fla (51.92%), alt (48.07%), ast (36.54%) 및 ser (34.62%). 항생제 감수성 시험에 있어서 아목시실린, 날리딕산 및 옥시테트라사이클린에 대한 내성은 모든 균주 (100%)에서 나타났으며, 분리 균주의 92.30%, 76.92%, 71.15%, 51.92%, 51.92% 및 50.00%가 각각 암피실린, 리팜피신, 이미페넴, 세팔로틴, 테트라사이클린 및 트리메소프림에 대해 저항성을 나타내었으며, 모든 분리 균주는 한 개 이상의 항균제에 대해 저항성을 보였다 (항생제 ≤ 3 등급에 내성). 그리고 분리 균주의 다중 항균제 내성지수 (MAR)는 0.22 ~ 0.67이었다. 또한 분리 균주에서 항생제 내성 유전자tetA (65.39%), qnrS (26.92%), tetE (25.00%) and tetB (15.38%), qnrB (17.31%), aphaAI-IAB (7.69%) 및 aac(6’)-Ib (3.84%)이 검출되었다. 또한 IntI 유전자도 분리 균주의 36.54%에서 검출되었고, 4종의 class 1 integron gene cassette profile (aadA2, qacE2-orfD, aadA2-catB2 및 dfrA12-aadA2)이 확인되었다. 상기의 결과로부터 국내에서 판매 사육되고 있는 금붕어와 같은 관상용 어류가 병원성이 있고, 여러 항균제에 내성을 갖는 Aeromonas spp.의 보균체 (reservoir) 역할을 하고 있음이 판명되었으며, 또한 이들 균주가 다양한 병원성 및 항균제 내성 유전자 뿐만 아니라 class 1 integron gene cassette 를 포함하고 있음을 밝혀 냈다. Aeromonas spp. are opportunistic pathogenic bacteria associated with a multitude of diseases in ornamental fish. In this study, a total of 160 aeromonads were studied which were isolated from popular ornamental fish species including goldfish, zebrafish and guppy in Korea. The goldfish-borne isolates were identified as A. hydrophila (n = 30), A. veronii (n = 32) and A. caviae (n = 3) by gyrB gene sequencing and each of the species formed distinct clusters in the neighbor-joining phylogenetic tree. The phenotypic virulence factors such as DNase, lipase, gelatinase, caseinase, biofilm and β-hemolysis activity were observed in 54 (83.08%), 47 (72.31%), 33 (50.08%), 32 (49.23%), 29 (44.62%) and 28 (43.08%) isolates. The virulence-related genes were detected by polymerase chain reaction (PCR) assay in the following proportions: aer (87.69%), ahyB (86.15%), fla (83.08%), ascV (67.69%), ser (67.69%), act (66.15%), ast (61.54%), alt (56.92%), gcaT (53.85%), and lip (49.23%). In antimicrobial susceptibility testing, most of the isolates were resistant to amoxicillin (100.00%), nalidixic acid (100.00%), ampicillin (98.46%), tetracycline (92.31%), rifampicin (86.15%), and cephalothin (61.54%) and each of the isolates showed multiple antimicrobial resistance phenotype (resistant to ≤ 3 classes of antimicrobials). The multiple antimicrobial resistance (MAR) index values ranged from 0.22-0.56. The antimicrobial resistance genes, qnrS (73.85%) qnrB (26.15%), aac(6’)-Ib-cr (4.60%), tetA (16.92%), tetE (21.54%), aac(6’)-Ib (29.23%) and aphAI-IAB (7.69%) were detected among the isolates. The IntI gene was found in 64.62% isolates, and four class 1 integron gene cassette profiles (incomplete dfrA1, catB3-aadA1, dfrA1-orfC, and qacE2-orfD) were identified. The A. veronii biovar veronii (n=26), A. veronii biovar sobria (n=3), Aeromonas hydrophila (n=8), A. caviae (n=3), A. enteropelogenes (n=2) and A. dhakensis ( n=1) were isolated from zebrafish and each species formed a distinct