Early View

PAPER ID: 1537

TITTLE: XNOR-Popcount, an Alternative Solution to the Accumulation Multiplication Method for Approximate Computations, to Improve Latency and Power Efficiency 

AUTHOR(S): Van-Khoa Pham*, Lai Le, Thanh-Kieu Tran Thi

ABSTRACT: 

Convolutional operations on neural networks are computationally intensive tasks that require significant processing time due to their reliance on calculations from multiplication circuits. In binarized neural networks, XNOR-popcount is a hardware solution designed to replace the conventional multiplied accumulator (MAC) method, which uses complex multipliers. XNOR-popcount helps optimize design area, reduce power consumption, and increase processing speed. This study implements and evaluates the performance of the XNOR-popcount design at the transistor-level on the Cadence circuit design software using 90nm CMOS technology. Based on the simulation results, for the same computational function, if MAC operation uses XNOR-popcount, the power consumption, processing time, and design complexity can be maximally reduced by up to 69%, 50%, and 48% respectively when compared to the method using conventional multipliers. Thus, the XNOR-popcount design is a useful method to apply to edge-computing platforms with minimalist hardware design, small memory space, and limited power supply.

VERSION OF RECORD ONLINE: 15 April 2024            VIEW PDF

*******************************************************************************

PAPER ID: 1464

TITTLE: Effect of Materials on the Mechanical Properties of Fused Deposition Modeling Three-Dimensional Printed Products

AUTHOR(S): , , Duy Phu Nguyen, Tu San Tran, Hai Yen Tran, Ngoc Phung Nguyen, Thong Minh Vo, Thi Hong Nga Pham, Vinh Tien Nguyen, Thanh Tan Nguyen*

ABSTRACT: 

This study evaluates the three-dimensional (3D) printing materials used in Fused Deposition Modeling (FDM) printing technology. 3D printing technology has been developing strongly, becoming an effective support tool in production and research. The 3D printing process involves many stages, with many parameters affecting the quality and properties of the product, in which 3D printing material is one of many essential factors affecting that process. The study conducts a comprehensive assessment of the most common materials in 3D printing technology to determine the advantages and limitations of precisely five types of materials: Polylactic acid, acrylonitrile butadiene styrene, polyethylene terephthalate glycol-modified, thermoplastic polyurethane, and acrylonitrile styrene acrylate. With 3D printing, parameters such as sintering temperature, printing speed, and layer thickness are kept constant. These parameters are applied equally to all five material samples. The experiment evaluates the tensile strength of materials. The study results provide an overview of the properties and applicability of 3D printing materials, helping to select materials suitable for specific FDM 3D printing technology applications.

VERSION OF RECORD ONLINE: 12 April 2024            VIEW PDF

*******************************************************************************

PAPER ID: 1516

TITTLE: Capacity to Organize and Manage Online Teaching of some Higher Education Institutions

AUTHOR(S): Khuong Nhu Nguyen,  Thuy Thanh Nguyen,  Tuan Van Nguyen*

ABSTRACT: 

Online teaching is a form of teaching developed in the late 1990s with Internet and hypertext technology. Online teaching applications are centrally located on the education service provider's web server and can be accessed by learners [1]. Online teaching has been widely applied in educational institutions, especially higher education in developed countries. In Vietnam, it has been widely deployed recently. There have been a number of research projects on the management of online teaching in higher education institutions such as that of Tran Thi Lan Thu [2], and research on the organization of online teaching for teachers, but there has been no research topic Evaluation of the capacity to organize and manage online teaching of higher education institutions. This article clarifies the theoretical basis for the capacity to organize and manage online teaching of higher education institutions, thereby assessing the current situation of the capacity to organize and manage online teaching through a number of popular higher education institutions through survey methods using questionnaires and interviews with teachers and administrators. Research results have shown some limitations in the capacity to organize and manage online teaching of higher education institutions.

VERSION OF RECORD ONLINE: 09 April 2024            VIEW PDF

*******************************************************************************

PAPER ID: 1522

TITTLE: Enhancing Accuracy in Classification Models for Skin Disease Diagnosis based on Segformer and ConvNeXt Approach

AUTHOR(S): Hai-Duong Le, Van-Dung Hoang^*

ABSTRACT: 

This study introduces an innovative methodology to enhance the precision of skin disease diagnosis classification models by integrating segmentation results. Employing advanced machine learning techniques, our approach involves predicting lesion areas in skin images by combining SegFormer for skin lesion segmentation and backbone ConvNeXt for classifying skin images that consist of benign and malignant diseases. Based on training the SegFormer model for skin lesion segmentation, it achieved the IoU (intersection over union) ratio of 0.861 on the test set, outperforming the top 1 entry on the ISIC 2018 Leaderboards, which had an IoU of 0.802. Furthermore, our skin classification model uses image cropping to generate input images that emphasize damaged skin areas, eliminating redundant information. Leveraging the segmentation model’s results, we define the bounding box for the lesion area, obtain a new image within the bounding box by adding padding, and then compare this new data with the original data. The disease classification model, using ConvNeXt as its backbone, exhibited superior performance on the new dataset compared with the original dataset, achieving a higher accuracy of 1.61%, precision of 26.42%, and recall of 26.49%. This research paves the way for novel approaches to address disease diagnosis challenges in medical images, particularly in skin diseases. It can improve the performance of classification models when trained on image datasets that do not have synchronization during acquisition.

VERSION OF RECORD ONLINE: 05 April 2024            VIEW PDF

*******************************************************************************