BRAC University Institutional Repository
Preserving Knowledge, Advancing Research, Sharing Scholarship
A digital platform for collecting, preserving, and sharing BRAC University’s scholarly, academic, and institutional outputs.
Contribute to the Repository
Faculty members and students are invited to submit their research publications, theses, dissertations, and scholarly works to increase visibility, access, and long-term preservation.
Recent Submissions
A qualitative study of the mental health experiences of cancer patients with the lowest survival rate and their caregivers
(Institute of Electrical and Electronics Engineers Inc., 2023-01-01) Orchi, Irin Hoque; Manzoor, Tahbib; Rahman Araf, Md. Wahidur; Sharker Omi, Monjurul; Abir, Tanvir Ahmed; Sadeque, Farig; Department of Computer Science and Engineering
Throughout the world, millions of people and their families are impacted by the serious illness of cancer. The total number of new cancer cases worldwide in 2020 was predicted to reach 18.1 million. Serious emotional disorders, like depression, are often present in cancer patients due to many factors, including the intensity of certain situations, the negative consequences of their long treatment, or the deaths of other cancer patients. Therefore, keeping an eye on the patient's moods is crucial to their ongoing treatment. Many cancer patients use online social media sites such as Facebook and Twitter to communicate their thoughts and emotions about their treatments, as well as the difficulties associated with them, in the form of posts or messages. From these sources, we can get good information about the mood of those patients, which will further help us with their treatment. After applying the necessary pre-processing to this data, we can apply sentimental analysis methods, which will help us predict the positive or negative emotions of cancer patients on these online platforms. We can give better psychological support to these patients after analyzing their mental health. So, our objective is to design a model capable of identifying such actions, as among all the cancers we are working on, five have the lowest survival percentage.
Trichophyton indotineae: Epidemiology, antifungal resistance and antifungal stewardship strategies
(John Wiley and Sons Inc, 2026-01-01) Gupta A.K.; Susmita; Nguyen H.C.; Liddy A.; Talukder, Mesbah; Wang T.; Magal L.; Chowdhary A.; Shemer A.; Saunte D.M.L.; Hay R.; Piguet V.; Journal of the European Academy of Dermatology and Venereology; School of Pharmacy
There has been a recent shift in the epidemiology of superficial fungal infections (tinea, dermatophytosis, dermatomycoses). Trichophyton indotineae is an emerging dermatophyte species of significant global concern for its contagious nature and antifungal drug resistance. This scoping review includes available clinical and laboratory assessments of T. indotineae to provide a comprehensive up-to-date overview of its epidemiology, clinical manifestations, diagnostic approaches, antifungal susceptibility patterns, resistance mechanisms and management strategies. We discuss T. indotineae resistance against standard and newer antifungals (terbinafine, griseofulvin and triazoles including fluconazole, itraconazole, voriconazole and posaconazole). In particular, the terbinafine susceptibility profile of T. indotineae can be linked to squalene epoxidase (SQLE) single-nucleotide variations. For diagnosis, it is not possible to separate T. indotineae from other members of the T. mentagrophytes complex (T. mentagrophytes and T. interdigitale) without access to molecular diagnostic methods. So, in patients presenting with extensive dermatophytoses, with a history of treatment resistance and/or recent travel, molecular diagnosis to confirm T. indotineae infection should be considered. Healthcare providers often face challenges in choosing between terbinafine and itraconazole treatments. While the use of terbinafine is limited due to resistance, itraconazole is hindered by erratic absorption, possible drug interactions and side effects as well as resistance in some cases. Newer treatments being investigated include super-bioavailable itraconazole, third-generation triazoles (voriconazole, posaconazole) and topical-oral combination regimens. The need for improved diagnostic accessibility, judicious antifungal prescribing, and implementing an effective antifungal stewardship program are highlighted.
A novel approach to efficient multilabel text classification: BERT-federated learning fusion
(Institute of Electrical and Electronics Engineers Inc., 2023-01-01) Sadot A.A.I.M.; Maliha Mehjabin M.; Mahafuz, Aziz; Department of Computer Science and Engineering
Large Language Model (LLM)-based transformers, such as Bidirectional Encoder Representations from Transformers (BERT), are currently gaining significant attention for various Natural Language Processing (NLP) tasks, such as machine translation, classification, and auto-completion. These transformer models demonstrate substantial performance improvements for text classification tasks. Multi-label classification problems often require more computation than binary and multi-class classification problems. Also, the computation requirements become more aggressive if large datasets are considered. Federated Learning (FL) offers a solution to train models in a distributed manner while preserving data privacy. This paper proposes a novel approach for building a machine learning model, which deals with a sizeable textual dataset for multi-label classification leveraging FL. FL has been used to train a compound model constructed by extending Bidirectional Encoder Representations from Transformers (BERT) with a "One-dimensional Convolutional Neural Network (1D CNN)". At first, The experiment was conducted in a single machine (Central) with the entire dataset. Then, the dataset was split into two groups, and the same experiment was performed in a Federated Learning fashion (BERT-FL Fusion). The FL setup considerably reduced the required computing power to derive an equivalent global model while increasing accuracy, precision, and F1 Score and minimizing Hamming Loss.
Sentiment analysis of amazon reviews using machine learning classifier
(Institute of Electrical and Electronics Engineers Inc., 2023-01-01) Monsoor, Razin Sumyta; Tamanna, Tania Sultana; Khan, Salequzzaman; Hoque, Shehrin; Islam, Mahdi; Rhythm, Ehsanur Rahman; Mehedi, Md Humaion Kabir; Rasel, Annajiat Alim; Department of Computer Science and Engineering
Reviews can significantly impact a company's reputation in the market, potentially influencing its overall business outcomes, either positively or negatively. This is especially crucial for companies that operate primarily through e-commerce platforms. Hence, it is vital for companies to pay close attention to customer reviews. Sentiment Analysis, often referred to as "opinion mining,"is a significant procedure in Natural Language Processing (NLP) which serves the purpose of ascertaining the emotional tone of a provided text and categorizing it into positive, negative, or neutral perspectives. In this paper, sentiment analysis methodology is presented for classifying Amazon reviews which utilizes a large dataset of reviews and employs Multinomial Naïve Bayesian (MNB), Support Vector Machine (SVM), Maximum Entropy (ME), and Logistic Regression as the primary classifiers by the authors. With the aid of machine learning, we employed a supervised learning approach to an extensive Amazon dataset in order to categorize it based on sentiment polarity, achieving a high level of accuracy for the results. Here, we utilized the Kaggle dataset that includes a substantial volume of reviews and associated metadata which comprises customer reviews and ratings on Amazon products.
ReVive Grip: Charting a path for affordable post-stroke hand rehabilitation
(Institute of Electrical and Electronics Engineers Inc., 2023-01-01) Chowdhury, Md Fahin Fuad Irfa; Adittya, Ashraf Shafayet; Ahnaf, Salman; Rafid, Sk Tahmed Salim; Islam, Sajidul; Huda, A. S. Nazmul; Department of Electrical and Electronic Engineering
This paper presents a cost-effective glove designed for post-stroke hand rehabilitation. The implemented system utilizes servo motors as actuators, collaborating with a spooling mechanism to delicately manipulate cords and replicate intricate gripping motions. The practical application of flex sensors is explored. The average response time of the prototype provides valuable insight into the efficiency of operation, resulting in a fast response time of 1.78 seconds. A demonstration of cost efficiency is found in the resulting prototype, realizable at approximately USD 100. At the end, an analysis based on 7th order polynomial regression model has also been included in order to improve the results further. This shows a significant reduction of average error.