Browsing by Author "Shamsul Arefin, Mohammad"

Now showing 1 - 3 of 3

International Conference on Electronics, Computer and Communication (ICECC 2008)
(Bangladesh Electronics Society: Rajshahi, Bangladesh, 27-Jun-2008) Bangladesh, University of Rajshahi,; Shamsul Arefin, Mohammad; Karim, Rezaul; Aziz, Galib Bin; Zakaria, Sheikh Md.; Khastagir, Shaibal; Ullah, M. Ahsan; Shikder, Md. Abu Al Syeed; Hoque, Mohammed Moshiul; Azad, Samina; Sarifunnahar; Ahmed, Md. Raju; Rizwan, Mustofa; Islam, Md. Shahidul; Kumar Sen Gupta, Ashoke; -Ul-Karim, Meher; Abdul Mohit, Kazi; Hossain, Md. Akbar
Current Mobile Message Editing Analysis and Adopting Customizing Facility in SMS Abstract: Modern days of mobile communication acquainted us with a technology of Short Messaging Service or shortly SMS. It helps the users to interact with other mobile subscriber using short message. This built in technology is not so efficient and offers limited options. The research area in this regard is to make it more efficient and flexible. In case of sending SMS, we do not have the opportunity to customize our message using different font styles. In this paper, the customizing facility has been introduced in case of SMS. The objective is to provide the user a freedom to select different types of font style. This application is very simple to install in every Java supported mobile handsets. The application takes about 200-250 KB storage, which is small enough to store in any java enabled mobile or other hand held devices. It can be regarded that this project work will create a new dimension in SMS service.
KOA-CCTNet: An Enhanced Knee Osteoarthritis Grade Assessment Framework Using Modified Compact Convolutional Transformer Model
(2024-07-29) Jahan, Mushrat; Hasan, Md. Zahid; Samia, Ismot Jahan; Fatema, Kaniz; Hossen Rony, Md. Awlad; Shamsul Arefin, Mohammad
Knee osteoarthritis (KOA) is a prevalent condition characterized by gradual progression, resulting in observable bone alterations in X-ray images. X-rays are the preferred diagnostic tool for their ease of use and cost-effectiveness. Physicians use the Kellgren and Lawrence (KL) grading system to understand the severity of an individual condition of KOA. This system categorizes the disease from normal to a severe stage. Early detection of the condition with this approach enables knee deterioration to be slowed down with therapy. In this study, we aggregated four datasets to generate an extensive dataset comprising 110,232 raw images by applying an augmentation technique called deep convolutional generative adversarial network (DCGAN). We employed advanced image pre-processing methods (adaptive histogram equalization (AHE), fast non-local means), including image resizing, to generate a substantial dataset and enhance image quality. Our proposed approach involved developing a modified compact convolutional transformer (CCT) model known as KOA-CCTNet as the foundational model. We further investigated optimal configurations by adjusting various parameters and hyperparameters in the final model to handle large datasets and address training time concerns efficiently. We investigated optimizing its configurations by adjusting numerous parameters and hyperparameters to efficiently manage extensive data and address concerns related to training time. Simulation results indicated that our proposed model outperforms other transfer learning models (Swin Transformer, Vision Transformer, Involutional Neural Network) in terms of accuracy. The test accuracy for the ResNet50, MobileNetv2, DenseNet201, InceptionV3, and VGG16 was 80.77%, 79.98%, 80.23%, 76.89%, and 79.58%, respectively. All of them were surpassed by our proposed KOA-CCTNet model, which had a test accuracy of 94.58% while classifying KOA X-ray images. Furthermore, we reduced the number of images to assess the model’s performance and compared it to existing models. However, by employing a large datahub, our proposed approach provides a unique and effective way to diagnose KOA grades with satisfying results.
Multilingual data management in distributed environment
(Department of Computer Science and Engineering, BUET, 2008-08) Shamsul Arefin, Mohammad; Latiful Hoque, Dr. Abu Sayed Md.
Efficient storage and query processing of data spanning multiple natural languages are of crucial importance in todays globalized world. As Internet has become the primary medium for information access and commerce, the multilingual data management can be treated as a vital issue for the availability of information in the native language of the Internet users. The necessities of multilingual applications include better searching and browsing capabilities in languages other than English, accessing information stored in different languages, accelerating globalization of businesses and implementing e-Commerce and e- Governance modules etc. While existing database systems provide some means of storing and querying multilingual data, they suffer from redundancy proportional to the number of language support. In our research, we propose a multilingual data management system that stores data in information theoretic way in encoded form with minimum redundancy. Multilingual data has been stored in language independent way, which is an easier method for database evolution. Query operation can be performed from the encoded data only with the help of a translator and the result is obtained by decompressing it using the corresponding language dictionaries for text data or without dictionary for other data. Schema evolution is simple and easier to maintain database consistency, which is difficult in the existing systems. Query performance is also significantly faster. Our algorithms for handling multilingual data have been evaluated by both syntactic data generated by a data generation program and real data sets. We have compared the performance of our system with the existing systems. From the experiment it is found that the proposed approach is better in terms of both storage and query processing speed than the existing systems.