Browsing by Author "Islam, Md. Saiful"

Now showing 1 - 20 of 98

3D fabrication of food through software implementation for patients of various diseases and dysphagia
(BRAC University, 2018-07) Farha, Anika; Muhtadi, Mantaka; Morshed, Ramisa Ibnat; Alam, Md. Ashraful; Islam, Md. Saiful
3D printing can be considered as a means of creating solid 3-dimensional objects using additive manufacturing with the aid of computer design software. With respect to time and newer technologies, 3D printing has advanced far more than we have ever anticipated. Though the potential of 3D printing has been under philosophical discussion for some time, creating food using 3D printers has always been a challenge. In order to print 3D food using 3D printing technology, a solid software is required that will count and personalize the nutrient information per patient and individual. In this research, a software has been developed that computes the required macronutrients according to the requirements per individual. Seven diseases have been outlined in this thesis and calculations for ingredients based on different ratios and equations have been made which have been derived after intensive research, from which two unique self-deduced algorithms have been developed to run the software logic. The system takes in the age, height, weight, gender and disease information from the hospital patients’ charts, uses the variables to run the first own algorithm to calculate personalized protein, fat and carbohydrate counts and then based on the diseases the software runs the second own algorithm to calculate the exact amounts of the ingredients required to prepare their meals per day. In this thesis, 3D models have also been designed to maximize user experience and the patient’s visual appeal for the food. The final output of the software is expected to print out a list of the ingredient count per day and display the selected 3D model. Further work of this thesis aims to explore the world of 3D printing and come up with nutrition specific solutions for people suffering from different diseases and have 3D printing technology make their lives easier because the potential of 3D printing of food is vast in a sense that it could provide an exciting alternative to help people customize their food, make them visually more appealing so that people requiring special care would be able to consume food with much more ease which would otherwise have been unachievable through conventional cooking and food preparation techniques
A CASE STUDY ON HATCHERY MANAGEMENT PRACTICES OPERATED BY JAMESWAY INCUBATOR AND HATCHER FOR COBB-500 BROILER PARENT STOCK EGG IN VALUKA HATCHERY. (CP BANGLADESH CO LTD) .
(Chattogram Veterinary & Animal Sciences Universiy, KHULSHI, CHITTAGONG-4225., 2013-02) Islam, Md. Saiful
The study was conducted to focus on the hatchery management practices, hatchability, quality of day-oldchick of broiler parent stock (Cobb-500) operated under Jamesway incubator & hatcher in Valuka Hatchery, CP Bangladesh Co. Ltd. at Mymensingh. The study was undertaken with 272000 day-oldchick over the month of December,2013. Result related to the average body weight gain of DOC was 40 gram. Highest body weight was 58 gm and lowest DOC body weight was 36 gm. The DOC were uniform in body weight. The highest hatchability was 87.86% in November,2013. The culling percentage of DOC was 2-3%. Lower grade chick quantity was negligible. Though feather sexing is uncommon in the context of BD, the studied hatchery was found to undertake this practice.
A comprehensive study of networking systems and their real-world implementation ‘’from theory to practice: my internship experience in SYSOLUTION”
(BRAC University, 2025-11) Tamanna, Sanjida; Islam, Md. Saiful
SYSSOLUTION is a Bangladesh-based software and IT solutions company offering a range of services across various sectors, including healthcare, real estate, fintech, telecom and e-commerce. This business has established itself as a pioneer in the telecom sector, delivering exceptional customer care and ensuring its presence at every step of the value chain. Moreover, it has a data center of its own to support cloud hosting and its managed services portfolio. Cloud-based solutions, enterprise systems, web and mobile application development, and custom software development are all included in the company’s portfolio. SYSSOLUTION seeks to improve user engagement, operational efficiency, and long-term growth for its clients by utilizing cutting-edge technologies and a client-centric strategy. I have learned how to manage a system and generate new ideas from my work at SYSSOLUTION. They encouraged me to learn more about emerging technologies and progress in my network engineering profession. My networking abilities have increased as a result of working with them. Therefore, I have gained new ideas about networking and customer management by working there.
