Scholarly Indexed Publications

Permanent URI for this collectionhttps://hdl.handle.net/10361/28605

Browse

Now showing 1 - 20 of 1040

Open Access
The jamkhed comprehensive rural health project and the alma-ata vision of primary health care
(American Public Health Association Inc., 2019-05-01) Perry, Henry B.; Rohde, Jon; BRAC James P Grant School of Public Health
The Jamkhed Comprehensive Rural Health Project (Jamkhed CRHP) was established in central India in 1970. The Jamkhed CRHP approach, developed by Rajanikant and Mabelle Arole, was instrumental in influencing the concepts and principles embedded in the 1978 Declaration of Alma-Ata. The Jamkhed CRHP pioneered provision of services close to people’s homes, use of health teams (including community workers), community engagement, integration of services, and promotion of equity, all key elements of the declaration. The extraordinary contributions that the Jamkhed CRHP has made as it approaches its 50th anniversary need to be recognized as the world celebrates the 40th anniversary of the International Conference on Primary Health Care and the writing of the declaration. We describe the early influence of the Jamkhed CRHP on the declaration as well as the work at Jamkhed, its notable influence in improving the health of the people it has served and continues to serve, the remarkable contributions it has made to training people from around India and the world, and its remarkable influences on programs and policies in India and beyond.
Metadata Only
Enhancing bidirectional sign language communication: Integrating YOLOv8 and NLP for real-time gesture recognition & translation
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Bhuiyan, Hasnat Jamil; Mozumder, Mubtasim Fuad; Khan, Md Rabiul Islam; Ahmed, Md Sabbir; Nahim, Nabuat Zaman; Department of Computer Science and Engineering
The primary concern of this research is to take American Sign Language (ASL) data through real time camera footage and be able to convert the data and information into text. Adding to that,we are also putting focus on creating a framework that can also convert text into sign language in real time which can help us break the language barrier for the people who are in need. In this work, for recognising American Sign Language (ASL), we have used the You Only Look Once(YOLO) model and Convolutional Neural Network (CNN) model. YOLO model is run in real time and automatically extracts discriminative spatial-temporal characteristics from the raw video stream without the need for any prior knowledge, eliminating design flaws.The CNN model here is also run in real time for sign language detection. We have introduced a novel method for converting text based input to sign language by making a framework that will take a sentence as input, identify keywords from that sentence and then show a video where sign language is performed with respect to the sentence given as input in real time.To the best of our knowledge, this is a rare study to demonstrate bidirectional sign language communication in real time in the American Sign Language (ASL).
Metadata Only
Template music transcription for different types of musical instruments
(Institute of Electrical and Electronics Engineers Inc., 2010-05-28) Faruqe, Omar; Department of Computer Science and Engineering
Automatically recovering the musical notation for a given signal allows reproducing and modifying the original performance. Music transcription can be considered as extracting the sequence of the notes that best models our audio signal. So it is understood as the process of analyzing a music signal in order to write down the parameters of the sounds occurring in it. Another principal application is structured audio coding: a MIDI-like representation which is highly compact but holds on to the identification and characteristics of a musical note to a significant extent. The applied notation can be the traditional musical notation which gives sufficient information for performing the piece using the available musical instruments. Automatic Music Transcription (extracting musical notes from an audio stream) is a complicated task due to the complexity of musical sounds that requires an effective solution. The scope of this article is in the automatic transcription of the harmonic and melodic parts of the real-world music signals. Algorithms are proposed that address two distinct sub-problems of music transcription. The main part of the article emphasizes the estimation of the several concurrent musical notes. The other sub-problem discusses musical meter estimation. This has to do with rhythmic aspects of music and refers to the estimation of the regular pattern of strong and weak beats in a piece of music. Signal processing methods for the automatic transcription of music are developed in this paper.