group in a neighbor-joining phylogenetic tree. The lipase production, biofilm formation, DNase activity, gelatinase production, caseinase production and β-hemolysis were phenotypically observed in 34 (79.07%), 33 (74.74%), 30 (69.77%), 25 (58.14%), 22 (51.18%) and 21 (48.84%) isolates. The virulence genes were detected in following frequencies- aer (86.05%), hlyA (83.72%), gcaT (83.72%), lip (72.09%), act (67.44%), fla (65.12%), ascV (58.14%), ast (55.81%), ser (41.86%), ahyB (39.53%) and alt (25.58%). The isolates were resistant to amoxicillin (100.00%), nalidixic acid (100.00%), oxytetracycline (100.00%), ampicillin (93.02%), tetracycline (74.42%), rifampicin (67.44%) and imipenem (65.15%), and every isolate displayed multi-drug resistance (resistant to ≤ 3 classes of antimicrobials). The multiple antimicrobial resistance (MAR) index values ranged from 0.22-0.50 among the isolates. The tetA (67.44%), qnrS (53.49%), tetB (30.23%), tetE (30.23%), qnrB (23.26%) and aac(6’)-Ib-cr (4.65%) genes were detected. Class 1 integrase (IntI1) gene was found in 46.51% of the isolates and two types of class 1 integron gene cassette profiles (qacG-aadA6-qacG and drfA1) were identified. The guppy-borne isolates were identified as A. veronii (n=34), A. dhakensis (n=10), A. hydrophila (n=3), A. caviae (n=3) and A. enteropelogenes (n=2). The phenotypic virulence factors such as β-hemolysis, biofilm, caseinase, DNase, gelatinase and lipase formation were observed in 28 (53.85%), 33 (63.46%), 28 (53.85%), 42 (80.77%), 37 (71.15%) and 42 (80.77%) isolates. The virulence genes such as act (84.61%), hly (80.77%), aer (73.08%), lip (73.08%), gcaT (73.08%), ascV (53.83%), ahyB (53.83%) fla (51.92%), alt (48.07%), ast (36.54%) and ser (34.62%) were detected, respectively. The resistance to amoxicillin, nalidixic acid and oxytetracycline was 100% among the tested isolates. 92.30%, 76.92%, 71.15%, 51.92%, 51.92% and 50.00% of the isolates were resistant to ampicillin, rifampicin, imipenem, cephalothin, tetracycline and trimethoprim, respectively and every isolate displayed multi-drug resistance (resistant to ≤ 3 classes of antimicrobials). The multiple antimicrobial resistance (MAR) index values ranged from 0.28 to 0.67 among the isolates. The occurrence of tetA (65.39%), qnrS (26.92%), tetE (25.00%) and tetB (15.38%), qnrB (17.31%), aphaAI-IAB (7.69%) and aac(6’)-Ib (3.84%) were observed in the isolates. The IntI gene was positive for 36.54% of the isolates and four class 1 integron gene cassette profiles (aadA2, qacE2-orfD, aadA2-catB2 and dfrA12-aadA2) were identified. These data suggest that ornamental fish can serve as a reservoir of virulent multi-drug resistant Aeromonas spp. which comprise different virulent and antimicrobial resistance genes as well as class 1 integron gene cassettes.