A Critical Review and Prospect of NO2 and SO2 Pollution Over Asia
(Elsevier, 2023-06-10) Jion, Most. Mastura Munia Farjana; Jannat, Jannatun Nahar; Mia, Md. Yousuf; Ali, Md. Arfan; Islam, Md. Saiful; Ibrahim, Sobhy M.; Pal, Subodh Chandra; Islam, Aznarul; Sarker, Aniruddha; Malafaia, Guilherme; Bilal, Muhammad; Islam, Abu Reza Md Towfiqul
Nitrogen dioxide (NO2) and sulfur dioxide (SO2) are two major atmospheric pollutants that significantly threaten human health, the environment, and ecosystems worldwide. Despite this, only some studies have investigated the spatiotemporal hotspots of NO2 and SO2, their trends, production, and sources in Asia. Our study presents a literature review covering the production, trends, and sources of NO2 and SO2 across Asian countries (e.g., Bangladesh, China, India, Iran, Japan, Pakistan, Malaysia, Kuwait, and Nepal). Based on the findings of the review, NO2 and SO2 pollution are increasing due to industrial activity, fossil fuel burning, biomass burning, heavy traffic movement, electricity generation, and power plants. There is significant concern about health risks associated with NO2 and SO2 emissions in Bangladesh, China, India, Malaysia, and Iran, as they pay less attention to managing and controlling pollution. Even though the lack of quality datasets and adequate research in most Asian countries further complicates the management and control of NO2 and SO2 pollution. This study has NO2 and SO2 pollution scenarios, including hotspots, trends, sources, and their influences on Asian countries. This study highlights the existing research gaps and recommends new research on identifying integrated sources, their variations, spatiotemporal trends, emission characteristics, and pollution level. Finally, the present study suggests a framework for controlling and monitoring these two pollutants' emissions.
Abundance, Characteristics and Ecological Risks of Microplastics in River Sediments around Dhaka City
(Department of Civil and Environmental Engineering (CEE), Islamic University of Technology (IUT), Board Bazar, Gazipur, Bangladesh, 2022-11-30) Islam, Md. Saiful
Microplastics (MPs), the small particles of plastics with a size less than 5 mm have been identified as an emerging pollutant in recent decades. Microplastics pose a higher risk in the aquatic environment and also a potential threat to human health. The aquatic species ranging from invertebrates to fishes can easily ingest microplastics along with other contaminants considering MPs as food sources due to their diverse characteristics (size, shape, and color), which accumulate in digestive tracts of aquatic species. Finally, MPs enter into the human body through gastrointestinal ingestion of aquatic species as well as from water consumption and thus create human health risks depending on their toxicity level. Microplastics (MPs) pollution has become an escalating problem in Bangladesh also due to its rapid urbanization, economic growth, and excessive uses of plastics, however data of MPs pollution of fresh water resources is very limited in Bangladesh. This study investigated microplastics pollution in the riverbed sediments in the peripheral rivers of Dhaka city. In total, 28 sediment samples were collected from the selected stations of Buriganga, Turag, and Balu River. A total of about 1 kg of riverbed sediment, 5-10 m away from the shoreline was sampled using an Ekman grab sampler (15×15×15 cm) from top 10 cm of the riverbed at each sampling station. Density separation and wet-peroxidation methods were employed to extract microplastic particles. Attenuated total reflectance-Fourier transform infrared spectroscopy was used to identify the polymers. Scanning electron microscopy (SEM) analysis was performed to examine the surface characteristics of weathered MPs. MPs in the river sediment were found to vary with sampling locations and the abundance of MPs varies from 46 to 534 items per kilogram (kg) of dry sediment. The mass concentration of MPs varies from 13.56 mg/kg to 430.65 mg/kg with an overall average value of 106.52 ± 73.17 mg/kg. The results indicated a medium-level abundance of microplastics in the riverbed sediment in comparison to other studies in the freshwater sediments around the world. The observed MPs particles were shorted into three shapes: films, fragments, and fibers. Films (53.89%) were the most abundant shapes followed by fragments (37.57%), and fiber (8.54%).. The white (18.77%) MPs were major abundance followed by transparent (14.90%), yellow (14.37%), blue (14.37%), red (12.03%), green (11.27%), black (8.40%) and grey (5.87%). Larger quantities of the plastics are generally used in Bangladesh for shopping bags, package products and textile materials, which are white or transparent in color. MPs are categorized into small microplastics (<1 mm) and large microplastics (1-5 mm). The results of this x investigation found that on an average, the riverbed sediments contain large sized MPs (67%) much higher than small sized MPs (33%). The most abundant polymers were polyethylene (PE), polypropylene (PP), and polyethylene terephthalate (PET). The pollution load index (PLI) values more than 1 were observed indicating that all sampling sites were polluted with microplastics. An assessment of ecological risks, using the abundance, polymer types, and toxicity of MPs in the sediment samples suggested a medium to very highlevel ecological risks of microplastics pollution of the rivers. The average ecological risk index (ERI) value suggested that both BR and TR have high ecological risk and BaR has medium ecological risk. In some sampling locations of both BR and TR, ERI value more than 1200 was observed, indicating very high ecological risk to those sampling locations. Higher abundance of MPs and presence of highly hazardous polymers such as polyurethane, acrylonitrile butadiene styrene, polyvinyl chloride, epoxy resin, and polyphenylene sulfide were caused the higher ecological risks. SEM images revealed that the PE, PP, and PET polymers with the carbonyl group had linear fractures, cracks, pits, grooves, granules, and flakes and along with some crystalline formation. However, the same types of particles without carbonyl group had experienced relatively stable surfaces but still contained rough and irregular textures. This textural analysis suggested that MPs particles in riverine sediment were weathering by various processes, producing smaller MPs, which are caused more potential ecological hazards in these river ecosystems. This study indicated that the river ecosystem of the peripheral rivers of Dhaka city is polluted by MPs from the anthropogenic sources both point and non-point in nature. MPs pollution of freshwater bodies is a new dimension of the widespread pollution because of increased use of plastic products, reckless and uncontrolled disposal of municipal solid wastes including plastic wastes, disposal of untreated industrial wastewater including plastic industries and excessive urbanization. Finally, this investigation provided a baseline information on microplastics pollution in the riverine freshwater ecosystem for more in-depth study on risk assessment and developing strategies for controlling microplastics pollution in the country.
