Artificial Intelligence-Based Breast Nodule Segmentation Using Multi-Scale Images and Convolutional Network

Quoc Tuan Hoang,Xuan Hien Pham,Anh Vu Le,Trung Thanh Bui 한국인터넷정보학회 2023 KSII Transactions on Internet and Information Syst Vol.17 No.3

Diagnosing breast diseases using ultrasound (US) images remains challenging because it is time-consuming and requires expert radiologist knowledge. As a result, the diagnostic performance is significantly biased. To assist radiologists in this process, computer-aided diagnosis (CAD) systems have been developed and used in practice. This type of system is used not only to assist radiologists in examining breast ultrasound images (BUS) but also to ensure the effectiveness of the diagnostic process. In this study, we propose a new approach for breast lesion localization and segmentation using a multi-scale pyramid of the ultrasound image of a breast organ and a convolutional semantic segmentation network. Unlike previous studies that used only a deep detection/segmentation neural network on a single breast ultrasound image, we propose to use multiple images generated from an input image at different scales for the localization and segmentation process. By combining the localization/segmentation results obtained from the input image at different scales, the system performance was enhanced compared with that of the previous studies. The experimental results with two public datasets confirmed the effectiveness of the proposed approach by producing superior localization/segmentation results compared with those obtained in previous studies.

Commercialization of Scientific Research Activity Results at Vietnam's Universities: Current Challenges and Solutions

Quoc Cuong Nguyen,Hoang Tuan Nguyen The Institute of Internet 2023 International journal of advanced smart convergenc Vol.12 No.2

Commercialization of research results plays an important role in the socio-economic development of a country based on science, technology and innovation. However, in recent years, the commercialization of research results by universities in Vietnam remains ineffective because scientists often carry out research on issues they find interesting, and don't listen to enterprises to find out what the latter need. In the other hand, the process of commercializing research results in universities encounters difficulties, including the issue of establishing intellectual property rights the limited quality of research results, and the limited access to capital resources to develop inventions. This paper focuses on analyzing the current challenges of commercialization of research results in universities and propose solutionsto promote commercialization of research results from the universities, suitable to the conditions of Vietnam.

The Innovation Activities of Vietnamese Enterprises: Current Status and Solutions

Quoc Cuong Nguyen,Hoang Tuan Nguyen,HyukDong Kwon 국제문화기술진흥원 2023 International Journal of Advanced Culture Technolo Vol.11 No.2

Innovation has been considered not only an important factor for creating and maintaining the competitiveness of nations but also a key determinant of enterprises performance in over the world. Innovation is the process by which an enterprise develops new products, services, processes or management systems to meet requirements due to the change of business environment, technology. The results of implementing innovation increase capacity of the business, thereby helping the business to develop sustainably in the face of market trends and competition from competitors. This paper aims to explore the current status of innovation activities as well as proposes solutions to develop innovation activities for Vietnamese enterprises. The proposed solutions in this study paper create a premise to improve the competitiveness and sustainable growth of enterprises in the context of deeper and more comprehensive international economic integration.

Back-stepping Approach for Rolling Motion Control of an Under-actuated Two-wheel Spherical Robot

Hoang Quoc Dong,Soon-Geul Lee,Soo Ho Woo,Tuan-Anh Le 제어로봇시스템학회 2020 제어로봇시스템학회 국제학술대회 논문집 Vol.2020 No.10

Spherical robots are the mobile robots with the locomoting by displacing its centre of gravity to generate torque and rotate itself. Therefore, the angle of the main body inside the robot determines the position and posture of the robot. There is only one contact point between the robot and the ground, and the inappropriate control strategy can generate the sizeable angular amplitude of the main body. As a result, the stable movement of the robot cannot be satisfied along with the appeared vibrations. This problem significantly impacts the tracking control quality and creates the clumsy gestures of the robot. In this research, an under-actuated dynamic model-based back-stepping control focusing on the rolling motion is developed and applied for a designed two-wheel spherical robot. With the provided closed-loop control law, both the precision and stability of the robot’s movement are guaranteed. The entire work’s efficiency is investigated by the experimental results.

