Browsing by Author "Alam, Ashraful"

Now showing 1 - 14 of 14

Acute lower respiratory tract infections in hospitalized patients with diarrhea in Dhaka, Bangladesh
(1990-11) Rahman, Mahbubur; Huq, Farida; Sack, David A.; Butler, Thomas; Azad, A.K.; Alam, Ashraful; Nahar, Nurun; Islam, Motiul
An efficient deep learning approach to detect bone fractures from X-ray images
(BRAC University, 2025-06) Rahman, MD Fahmidur; Ridoy, MD Ashaduzaman; Mahajabin, SK Fahema; Tasawar, Sadman Rahman; Alam, Ashraful
We recommend an effective,reliable and efficient deep-learning approach to classify bone fractures using X-ray images. The proposed system comprises several steps as dataset collection, preprocessing, and categorization of fractures. The dataset contains different types of X-ray images for various fractures and annotated as fractured and non-fractured cases. Dataset was collected from Kaggle. The images were processed through resizing,augmentation to be standardized for deep learning models. The classification of fractures utilizes neural network paged deep learning models as the last process. The research uses Inception,VGG-19 and a custom CNN model for fracture detection while using Grad-CAM to demonstrate how each model makes its decisions. These models utilize CNN architectures to overcome current models accuracy limitations when detecting fractures. The study also shows how CNN-based approaches elevate the precision of bone fracture detection which creates potential benefits for medical use and improved patient healthcare.
An efficient deep learning approach to detect multiple neurodegenerative diseases using image data
(BRAC University, 2025-06) Sameen, Shadman Rahman; Khan, Shahabuddin Ahmed; Sakib, Mir Md. Muktasif; Chowdhury, Tushar; Alam, Ashraful
In today’s time early and accurate diagnosis of the neurodegenerative diseases (such as Alzheimer Disease (AD), Parkinson Disease (PD), Frontotemporal Dementia (FTD)) is essential to timely intervention and is quite laborious since the Magnetic Resonance Imaging (MRI) scans of all of them carry same set of characteristics. The deep learning pipeline proposed in this study will utilize 2D sagittal sectioning of 3D MRI columns and will be focusing on such crucial regions as the hippocampus, substantia nigra, and forehead lobe. These pictures were downsized to the identical standard of 224x224x3 and heavily augmented to promote generalization. We benchmarked the transfer learning classification performance of a number of pre-learning CNN models. Among the standard models, MobileNetV2 exhibited superior performance on the test set (94.36%), as opposed to EfficientNetB0 (85.47%), ResNet50 (68.29%), DenseNet121 (62.65%), and VGG19 (33.33%). However, MobileNetV2 exhibited a little overfitting. To address the challenges of accuracy problem of multiclass disease detection and overfitting issue we proposed SadNetV1 that was constructed by integrating MobileNetV2, Squeeze-and-Excitation (SE) blocks and Spatial Attention mechanism to facilitate effective dimensionality reduction in capturing channel and spatial dependence. The proposed SadNetV1 model showed improved performance of 96.15% test, 96.84% train, and 97.11% validation accuracy and can offer generalization in complex MRI images and could differentiate between AD and PD, and FTD. These results give credence to the potential of attention-augmented lightweight networks in effective categorization of neurodegenerative diseases in clinics.
An Evaluation of Human Resource Management Practices Of Labaid Cancer Hospital & Super Speciality Centre
(Daffodil International University, 2025-03-25) Alam, Ashraful
This internship report encapsulates three months of human resource practices at Labaid Cancer Hospital & Super Speciality Center (LCH). It aims to provide a comprehensive overview of LCH's patient support services and HR practices. LCH commenced operations in March 2021, and this study focuses on examining the efficiency of its HR practices over time. The main objective of this report is to gain a thorough understanding of LCH's HR practices. The study is divided into six sections, primarily based on firsthand insights from LCH. The first section summarizes the report, covering its context, objectives, methodology, and limitations. The second section provides an overview of LCH, detailing its patient-related activities and processes. The third section reviews relevant literature on human resource management practices. The fourth chapter focuses on the current HR practices at LCH, including the functions of the Human Resource Department and the performance outcomes of its staff. Chapter five presents a SWOT analysis of LCH’s HR Department, assessing its internal strengths and weaknesses, as well as external opportunities and threats. Key findings from the study of LCH’s HRM practices are included in this section. Finally, chapter six outlines the study's results and analysis, offers recommendations, and provides a conclusion. The report's main findings illustrate how LCH plans, recruits, selects, trains, evaluates, and rewards its employees through its HR Department.