INTER-SPECIFIC RELATIONSHIP OF TWO SYMPATRIC THRIPS, Frankliniella occidentalis (PERGANDE) AND Frankliniella intonsa (TRYBOM) (THYSANOPTERA: THRIPIDAE)

MOHAMMAD MOSHAROF HOSSAIN BHUYAIN 안동대학교 2015 국내박사

GENERAL ABSTRACT INTER-SPECIFIC RELATIONSHIP OF TWO SYMPATRIC THRIPS, Frankliniella occidentalis (PERGANDE) AND Frankliniella intonsa (TRYBOM) (THYSANOPTERA: THRIPIDAE) MOHAMMAD MOSHAROF HOSSAIN BHUYAIN MAJOR IN APPLIED ENTOMOLOGY DEPARTMENT OF BIORESOURCE SCIENCES ANDONG NATIONAL UNIVERSITY The western flower thrips, Frankliniella occidentalis (Pergande) (Thysanoptera: Thripidae) and the garden thrips, F. intonsa (Trybom), are flower-inhabiting thrips infesting many crops of agricultural importance. By direct damage as well as tospovirus transmission, thrips cause significant economic loss worldwide. Asymmetrical seasonal occurrence and distribution pattern of F. intonsa dominate over F. occidentalis was found regionally in Andong in recent years. I surveyed dynamics of thrips populations on greenhouse strawberry by installing sticky cards and on field pepper by destructive flowers sampling during the season of 2012, 2013, and 2014 at Songcheon in Andong. F. intonsa was 5.8, 9.4, and 120.5 times more abundant on pepper flowers while 34.7, 25.3 and 14.4 times in greenhouse strawberry than F. occidentalis in 2012, 2013 and 2014, respectively. In the second experiment, I assessed relative insecticide susceptibility of both species to explain the differential occurrence pattern in the field. Acute toxicity of F. occidentalis and F. intonsa to different pesticides was measured in laboratory condition by using direct spray, residual, and oral assays. Results showed that both species were highly sensitive to chlorfenapyr, and its mortality reached 100% within 12 h irrespective of assay methods in F. occidentalis, but the time to death was shorter (9 h) in direct spray and residual assay in F. intonsa. As another hypothesis to explain the differential occurrence pattern, I evaluated interactive influence of temperatures (20, 25, and 30°C) and relative humidities [Low (50-55%), Medium (70-75%), and High (90-95%)] on larval, pupal, and adult stage of two thrips species in the third chapter. In larval survivorship experiment, considering all temperatures and RHs, F. intonsa larvae survived shorter period of time than F. occidentalis. The median longevity time values of F. occidentalis larvae were higher than F. intonsa at all the tested temperatures and RHs. Survival period of larvae of both thrips generally increased with increasing RH expect 20°C at high humid condition for F. intonsa. In pupal mortality and its’ duration of both species decreased as the relative humidity (RH) increased in all the temperatures tested, while it was significantly shorter in F. intonsa (pupal period). In adult stage, F. intonsa survived better than F. occidentalis and exhibited 2.5 and 2.4 times longer lifespan whereas F. occidentalis extended the lifespan by 1.8 and 1.6 times as RH increased from low to high at 20 and 25°C, respectively. In the fourth chapter, I evaluated the existence of interspecific competition between the two species in the laboratory by providing honey or pollen as food source. Five behavioral parameters were recorded: confronting, wandering, resting, food guarding, and feeding. Results showed that adult F. intonsa spends more time in guarding and feeding on both honey and pollen than F. occidentalis, although such a pattern was not observed in the larval stage. In the presence of opponent, guarding and feeding time (in seconds) of adult F. intonsa was 8.45 and 1.45 times longer on honey and 42.76 and 1.33 times longer on pollen than F. occidentalis, respectively. From these results I conclude that both interference and exploitation competition occur between F. occidentalis and F. intonsa and that F. intonsa is better at feeding and guarding food sources, especially in the adult stage. As continuation of the competition experiment, effect of the competition on adult reproduction and longevity of the two thrips species was assessed in leaf cage and micro-tube arena, respectively in the fifth chapter. The number of reproduced progeny was reduced by 79% for F. occidentalis and only 43% for F. intonsa in competition condition. The mean longevity was also reduced greater in F. occidentalis (by 41%) than in F. intonsa (23%). With kidney bean leaves, area of feeding marks on bean leaf made by both F. intonsa and F. occidentalis (competition condition) was not significantly different from those made by each species alone (no-competition condition). These results verify that interference competition exists between the two species Therefore, interspecific competition might affect field occurrence pattern of the two thrips species. Key words: Frankliniella occidentalis, Frankliniella intonsa, asymmetrical seasonal occurrence, population dynamics, competition, climate change, body water retention ability, anthophilous thrips, Thripidae, clip cage design, food guarding, feeding, invasive species.