Design of DSTBC and HARQ schemes for Turbo-coded Cooperative Wireless Relay Networks over Frequency Selective Fading Channels

Quoc-Tuan Vien,Dao-Tan-Hoang Nguyen,Le-Nam Tran,Een-Kee Hong 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7

This paper proposes a new distributed space-time block code (DSTBC) over frequency selective fading channels in amplify-and-forward (AF) turbo-coded cooperative wireless relay networks with hybrid automatic repeat request (HARQ) combining schemes. The proposed DSTBC and HARQ schemes are designed to achieve spatial diversity gain and low decoding complexity. These goals are obtained by precoding the received signal at the relay to form the conventional space-time block codes. In addition, the coding gain is combined with the diversity gain to enhance the performance of DSTBC by adding turbo coding at the source and iterative decoding at the destination. Together with turbo coding, two fundamental HARQ combining schemes are considered in our design, including Chase combining (CC) and incremental redundancy (IR).

Adaptive Neural Network Sliding Mode Control for an Unmanned Surface Vessels

Tuan Pham,Jonggyu Park,Soon-Geul Lee,Quoc-Dong Hoang 제어로봇시스템학회 2020 제어로봇시스템학회 국제학술대회 논문집 Vol.2020 No.10

Autonomous vessels have been given interesting research topics in both the maritime area and the control engineering field with increasingly numerous applications. The practical system is impacted by a vast range of uncertainties, including the high nonlinearities of the dynamic model, the variable working environments, and the influences of ocean currents, waves, and wind. These factors give the big challenges for control, especially precise trajectory tracking control. In this research, a new stage adaptive neural network back-stepping control based on sliding-mode control for the Fossen’s ship model is proposed. By using the artificial neural network to approximate the composite time-varying disturbance function, the controller can solve the problems of accurate control tracking the displacement of the surface vessel following the convergence of the Lyapunov function. The numerical simulation results illustrate the effectiveness of the designed control law.

Research and Calculate 29/34-Seat Passenger Cars to Ensure Safety for Occupants in the Event of a Collision According to ECE R94 Standards

Vu Hoang Phuong,Nguyen Cong Thanh,Nguyen Quoc Tuan,Ta Hong Thanh Tu 한국인터넷방송통신학회 2023 International Journal of Internet, Broadcasting an Vol.15 No.1

In recent years, there are so many serious crashes involving coaches, especially the frontal collision occupies 40% of the front of the vehicle, Frontal collisions account for 100% of the front of the vehicle affecting the driver and side-impact collisions that injure the person in the vehicle. Therefore, the research into improving and optimizing the structure is necessary for risk of injury for passengers in frontal accidents. In this paper, we have designed a Shock absorber that can absorb collision energy. Research using HYPERMESH software. to build the finite element model and calculate the meshing to suit the mesh size of 5mm. apply LSDYNA software to calculate structural strength. In the study, for a vehicle to collide with a hard obstacle occupying 100% of the head of the vehicle. Then, the experimental design method, Minitab is used for find the structural parameters in the design. Improvement results showed that the acceleration of the impact on passengers and the driver is decreased by 55,17%. The mass of texture improvements is reduced by 11%, according to the requirements of European Standards ECE R94.