Open Access
Pregnant women's and community health workers' perceptions of root causes of malnutrition among infants and young children in the slums of Dhaka, Bangladesh
(2011-07-01) Goudet, Sophie M.; Faiz, Sabina; Bogin, Barry A.; Griffiths, Paula L.; BRAC James P Grant School of Public Health
Research in Bangladesh shows that malnutrition among infants and young children is most severe in urban slums. We examined the root causes of malnutrition as perceived by pregnant women and community health workers. We conducted 10 focus group discussions in the slums of Dhaka in 2008 and 2009. Participants accurately perceived inappropriate care, inappropriate environment, inappropriate food, and flooding to be major causes. Recurrent flooding has not traditionally been identified by experts as a cause of malnutrition. We recommend further research to address the nutritional risks flooding creates for vulnerable slum populations.
Open Access
Will nirsevimab be the holy grail for prevention of respiratory syncytial virus lower respiratory tract infections in infants?
(AME Publishing Company, 2024-03-27) Homaira, Nusrat; BRAC James P Grant School of Public Health
Metadata Only
ViTDeBERTaNews: A comparative study of single-modal, multimodal, and LLM techniques for detecting fake news
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Sultan T.; Sarker M.M.H.; Bhuiyan, Md. Khairul Bashar; Saib M.; Islam M.S.; Hossain M.R.; El-Shafai W.; Azar A.T.; Njima C.B.; Department of Electrical and Electronic Engineering
The spread of fake news online poses a significant challenge, particularly as social media increasingly combines images and text. To address this, we present the ViT+DeBERTaNews model, which effectively merges visual and textual information for fake news detection. This model utilizes ViT for detailed visual feature extraction and DeBERTa for deep textual understanding. Our experiments demonstrate its effectiveness, achieving 94.8% accuracy on the Weibo dataset and 92.6% on Twitter. The model's precision, recall, and F1 scores for fake news detection were 0.967, 0.945, and 0.956 on Weibo, and 0.929, 0.930, and 0.933 on Twitter, respectively. For real news, it scored 0.968, 0.944, and 0.956 on Weibo, and 0.925, 0.944, and 0.956 on Twitter. In contrast, text-based models like GPT-2 Epoch 3, while strong in precision and recall, are limited by their text-only approach. GPT-4 also faced challenges in recall on the Weibo dataset, indicating the need for task-specific optimizations. These findings underscore the necessity of advanced multimodal models for effective fake news detection.
Metadata Only
MemeFusionNet: A cross-linguistic multimodal model for identifying troll memes
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Sultan T.; Akbarpour H.A.; El-Shafai W.; Saib M.; Bhuiyan, Md. Khairul Bashar; Islam M.S.; Azar A.T.; Njima C.B.; Department of Electrical and Electronic Engineering
The proliferation of troll memes, which exploit textual and visual elements to propagate misinformation and incite negativity, presents a critical challenge for online content moderation. Existing methods often struggle with cross-linguistic generalization, multimodal fusion, and contextual understanding, limiting their effectiveness in multilingual environments. To address these gaps, we propose MemeFusionNet, a transformer-driven multimodal fusion framework that effectively captures the intricate relationships between images and text. MemeFusionNet integrates a cross-modal attention mechanism based on ViLT to enhance contextual awareness and better detect implicit troll content, such as sarcasm and cultural nuances. Our model demonstrates superior performance on Bangla and English meme datasets, achieving 86% and 96% accuracy, respectively, outperforming all existing benchmarks. Its scalable architecture ensures robust cross-lingual adaptability, making it well-suited for large-scale, real-time content moderation.