Legal Comparison of Corporate Governance Practices between Bangladesh and South Korea

Hossain Mohammad Mosarraf 충북대학교 대학원 2020 국내박사

With focusing on corporate governance code, commercial act and company act; this study has delineated, examined and compared the nature of practices of corporate governance of Bangladesh and South Korea. It has also identified the critical issues, which hider the systematic corporate governance in developing countries. By studying corporate literatures, this study has further described the different frameworks of corporate governance along with scrutinization of different models and it appropriateness for Bangladesh and South Korea. As we know that many corporate governance practices and reforms in emerging markets are based on corporate governance systems of developed markets. In Bangladesh, most corporate governance rules, practices and code system are developed by following the British practices, and they are largely recommended through voluntary codes and company act allowing companies to have considerable deceleration in their implementation. The ownership structure in Bangladeshi companies is characterized by a prevalence of family ownership and concentrated ownership resembling the characteristics of the relationship-based model of corporate governance. Therefore the application of corporate governance practices originating in Anglo-Saxon countries, such as Britain, which typically experience dispersed ownership, creates many incompatible issues affecting overall efficiency of corporate governance system in Bangladesh. So this study has showed the ins and outs of Bangladesh corporate codes, company acts, and Bangladesh socio-economic culture that influence corporate practices in Bangladesh. And, in the case of South Korean corporate governance practices; it is seen that corporations pursue bank-centered financing system by devising internal market transactions, especially payment guarantees and collateral provisions. And structurally Korean corporate governance practices are also family-held, hierarchical and centralized. In term of family ownership; blood relations are of utmost importance in Korean society, so the family name carries more weight and thus children can become the primary heir. But even though it is utmost centralized, it is decently seen that Korean corporate governance system has been particularly effective in generating economic development in the form of a non western model of state-driven economic growth. By focusing on its code provisons, commercial act and cultural phenomena; this study has shown South Korean corporate practices; and how it works in internal mechanism. In Particular, this study has found out some kernel factors that impede the systematic corporate governance practices in Bangladesh. As crucial factors of impediment, this study has pointed out that, fragile and nominal regulations; unawareness and ignorance about corporate; immoral influences by bureaucrat; unethical political pressure, illegal dominance by controlling shareholders, influence of family-held business, non existence of pressure group, lack of compliance on corporate code, inappropriate contents in corporate code; and lack of proper infrastructure are commonly seen in Bangladesh corporate governance practices. And, for the purpose of getting good corporate governance practices; this study has postulated some recommendations, such as strong management; proper board structure; legal, moral & ethical decision; integrated audit report; maintaining and timely disclosing financial transactions; honoring shareholders' rights; systematic risk management; and finally resonable & satisfactory remuneration for corporate managers. By using self-administered scrutinization in corporate governance practices, corporate governance code contents, commercial & company act; this study has aimed at divulging out the full gamut of corporate regulatory practices for enhancing the governance scenario in Bangladesh and Republic of Korea. This research findings proves impediments with clarity that corporations in both countries are, in principle, moderately compliant with the provisions of code & commercial act. The hypothetical or suppositional or speculative findings of this study has been with its theoretical perspective. As this study has described the full gamut of corporate system so, it can help as catalysts to implement better policies for good governance practices; recognize the areas of non-compliance, understand the why, how and where corporations are falling behind from proper governance practices. Finally I hope, this study can certainly contribute to lessen the interstice of lacks of corporate academic literatures in Bangladesh and South Korea.

To Build a Front-End Web Application Which JavaScript Framework is Trending Nowadays

Hossain Mohammad Faruque 울산대학교 대학원 2022 국내석사

It is the year 2022, and we have already reached the last decade of the twenty-first century. This era is changing so rapidly, especially with the blessing of the internet. Web development is the most crucial part of this web 3.0 era. There are over 1.6 billion websites now on the internet. It’s the fastest-growing industry in the web world. Because every small enterprise to big giant everyone needs their website to promote themselves and their products all over the world. However, It’s not easy to make a website in a day or a week, or a month. It takes time to make a smooth and user-friendly web application. But without the help of a framework, it’s quite impossible to make a dynamic web application. Because web development frameworks are a useful and most necessary tool nowadays. For front-end web applications there are so many JavaScript frameworks over the internet. In this thesis, we are talking mostly about using frameworks by a front-end developer and they are React, Angular, Vue, and Ember. All previous research is too much old which is against to now trending world and also previous research we found an only comparison between frameworks and their pros and cons. But no one thinks about what a developer chooses to build a web application. In this thought, we design this thesis theoretically and practically to survey developers' preferences. It will be helpful for those who want to come into this industry as newcomers and also experts can take an easy decision for making good web applications. So, We can say that after studying all of the above information why need this thesis to find out which web frameworks are trending nowadays and what is the next.