Inverse MR and Dual-AMR Phenomena in Co/CoO/Ag/Co Sandwiches

Nguyen Anh Tuan,Luong Van Su,Hoang Quoc Khanh,Nguyen Anh Tue,Nguyen Thi Luyen 한국물리학회 2018 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.72 No.7

Band-form Co/CoO/Ag/Co sandwiches were prepared using radio-frequency magnetron sputtering. The sandwiches had rather thick Co (tCO = 63 nm) and Ag (tAg = 2 ÷ 65 nm) layers and a super-thin paramagnetic cobalt monoxide (CoO) layer (< 1 nm) inserted in the Co/Ag interface. The sandwiches exhibited anomalous magnetoresistance (MR) behavior depending on the tAg value. Inverse magnetoresistance (IMR) effect was observed only at thin tAg values of 2 nm and 6 nm, whereas the dual-anisotropic MR (dual-AMR) occurred at thicker tAg values of 12 nm, 25 nm and 65 nm. A superposition of the dual-AMR and the weak IMR effects was obtained at the thickest Ag layer of 65 nm. The origin of these anomalous MR behaviors was discussed to show the prominent role of the CoO layer and large thickness of the Co and Ag layers.

Risk factors for cannula-associated arterial thrombosis following extracorporeal membrane oxygenation support: a retrospective study

Trieu Ngan Hoang Kim,Phan Xuan Thi,Tran Linh Thanh,Pham Huy Minh,Huynh Dai Quang,Nguyen Tuan Manh,Mai Anh Tuan,Du Quan Quoc Minh,Nguyen Bach Xuan,Pham Thao Thi Ngoc 대한중환자의학회 2023 Acute and Critical Care Vol.38 No.3

Background: Hemostatic dysfunction during extracorporeal membrane oxygenation (ECMO) due to blood-circuit interaction and the consequences of shear stress imposed by flow rates lead to rapid coagulation cascade and thrombus formation in the ECMO system and blood vessels. We aimed to identify the incidence and risk factors for cannula-associated arterial thrombosis (CaAT) post-decannulation. Methods: A retrospective study of patients undergoing arterial cannula removal following ECMO was performed. We evaluated the incidence of CaAT and compared the characteristics, ECMO machine parameters, cannula sizes, number of blood products transfused during ECMO, and daily hemostasis parameters in patients with and without CaAT. Multivariate analysis identified the risk factors for CaAT. Results: Forty-seven patients requiring venoarterial ECMO (VA-ECMO) or hybrid methods were recruited for thrombosis screening. The median Sequential Organ Failure Assessment score was 11 (interquartile range, 8–13). CaAT occurred in 29 patients (61.7%), with thrombosis in the superficial femoral artery accounting for 51.7% of cases. The rate of limb ischemia complications in the CaAT group was 17.2%. Multivariate analysis determined that the ECMO flow rate–body surface area (BSA) ratio (100 ml/min/m2) was an independent factor for CaAT, with an odds ratio of 0.79 (95% confidence interval, 0.66–0.95, P=0.014). Conclusions: We found that the incidence of CaAT was 61.7% following successful decannulation from VA-ECMO or hybrid modes, and the ECMO flow rate–BSA ratio was an independent risk factor for CaAT. We suggest screening for arterial thrombosis following VA-ECMO, and further research is needed to determine the risks and benefits of such screening.

The Impact of Digital Transformation on Training Activities of the Accounting Major in Universities in Vietnam

Thi Huyen Tran,Hoang Tuan Nguyen,Quoc Cuong Nguyen 국제문화기술진흥원 2024 International Journal of Advanced Culture Technolo Vol.12 No.1

Accounting is one of the fields greatly affected by the Industrial Revolution 4.0 in general and the digital transformation trend in particular. However, the digital transformation process will also be a weakness if domestic accounting and auditing training activities do not adapt promptly and keep up with the practical requirements of social needs. Job opportunities in the accounting industry will shrink due to increased labor productivity and increasing support technologies, leading to a decrease in the need for traditional accountants. This paper focuses on clarifying the shortcomings and difficulties in accounting training activities by surveying 120 accounting lecturers teaching at domestic universities. Research results show that there are six existing factors and difficulties in accounting training activities due to the digital transformation process, including: facilities, lecturer capacity, training programs, teaching methods. teaching, assessment methods and practice and practice activities. Based on the survey results, we propose recommendations for training activities due to the impact of digital transformation at universities in Vietnam.