Open Access
Analysis of the correlation between SARS-CoV-2 transmission and meteorological parameters in Bangladesh
(SIPISS- Edizioni FS Publishers, 2022-01-01) Rahaman, Md. Mushfiqur; Khan, Risala T.; Zaman, Shakila; Rahman, Md. Tanvir; Shuva, Taslima F.; Department of Computer Science and Engineering
Introduction: Since COVID-19 has been characterized as a worldwide epidemic, multiple studies have suggested that weather may have a role in virus transmission. This research aims to examine the correlation between meteorological parameters and SARS-CoV-2 transmission, as well as to forecast cumulative COVID-19 cases in Bangladesh. Methods: In this study, an average incubation period of 5-6 days was used to examine the real effect of environmental parameters on SARS-CoV-2 transmission. Therefore, considering the incubation period and reporting time a standard 7-day shift in meteorological parameters from the daily COVID-19 cases was applied to measure the actual correlation. In this regard, the non-parametric correlation test (Spearman's Rank Correlation) was performed where 95% (p < 0.05) and 99% (p < 0.01) confidence intervals were considered as an acceptance criterion. Results: This work found a significant positive correlation (p < 0.01) for COVID-19 cases with minimum temperature, average temperature (only for division-specific analysis), wind speed, rainfall, humidity, and cloud. Furthermore, a significant negative correlation (p < 0.01) was found with atmospheric pressure and sun hours. However, the impact of maximum temperature (except for some divisions) or UV index was significantly low. Discussion: Our findings showed that the strength of the correlation coefficient is higher for the test positivity rate rather than the confirmed case count. However, to forecast the cumulative cases of COVID-19, ARIMA (Autoregressive Integrated Moving Average) may be considered the best-fitting model according to AIC (Akaike Information Criterion) and performs slightly better than Holt's exponential smoothing model. Additionally, this study represents the comparative analysis between predicted and actual Coronavirus-19 cases during December 2021 to show how close the predicted result is to the selected model. Take-home message: In this study, the overall analysis indicates that COVID-19 outbreaks in Bangladesh are more likely to hit massively during the pre-monsoon (March to May) season and the monsoon (June to October) season than in winter (November to February). These findings might be useful for decision-makers and authorities to know more about the seasonal impact of the outbreak and plan accordingly before the country enters a new weather season. © 2022 by the authors.
Open Access
People with disabilities and transport access: Evidence from PENDA and inclusive futures1
(Institute of Development Studies, 2026-01-01) Carew, Mark T.; Das, Narayan; Thompson, Stephen; BRAC Institute of Governance and Development
Transport is a fundamental enabler of participation in society yet people with disabilities frequently experience barriers to accessing it. These barriers are often greater in low- and middle-income countries (LMICs). Despite its importance, transport remains neglected in disability-inclusive development interventions. This article synthesises evidence from the Disability Inclusive Development Inclusive Futures programme and the Programme for Evidence to Inform Disability Action, which implement and evaluate disability inclusion interventions across multiple LMICs. We draw on their findings to examine the transport barriers experienced by people with disabilities and how these shaped intervention participation and outcomes. Findings highlight inaccessible infrastructure, a lack of accessible public transport options, unavailability of assistive technology, and transport-related stigma and discrimination as barriers. These barriers were shaped by intersectional disadvantage and risk constraining intervention participation and impact. We discuss the implications for disability-inclusive development intervention design and national transport policy, highlighting that transport access is a prerequisite for meaningful inclusion. © 2026 The Authors. IDS Bulletin © Institute of Development Studies
Metadata Only
PrecisionStroke: Optimized stroke risk prediction through advanced hyperparameter tuning and machine learning techniques
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Arman, Mithila; Shuba S.J.; Rhaman M.M.; Choya S.B.Z.; Chowdhury M.A.A.; Islam M.; Sheikh I.A.; Jahan M.K.; Department of Computer Science and Engineering
Stroke prediction is a challenging problem in the healthcare domain, especially due to high class imbalance in datasets where data about stroke cases is a very small portion. In this study we develop a machine learning based framework for stroke prediction that tackles through powerful data preprocessing methods, such as Synthetic Minority Oversampling Technique (SMOTE) for balancing the classes. (Situated within the entire framework of a comparative evaluation of the Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), CatBoost, XGBoost, LightGBM, and Multi-Layer Perceptron (MLP) and TabNet machine learning classifiers as per unique designs. In conclusion, the Random Forest classifier proved to be the best performing model with an accuracy of 99.38%, followed closely by LightGBM and CatBoost. A detailed analysis including KP song metrics including precision, F1 score, and confusion matrices were carried out to ensure thorough evaluation. The study also emphasizes the importance of explainable AI tools which improve the explainability of predictions. This framework shows that machine learning has the potential to assist with early detection of stroke and may ultimately be integrated into clinical decision-making systems.