An edge segment based moving object detection for automated video surveillance

Hossain, Mohammad Julius 경희대학교 2008 국내박사

This dissertation presents a novel moving object detection algorithm for sequence image using edge information as segment. Various methods have been proposed to solve this problem and most of them work very well in comparatively static scenario. However, the real environment is very dynamic due to various phenomena including illumination variation, calibration error, camera motion, shadow, inter-reflection, camouflage and the temporal changes in background contents. Most of the existing methods are not much successful to address these issues effectively to ensure the robustness against these variations. The proposed method is introduced to address some of these challenges including variations in ambient lighting and background contents, movement of camera and temporal changes in the content of background. This work also investigates the suitability of the proposed edge segment based algorithm in comparison with the traditional intensity based as well as edge pixel based moving object detection methods. The proposed method works with contour based features rather than intensity based features as edge information is more robust than pixel value in illumination change and also most of the structural information of image lies in the boundaries of image contents that characterize edges. Nonetheless, variations in edge information in different frames are inevitable and a robust detection method needs to keep track of these variations to increase the accuracy of detected moving edge as well as to decrease the false detection rate. Existing edge pixel based methods treat each edge point independently without considering the continuity/neighborhood information. This individual edge pixel does not carry the shape information of image contents and it is unable to keep the track of the variations in different frames. As a result the accuracy of detection process decreases due to the incapability of handling challenges including change of ambient lighting, edge localization error and minor movement of camera. It also results scattered moving edges which are not suitable for further processing including segmentation and tracking. However, it is not possible to propose a fast solution to solve these problems without representing the edge pixels with certain data structure. To overcome these problems, the proposed method extracts edges from video frames and represents as segments using an efficiently designed edge class. An edge pixel is not handled independently rather all the points belonging to a segment are considered as a unit and are processed together. This representation helps to obtain the geometric information of edges and object shape, facilitates local transformation and incorporates knowledge into the edge segment. The shape information helps to analyze the variations of edges in different frames and these statistics are stored as knowledge with the edge segments to increase the effectiveness of further processes including reference initialization and update, edge matching and tracking, moving object segmentation, classification and recognition. The proposed work presents an efficient approach for edge segment based background initialization which is capable of being initialized while moving objects exist in the scene. Each segment in the background holds certain knowledge about its variations in different frames. This knowledge helps to determine the amount of allowable variation by each of the segments in the detection process. It also helps to select different level of flexibilities/search window for different segments that reduces the possibility of false matching and increases the accuracy of detection. The changes in the constituents of background scene are dealt by maintaining a separate background component. It takes care of background edge segments that are missed during the background initialization process and also the edges of moving object that work as a background temporarily. A Hausdorff distance based flexible edge matching is proposed utilizing integer approximations with Chamfer ¾ distance having lower relative error rate than other approximations. Decision on matching is performed on a complete edge segment instead of an individual pixel which reduces the occurrence of scattered moving edge pixels in the detection result. The matching method can tolerate the edge localization error by allowing suitable flexibilities, automatically determined from the statistics gathered during the background initialization process. Overall matching process is also very fast and suitable for real time detection of moving edges, as distance transformation is performed in linear time with integer approximations. With the help of segment based representation of edges, the proposed method is able to adjust the minor movement of camera with a modest complexity. Thus, it can effectively reduce the risk of false alarm due to illumination change, edge localization error and camera motion while maintaining the high sensitivity to the presence of moving object. Detected moving edges are utilized to determine the region of interest (ROI) of moving object, watershed algorithm applied on ROI of current frame and it is followed by the background segment removal procedure to extract the region of moving object. Segmented boundary of moving object is more accurate due to the application of watershed algorithm. Experimental results with real image sequences are presented and evaluated in both qualitative and quantitative perspectives. The evaluation records reflect that the proposed edge segment based moving object detection method is suitable for automated video surveillance applications in various monitoring systems. Region information of extracted moving object can be also utilized in various applications including content based video coding and multimedia communication where extraction of motion information is necessary. Moreover, with the introduction of edge segment based detection method, a good number of research issues have been explored in this area including segment based tracking and motion analysis.