An Investigative Approach To Employ Different Machine Learning Algorithms in Detecting Brain Cancer
(Department of Electrical and Electronic Engineering, Islamic University of Technology (IUT) The Organization of Islamic Cooperation (OIC) Board Bazar, Gazipur-1704, Bangladesh, 2022-05-30) Mohtasim, Md. Akib; Islam, Zahidul; Alam, Ashraful
This thesis takes an exploratory method to investigate the performance of several machine learning algorithms in more accurately detecting brain tumors. It primarily accomplishes this by detecting both malignant and benign lesions in the brain. In recent years, brain cancer has been linked to the highest newborn cancer fatality rates worldwide. Various machine learning techniques have been shown to be an excellent tool for detecting brain while it is still in its early stages. To train and test the model classifier, a dataset containing 700 instances and 1500 features from the Kaggle Data repository was used. Thirteen Eleven machine learning algorithms were studied and their performance parameters like confusion matrix and accuracy were analyzed. Furthermore, a thorough comparison was carried out through the compuration of precision, sensitivity, F1 score, recall, specificity, cross validation score and error rate of each algorithm.
Antimicrobial susceptibility of Neisseria gonorrhoeae isolated in Bangladesh (1997 to 1999): rapid shift to fluoroquinolone resistance
(2002) Rahman, Motiur; Sultan, Zafar; Monira, Shirajum; Alam, Ashraful; Nessa, Khairun; Islam, Sonia; Nahar, Shamsun; Shama-A-Waris; Khan, Shahnewaz Alam; Bogaerts, Jozef; Islam, Nazrul; Albert, John
Assessment of Water Pollution and its Impact on Fisheries Resources of Posna Beel, Kalihati, Tangail
(University of Rajshahi, 2006) Alam, Ashraful; Jahan, M. Sarwar; Zaman, M.
Posna Beel is an important perennial waterbody in Kalihati Upazila of Tangail District. The heel area is 121.5 ha in monsoon and 9 .0 ha in dry period. The heel is the main source of fish for Pouzan, Balla, Rampur and Elenga Bazars. A research work for two years was carried out on the water quality, Agro-chemical use, and pesticide residue and beel sediment of Posna heel. From water quality monitoring, it was found that the heel water was very suitable for open water fisheries. The levels of concentration of some water pollutants such as ammonia (0.012-0.076 mg/I) and nitrate (0.15-0.48 mg/1) were very low while the biochemical parameters such as BOD5 (2.70-5.43 mg/I) and COD (5.20-8.15 mg/l) were slightly high. Nitrite was found zero round the year. The values of DO vary from 6.66 mg/I to 9.66 mg/1, which was suitable for fishes and other biota. According to the field survey, the farmers in the beet area was found to using 113.14 kg urea per acre against the government recommendation level of 87.0 kg per acre. On the other hand, application of TSP, MP and Gypsum were lower than the government recommendation level. Both the bee! water and sediment were free from Organophosphorus and Carbamate pesticides. However, a very trace amount of Organochlorine (DDT, heptachlor and dieldrin) pesticide was found in the beel sediment and water sample, which was negligible in the context of pollution. From the survey on agrochemical use, it was found that farmers around the Posna beet were not using organochlorine pesticides in HYV Boro field. The presence of a trace amount of Organochlorine pesticides in water and sediment samples may be for the past use of these pesticides for mosquito eradication in this area.
Elemental image and audio synchronization and transmission technique for glasses-free 3D TV system
(BRAC University, 2019-04) Khan, Amit Hasan; Khan, Fairoz Nower; Nisa, Noor E; Jannat, Ashraful; Alam, Ashraful
Previously integral imaging based 3D technology was proposed only for still image. We propose a novel technique to synchronize elemental images and audio signal and the transmission technique for glasses-free 3D TV system based on integral imaging. The main idea behind the method is to generate 3D video based on elemental images synchronized with audio stream. The system uses the depth information and RGB data of per frame of a video through Intel RealSense 3D camera and the audio stream from microphone. The audio le is sampled according to per frame duration of the video and kept in di erent bu ers but having same index. The frames are divided into elemental images using Elemental Image Generation algorithm and the audio signal is synchronized according to the index. Then the stream of elemental images and corresponding audio data is transmitted to data server for storage. HLS streaming protocol is used to stream the TV content. A dedicated web application was made that fetches data from the server and plays video on the user end display device. By using multi-array of lenses in front of display, the video is viewed as three-dimensional with the help of integral imaging technology. The 3D TV content has an output rate of 30 frames per second and a calculated viewing angle of 18.9 . As integral imaging is an auto stereoscopic method to represent depth perception, it frees the viewer from wearing any 3D glasses.