Open Access
Public university students’ experiences of anti-feminist backlash in Dhaka, Bangladesh*††
(Institute of Development Studies, 2024-03-01) Ahmed, Adeepto Intisar; Jahan, Ishrat; Hasan, Israr; Rashid, Sabina Faiz; Naomi, Sharin Shajahan; BRAC James P Grant School of Public Health
Public university campuses in Bangladesh have been historically significant sites of negotiating with social and political orders. Based on in-depth interviews with male and female students from three public universities in Dhaka, conducted between 2022 and 2023, this article identifies the ways in which formal and informal structures of power on campus reproduce patriarchal norms and gendered inequities. The students’ narratives shed light on how the culture of residential halls, and practices of policing and surveillance, interact with patriarchal norms to limit women’s agency and mobility. They also show the ways in which masculine practices which draw from hypersexual views on women and glorify violence become enabled and sustained by institutional power dynamics, wherein harassment and policing become instruments to negotiate power. The article provides new insights into the ways in which patriarchal power dynamics and gender norms promoted and practised within an institutional space create drivers of anti-feminist backlash. © 2024 The Authors. IDS Bulletin © Institute of Development Studies |.
Open Access
‘It’s a family matter’: Inaction and denial of domestic violence
(Institute of Development Studies, 2024-03-01) Sultan, Maheen; Mahpara, Pragyna; BRAC Institute of Governance and Development
This article provides a grounded example of backlash in action surrounding the implementation of the Domestic Violence (Prevention and Protection) Act 2010 in Bangladesh. While formulation and enactment of the law marked significant achievements, its implementation has been weak. Unlike conventional analyses that concentrate on backlash in gender equality policy formulation, this study focuses on the obstacles encountered during the implementation phase. Through in-depth interviews with advocates and stakeholders responsible for implementation, the article examines their attitudes and interests concerning the law and women’s rights to life, dignity, and bodily integrity. Prevailing gender norms perpetuate the trivialisation of domestic violence, framing it as a personal issue of minimal importance. Consequently, service providers tend to delegitimise and deprioritise it. This article investigates the strategies and tactics of deliberate inaction employed by backlash proponents, which is different from lack of capacity, and explores the counter-strategies deployed by advocates aiming to ensure the Act’s effective implementation. © 2024 The Authors.
Metadata Only
Multi-head attention based LSTM model for forecasting Bangladeshi stock market prices
(Institute of Electrical and Electronics Engineers Inc., 2025-01-01) Jahan M.K.; Chowdhury M.E.; Shuba S.J.; Sabahat Khan F.; Rahman M.M.; Faruk M.O.; Azam Chowdhury M.A.; Jahidul Islam S.M.; Arman, Mithila; Department of Computer Science and Engineering
Forecasting stock prices is crucial for investment and risk management, offering insights into market trends and aiding investors in making informed decisions. Traditional time-series models, such as LSTM and GRU, have demonstrated promise in capturing temporal dependencies within stock market data. However, these models often fall short in identifying complex patterns due to limited attention mechanisms, particularly in emerging markets with unique economic factors like the Bangla stock market. Addressing this gap, we propose a novel Multi-Head Attention and LSTM-based architecture that combines the sequential processing capabilities of LSTM with multi-head attention layers to better capture contextual dependencies and subtle fluctuations in stock price movements. Our approach is evaluated on datasets from six major Bangladeshi banks, including Islami Bank, Dhaka Bank, Arab Bangladesh (AB) Bank, Bank Asia, Eastern Bank Limited (EBL), and City Bank. Results demonstrate that our model outperforms traditional models across all metrics. The integration of multi-head attention allows for nuanced attention allocation across time steps, significantly enhancing forecasting accuracy. This enhanced model can serve as a valuable tool for investors and financial analysts, supporting better forecasting for risk assessment and investment strategy formulation. The framework demonstrates potential scalability across other emerging markets, offering an adaptable solution for financial forecasting in regions with complex market dynamics.