Integration Testing Approach based on Usage Patterns of Global Variables in Embedded Software : 임베디드 소프트웨어의 전역 변수 사용 패턴 기반의 통합 테스트 접근법

Hossain, Muhammad Iqbal 경북대학교 대학원 2012 국내석사

In embedded software global variables are used extensively to pass information between sections of code that do not share a caller/callee relation. It is a common practice to use global variables to read or modify throughout the source code. In integration testing the dependencies of global variables among the functions makes the testing procedure more complicated. Moreover, global variables potentially cause many issues such as lack of access control and implicit coupling, which make the program more difficult to test. Moreover for large number of integration module and its associated global variables introduce the problem of scalability. Due to these side effects developers sometimes ignore global variables which mark a big question on the reliability of the system. In this thesis we propose and implement an automated integration testing approach using the usages pattern of global variables and generate scalable test cases according to feasible test sequences. In our approach different graph analysis techniques are used with definition-use of global variables. Basically the procedure is divided into two parts, test model generation and test case generation from test model. In first part, extension of call graph and control flow graph is used for generating a test model. Graph reduction criterion is introduced to minimize number of test path. The reduced graphs are combined in a single graph to generate the test model. In second part we generate valid DU pair and sequential integration module path to generate test paths from the test model. Finally symbolic execution technique is used to generate the test cases from the test paths. This approach is automated, cost effective and the case study indicates high scalability on generating test cases. 임베디드 소프트웨어에서 전역 변수는 함수 호출 관계가 없은 소스 코드 간에 정보를 전달하기 위해 널리 사용된다. 통합 테스팅 단계에서 존재하는 전역 변수의 의존성은 테스팅을 더욱 복잡하게 만든다. 전역 변수는 접근 제어 결여나 암시적 의존성과 같은 문제를 야기한다. 이러한 부작용으로 인해 개발자는 시스템의 신뢰성에 문제를 일으키는 전역 변수를 사용하지 않기도 한다. 본 논문에서는 전역 변수의 사용 패턴을 이용하여 자동화된 통합 테스트 방법을 제안하고 실현 가능한 테스트 시퀀스에 따라 확장 가능한 테스트 케이스를 생성한다. 제안된 접근법은 전역 변수의 정의-사용 패턴을 이용한 그래프 분석을 통해 테스팅 비용과 복잡도를 줄이고 전체적인 과정을 단순화하는 축약 기준을 제시한다. 이 과정은 테스팅 모델 생성과 테스트 케이스 생성으로 구성된다. 테스트 모델 생성을 위해 확장된 호출 그래프와 제어 흐름 그래프가 이용된다. 그래프 축약 기준은 테스트 경로의 수를 최소화하며, 축약된 그래프는 테스트 모델을 생성하기 위해 하나의 그래프로 통합된다. 테스트 케이스 생성 단계에서는 테스트 모델로부터 테스트 경로를 생성하기 위해 유효한 정의-사용 쌍과 순차적인 통합 모듈 경로를 생성한 후, 심벌릭 실행을 이용해 테스트 경로로부터 테스트 케이스를 생성한다.

Probabilistic Risk Analysis for the Underground Box Structure Construction Based on Artificial Neural Network (ANN) Technique