Etiology of sexually transmitted infections among street-based female sex workers in Dhaka, Bangladesh
(2000) Rahman, Motiur; Alam, Ashraful; Nessa, Khairun; Hossain, Anowar; Nahar, Shamsun; Datta, Dilip; Alam, Shahnewaz Khan; Mian, Ruhul Amin; Albert, M. John
Human Activity Recognition Using Smartphone
(Daffodil International University, 2021-01-31) Alam, Ashraful; Das, Anik; Tasjid, Md Shahriar; Marma, Singnuching
Smart devices like smartphones, smartwatches have made this world smarter than any other time at every scale. A lot of facilities can be taken from these devices. Proper use of built-in sensors such as accelerometer, gyroscope, GPS is a few of them. In everyday life, people do a lot of physical activities which can be important for analysis like health state prediction, how much exercise they do etc. by using those sensors based on Artificial Intelligence. In this paper we have implemented both machine learning and deep learning to detect and recognize eight activities with a maximum of 99.3% accuracy. Of those activities few are similar in physical movements and actions like sitting in a chair at home, standing, and sitting in a car. These are almost similar and difficult to distinguish. Going upstairs and downstairs are also almost similar to separate. So we showed that with more sensors and data collection points a wide range of activities can be recognized and the accuracies can be increased. We proved our point by comparing the results of using fewer sensors and again using data of only one position of either pocket or wrist. Then finally we showed that by putting all the sensors and data of pocket, wrist together, we can recognize those activities accurately and in this way, a wide range of activities can be recognized with precision.
Leveraging Sensor Fusion and Sensor-Body Position for Activity Recognition for Wearable Mobile Technologies
(Scopus, 2021) Alam, Ashraful; Das, Anik; Tasjid, Shahriar; Marouf, Ahmed Al
—Smart devices like smartphones and smartwatches have made this world smarter. These wearable devices are created through complex research methodologies to make them more usable and interactive with its user. Various interactive mobile applications such as augmented reality (AR), virtual reality (VR) or mixed reality (MR) applications solely depend on the in-built sensors of the smart devices. A lot of facilities can be taken from these devices with sensors such as accelerometer and gyroscope. Different physical activities such as walking, jogging, sitting, etc., can be important for analysis like health state prediction and duration of exercise by using those sensors based on artificial intelligence. In this paper, we have implemented machine learning and deep learning algorithms to detect and recognize eight activities namely, walking, jogging, standing, walking upstairs, walking downstairs, sitting, sitting-in-a-car and cycling; with a maximum of 99.3% accuracy. A few activities are almost similar in action, such as sitting and sitting-in-a-car, but difficult to distinguish; which makes it more challenging to predict tasks. In this paper, we have hypothesized that with more sensors (sensor fusion) and data collection points (sensor-body positions) a wide range of activities can be recognized and the recognition accuracies can be increased. Finally, we showed that the combination of all the sensors data of both pocket/waist and wrist can be used to recognize a wide range of activities accurately. The possibility of using the proposed methodologies for futuristic mobile technologies is quite significant. The adaptation of most recent deep learning algorithms such as convolutional neural network (CNN) and bi-directional Long Short Time Memory (Bi-LSTM) demonstrated high credibility of the methods presented as experimentation.
Nature of ageing and family care for the elderly in rural Bangladesh
(Dhaka: International Centre for Diarrhoeal Diseases Research, Bangladesh; 2003, 2003) Alam, Ashraful
Performance Analysis of BPSK and QPSK Under Rayleigh Fading Channel
(East West University, 8/23/2017) Hasan, MD Momen; Sarker, Sudip; Alam, Ashraful
A well-defined characterization of the distortions in the wireless signal after affected by noise is analyzed to improve channel estimation and increase detection robustness. The Bit Error Rate (BER) and Signal to Noise Ratio (SNR) of the wireless signal under Rayleigh fading channel has been examined. Simulations confirmed the comparison of BER curves for Binary Phase Shift Keying (BPSK) and Quadrature Phase Shift keying (QPSK) from Adaptive White Gaussian Noise (AWGN) channel model
Treatment failure with the use of ciprofloxacin for gonorrhea correlates with the prevalence of fluoroquinolone-resistant Neisseria gonorrhoeae strains in Bangladesh
(2001) Rahman, Motiur; Alam, Ashraful; Nessa, Khairun; Nahar, Shamsun; Dutta, Dilip K.; Yasmin, Lubna; Monira, Shirajum; Sultan, Zafar; Khan, Shahnewaz Alam; Albert, M. John