Metadata Only
Autonomous mobile chargers for rechargeable sensor networks using space filling curve
(Institute of Electrical and Electronics Engineers Inc., 2018-07-27) Chowdhury S.A.; Benslimane A.; Akhter, Farzana; Department of Electrical and Electronic Engineering
Recent breakthrough in the wireless energy transfer have attracted attention of the researchers due to its potential to prolong the lifetime of Wireless Sensor Networks (WSNs) and to eliminate energy bottlenecks. Through wireless charging vehicles sensor node's energy can be replenished periodically by using Wireless Charging Vehicles (WCVs). However, for a large scale WSN, the capacity of WCVs, positioning and coordination between them, and limited resources of WCVs have to be considered to implement a scalable charging system. Additionally, creating an efficient charging route for the WCVs considering the dynamic energy consumption of sensor nodes is also crucial. Since, WCVs have limited capacity and low system resources, the classic TSP-based Optimization algorithms, which requires high-performance computing, are not suitable for them. In this paper, we proposed GHSC (Guided Hilbert for Specified Cluster), which is based on the Hilbert space filling curve. We consider a clustered WSN with multiple WCVs, where each cluster is assigned to a single WCV. GHSC creates a primary charging route by considering a cluster as an unit square and dividing the square into small sub-squares. As the Hilbert curve goes through each sub-squares, the algorithm assigns a rank for each of them based on their position along the curve. The algorithm then map the nodes into vertices of the sub-squares and rank the nodes as well. Once ranking is done, the algorithm creates a charging route based on the node's ranking. This heuristic approach of the GHSC algorithm can find effective charging tours for the WCVs with a time complexity of (O(n log n) + ?(nk)). Compared to other Global Optimization algorithms, GHSC was able to find tours, which was at most a logarithmic factor longer than the shortest tour.
Metadata Only
Fair and efficient weighted sum rate maximization for multi-rate secondary users in cognitive radio network
(Institute of Electrical and Electronics Engineers Inc., 2017-07-28) Chowdhury S.A.; Benslimane A.; Akhter, Farzana; Department of Electrical and Electronic Engineering
Distributing scarce spectral resources among the unlicensed users has been an attractive research area for Cognitive Radio Network's (CRN) research community. A resource distribution technique, which emphasizes fairness, ensures allocation of resources for all Secondary Users (SUs) irrespective of their data rates and may cause efficiency loss for the CRN. On the other hand, a throughput based resource allocation approach considers SUs with high data rates only and consequently SUs with low data rates face negative experiences as they starve from resources. Our work aims to balance between the fairness and the efficiency of a CRN. We formulate an objective function, which is a nonlinear convex function, to achieve maximum weighted sum rate for the SUs while ensuring fairness to all. We use Primal Dual Interior Point Method to solve the optimization problem and define a weight factor to obtain balance between the fairness and the throughput of the CRN. Finally we present an online iterative algorithm which maximizes the weighted sum rate of the SUs while guaranteeing QoS to the Primary Users (PUs). Numerical results exhibit that our method achieves higher throughput while ensuring adequate fairness to the SUs, compared to other traditional fairness schemes.
Metadata Only
Throughput maximization of cognitive radio network by conflict-free link allocation using neural network
(Institute of Electrical and Electronics Engineers Inc., 2017-07-28) Chowdhury, Sakil Ahmed; Benslimane, Abderrahim; Akhter, Farzana; Department of Electrical and Electronic Engineering
Interference, which can be caused by Secondary Users (SUs) to other Primary Users (PUs) or SUs, is of major concern for Cognitive Radio Networks (CRNs). We study a spectrum allocation problem, where our primary focus is on maximizing the throughput of the CRN while protecting the PUs from interference and controlling the co-channel interference among the SUs. We build a CRN, where initially, all SUs explore prospective links for communicating between them and send the sensed data to a centralized Base Station (SU BS). The SU BS aggregates the received information and discovers conflicting links among the SUs by using an interference graph. Conflict among links affects the SINR, which in essence hampers the efficiency of the CRN. We form an optimization problem, which is a binary integer linear programming problem, with an objective to determine the optimal set of conflict-free links that satisfies the interference constraints. We propose the CFLA (Conflict-free Link Assignment) algorithm, which uses a neural network for capacity-aware link distribution. Simulation results show that the algorithm has less time complexity than the evolutionary algorithms such as genetic algorithm in particular and perform steadily for large scale CRNs as well.