Hossain, Md Sharif 한국교통대학교 일반대학원 2019 국내석사

Congestion from population and traffic has rapidly increased in urban areas, which resulted in the construction of underground transportation systems. The construction of tunnels in urban areas may cause any sudden and substantial ground displacement, which in turn may significantly distort and damage overlying adjacent buildings, roads, and services. To minimize the possible detrimental effects of ground settlement, an essential method needs to be developed to predict the maximum ground settlement as well as the risk potential during the installation of underground box structure. While most of the conventional methods (analytical methods, empirical methods) are predicting and estimating ground settlement for the circular tunnel structure. The aim of this study is to establish a probability risk model based on Artificial Neural Network (ANN) technique to assess the risk potential of underground box structure installation. Artificial Neural Network (ANN) is a complex mathematical model that is inspired by the functioning of the human brain and nerve system, which are widely used in the field of nonlinear systems and system identification. A total of 160 underground box structure models were generated from finite element analysis (using PLAXIS 2D 8.6), taking into account ground conditions, structure sizes and shapes, traffic loads, existing structures and corresponding maximum surface settlements, building settlement, building deflection ratio, and pipe pull-out distortion. A MATLAB-based multi-layer feed forward back propagation optimum ANN model is developed with six input parameters including 1) soil modulus of elasticity (E), 2) soil cohesion (c),3) box width (B), 4) overcut size (s), 5) building distance (d1), and 6) pipe distance (d2), and one hidden layer with 10 neurons. Furthermore, ANN model was expanded, trained, and tested with consideration of the variability of input parameters and limit state functions for individual components (road/railroad settlements, building settlement, building deflection ratio, and pipe joint pull-out displacement) to predict the component/system failure probability. The performance accuracy of ANN model was verified via mean square error (MSE) and linear regression analysis results. The average positive percentage of error of the model was found to be 14.6%, which indicates a fairly accurate performance. Moreover, a sensitivity analysis and a case study were performed to verify the performance of the risk model. It was confirmed that larger the size of the box structure, the greater the risk. It was also observed that the modulus of elasticity of soil and building distance is the most sensitive factor, rather than soil cohesion and pipe distance in terms of probability of failure. In addition, the ANN risk model was able to predict the probability of failure of the real case project. Experimental programs were also performed to investigate the effect of the umbrella method, which contributes to the decrease of risk potential in a practice. A limited field test evaluation using ground-penetrating radar (GPR) along with pneumatic dynamic cone penetrometer (PDCP) was found to be promising in the assessment of risk potential at shallow depths.

Inhibition of Acyl-homoserine Lactone-mediated Quorum Sensing by Novel Phenolic Compounds of Nymphaea tetragona (Water Lily) : Nymphaea tetragona의 페놀 화합물에 의한 아실호모세린락톤-중계 쿼럼 센싱의 억제