Metadata Only
Real-time intruder detection in surveillance networks using adaptive kernel methods
(Institute of Electrical and Electronics Engineers Inc., 2010-08-13) Ahmed, Tarem; Ahmed, Sabrina; Ahmed, Supriyo; Motiwala, Murtaza; Department of Electrical and Electronic Engineering
In this paper we apply a recursive algorithm based on kernel mappings to propose an automated, real-time intruder detection mechanism for surveillance networks. Our proposed method is portable and adaptive, and does not require any expensive or sophisticated components. Through application to real images from BRAC University's closed-circuit television system and comparison with common methods based on Principle Component Analysis (PCA), we show that it is possible to obtain high detection accuracy with low complexity.
Metadata Only
Flow vector prediction using EM algorithms
(Institute of Electrical and Electronics Engineers Inc., 2010-08-13) Ahmed, Tarem; Department of Electrical and Electronic Engineering
This paper considers the problem of predicting the number, length and distribution of IP traffic flows some time into the future, based upon packets collected in the present. Two versions of the Expectation-Maximization (EM) algorithm are used to predict the mean flow length and complete flow distributions for subsequent timesteps. A model is first used to represent the histogram of flows corresponding to any given time interval, and the EM algorithms are then used to estimate the parameters of the model. The proposed algorithms are tested on a large number of commonly-available data traces and both show high prediction accuracy.
Metadata Only
Challenges, comparative analysis and a proposed methodology to predict sentiment from movie reviews using machine learning
(Institute of Electrical and Electronics Engineers Inc., 2017-10-17) Ahmed, Erfan; Sazzad, Md. Asad Uzzaman; Islam, Md. Tanzim; Azad, Muhitun; Islam, Samiul; Ali, Mohammad Haider; Department of Computer Science and Engineering
This paper investigates a new approach of finding sentence level sentiment analysis using different machine learning algorithms. Three different machine learning algorithms - SVM (Support Vector Machine), Naïve Bayes and MLP (Multilayer Layer Perceptron) have been used both for sentiment analysis. Moreover two different classifiers of Naïve Bayes and two different types of SVM kernels have been used in this work to identify and analyze the difference in accuracy as well as to find the best outcome among all the experiments. For sentiment analysis aclimdb movie review dataset has been used. Lastly, the impact of stop words and number of attributes in accuracy for sentiment analysis has also been illustrated.
Metadata Only
Feature cloning and feature fusion based transportation mode detection using convolutional neural network
(Institute of Electrical and Electronics Engineers Inc., 2023-04-01) Alam, Md. Golam Rabiul; Haque M.; Hassan M.R.; Huda S.; Hassan M.M.; Strickland F.L.; Alqahtani S.A.
The smartphone-based sensors (including accelerometer, proximity, and gyroscope sensors) are ubiquitous and emerging mobility data sources that could be used for transportation modes (i.e. bus, train, car, walking, and stationary) detection. One of the important challenges in transportation modes detection is to build an appropriate model that can extract useful data from the sensor outputs and that can reduce misclassifications. Several factors make the feature modeling difficult including inappropriate sampling frequency of input signals, wavering behavior of devices (e.g. the changing orientation of a device relative to the human body), and continuous base vibration causing similar sensor outputs for both stationary and non-stationary states and related threshold values of velocity. This paper proposes novel approaches to address these challenges by developing a robust transportation mode detector based on a convolution neural network (CNN). The proposed robust detector develops a feature modeling technique by novel feature fusion and cloning techniques. Pre-trained features are constructed using a separate vanilla neural network (VNN) framework to extract the distinguishing components from the original features that are combined with the original and cloned features. The proposed feature fusion technique is successfully able to overcome the noise from the base vibration and the minimal informative outputs from the lower sampling frequency. This enables the CNN to be trained with more efficient and discriminative features that result in a better classification model. The proposed approaches have been validated using a large volume of mobile sensor data based on the movements of travelers. Different types of mobile sensors have been used to collect data including accelerometer, proximity, and gyroscope. Experimental results demonstrate that the proposed approaches can improve the performance of the detection engine significantly over conventional techniques and reduces the misclassification rate.

Browse

Recent Submissions