Hossain, Akil Md 경북대학교 대학원 2016 국내박사

쿼럼 센싱(QS)이란 생존을 위해 세균이 정보를 공유하거나 유전자 발현 조절 등을 통해 세균의 정족수 밀도와 관련된 신호전달 체계이다. 이러한 신호전달 체계를 이용하여 세균은 정보를 공유하고 유전자 발현을 조절한다. 세균은 QS를 이용하여 바이오필름의 형성, 운동성, 병원성 인자 활성 같은 다양한 표현형 조절을 한다. 여러가지 병원성 인자의 활성을 통하여 숙주세포와 세균의 상호관계를 설정한다. 식물은 매우 다양하면서 구체적인 생물학적 활동들로 구성되어 있기 때문에 수천년 동안 의약소재로 사용되고 있다. 따라서 본 연구의 목적은 QS억제제를 탐색하고 개발하여 세균이 항생제 저항성을 갖는 것을 막고, 궁극적으로는 이러한 소재를 이용하여 병원성 세균 감염증를 제어하는 것이다. Nymphaea tetragona의 50% 메탄올 추출물 (NTME)에 대하여 QS 억제효과가 있는지 그리고 세균의 병원성을 효울적으로 제어할 수 있는지 조사하였다. 먼저 NTME 단독 용매 분획물질과 기존의 항생제와의 항균 효과를 본 연구에서 조사하였다. 한편, QS 억제에 가장 강력한 추출 분획 물질을 확인하였으며 활성화합물도 분석 및 확인하였다. 최종적으로 NTME에 다량 포함되어 있는 메틸 갈레이트와 피로갈롤을 포함한 5가지 페놀 화합물의 효과를 확인하였다.수련추출물은 농도에 비례하여 항균활성을 보였고 NTME는 강력한 QS 및 QS 관련 세균의 병원성 발현을 억제하였다. 그리고 본 연구에서 NTME는 in vitro나 in vivo에서 독성이 확인 되지 않았다. NTME의 아세테이트 분획물(EFNTE)은 좋은 항균 활성을 보여 주었다. S. typhimurium에 대한 EFNTE-Tylosin과 EFNTE-streptomycin의 Fractional Inhibitory Concentration Index는 각각 상승 및 상가효과를 보였다. 분획물질 중 EFNTE와 부탄올 분획물에서 확실한 QS 억제 효과를 보였다. 페놀 함유 항산화물질은 EFNTE와 부탄올 분획물의 주요 페놀화합물로 5가지 성분중에서 메틸갈레이트만이 정량적 분석에서 QS의 억제 효과를 보였다. 이 화합물은 아실호모세린락톤(AHL) 합성 및 활성이 모두 억제를 하였다. 또한 P. aeruginosa의 바이오필름 형성 및 QS 관련 병원성 발현도 억제하였다. 메틸갈레이트는 P. aeruginosa의 유전자 lasI/R, rhlI/R 및 pqsA의 발현도 억제를 하였다. 이 연구의 결과, 메틸갈레이트는 QS 조절을 통한 치료제로써 가능성을 보였다. 이러한 결과는 최초의 보고로 메틸칼레이트가 병원성 세균의 QS-제어 표현형을 억제하는 QS 조절 물질로의 AHL 합성과 AHL 활성작용을 동시에 억제한다는 기전을 구명하였다. 따라서 본 논문에서는 항생제 내성 및 병원 세균의 병원성 감소시키기 위하여 항쿼럼센싱 물질을 수련추출물에서 탐색하여 주요 성분인 메틸갈레이트를 확인하였고 이 성분에 대한 항퉈럼센싱의 기전을 분자생물학적인 방법으로 밝혀 최초의 활성기전을 구명하였다. 이 결과는 새로운 항생제 내성 획득을 억제하고 병원성 발현을 조절하는 소재로 활용이 가능할 것으로 기대된다.

Use of medicinal plants and probiotics as combination functional feed additives for non-ruminant

Hossain, Md. Elias Sunchon National Univ. 2012 국내박사

Function of Sinapic acid on induction of thermogenic signature in brown adipocyte

Hossain, Monir Soonchunhyang University Graduate School 2019 국내석사

Obesity is now a worldwide epidemic and categorizes as one of the chronic diseases by World Health Organization in the 21st century. Lipid accumulation in white adipose tissue is the key contributor to the obesity and orchestrates numerous metabolic health problems such as type 2 diabetes, hypertension, atherosclerosis, and cancer. Nonetheless, the prevention and treatment of obesity are still inadequate. Recently, scientists found that brown adipose tissue (BAT) in adult humans has functions that are diametrically opposite to those of white adipose tissue and that BAT holds promise for a new strategy to counteract obesity. Findings have revealed that increase in the mass of brown adipose tissue or activating its function can reduce body mass in rodent models. While white adipose tissue (WAT) contributes to obesity by storing triglycerides, brown adipose tissue (BAT) dissipates energy in the form of heat. In this study, we evaluated the potential of sinapic acid (SA) to promote the thermogenic program and lipolysis in BAT. SA treatment of brown adipocytes induced the expression of brown-adipocyte activation–related genes such as Ucp1, Pgc-1α, and Prdm16 and mitochondrial biogenesis-related genes including Nrf1. Additionally, we observed that SA treatment can reduce the expression of adipogenesis inhibiting factors PREF, WNT10a, and WNT10b. Furthermore, structural analysis and western blot revealed that SA competitively inhibits pan-PKA inhibitor, H89 and can upregulate protein kinase A (PKA) phosphorylation. SA binds to the adenosine triphosphate (ATP) site on the PKA catalytic subunit where H89 binds specifically. It is confirmed by PKA-cat-α1 gene–silencing experiments that SA can activate the thermogenic program via a mechanism involving PKA and cyclic AMP response element–binding protein (CREB) signaling. Moreover, SA treatment promotes lipolysis via a PKA/p38-mediated pathway. Our findings may allow us to open a new avenue of strategies against obesity and need further investigation.