An Affordable CanSat Design and Implimentation to Study Space Science for Bangladeshi Students
(IEEE, 2020-06) Raian, F.M. Tanvir Hasan; Islam, H.M. Jahirul; Islam, Md. Saiful; Azam, Rafiul; Islam, H.M. Jahidul; Debnath, Sutapa
This paper presents an educational project of CanSat where the main goal is to build it with the resources available in the local market of Bangladesh within a limited cost so that most of the students can afford it. This project can be used by the students of different engineering disciplines to begin the study of satellite technology. This CanSat can be launched by water rocket and different parameters like temperature, barometric pressure, altitude, latitude, longitude, GPS time, probe orientation, mission states are sent to the ground station right after the launch. The real-time data are received and decoded at the ground station and the computer of the ground station graphically visualizes the data.
An Analysis on Bengali handwritten conjunct character recognition and prediction
(BRAC University, 2021-01) Munawar, Maazin; Roy, Yagghaseni Saha; Hussain, Mohammed Mudabbir; Islam, Md. Saiful
In the very active field of handwriting recognition, a lot of research can be found in the detection of the handwriting of various languages, especially English. However, for languages like Bengali, while they hold some success in handwritten character recognition, a big roadblock is Bengali conjunct characters or “Juktakkhor”. As Bengali conjunct characters are very complex, even today many institutions in Bangladesh still maintain documents as handwritten copies. In this paper, we will present a model that focuses on conjunct character recognition and conversion to textformat. OurproposedsystemwillbetrainedandtestedusingCNNmodelslike VGG19, ResNet-50, GoogleNet, LSTM, ShuffleNet etc. The results generated from preliminary analysis yield that ShuffleNet gives the most accurate results with an accuracy of 91.2% followed by GoogleNet with 73.3%.
An Analytical Technique for Solving Second Order Strongly Damped Nonlinear Oscillator with a Fractional Power Restoring Force
(Khulna University of Engineering & Technology (KUET), Khulna, Bangladesh, 2017-03) Islam, Md. Saiful; Uddin, Dr. Md. Alhaz
In this thesis, an analytical technique has been developed for solving strongly nonlinear damped systems with 1/ 3 x restoring force by combining He’s homotopy perturbation method (HPM) and the extended form of the Krylov-Bogoliubov-Mitropolskii (KBM) method. The presented method has been justified by an example. We have also established the relationship between amplitude and approximate angular frequency. In this study, the presented technique gives desired results avoiding any numerical complexity. Graphical representation of any physical system is important. So, approximate solutions are compared with those numerical solutions obtained by fourth order Runge-Kutta method in graphically. The results in figures show that the approximations are of extreme accuracy with small and significant damping. The presented method is simple and suitable for solving the above mentioned nonlinear damped systems.
An interactive knowledge based recommender system for tourism
(BRAC University, 2017-04) Ahmed, Wakil; Nur, Farah; Hema, Nabila Bhuyan; Ahmed, Nasim; Tairin, Suraiya; Islam, Md. Saiful
The purpose of this project was to construct a centralized tourist management system that would serve as a consolidated platform to provide an effective and efficient mechanism for tourist management and means for availing various travels and booking related services. Providing optimal solution using interactive knowledge-based system using dynamic information of the users, e.g. compare different options for traveling within a budget of the user and providing recommendations. The entire system is designed to streamline the travel management system by primarily targeting all the districts and division of Bangladesh and then centralizing the process maintaining schedules, queues and confirming hotel and transport bookings. Providing optimal solution using interactive knowledge-based system using dynamic information of the users, e.g. compare different options for traveling within a budget of the user and providing recommendations. Currently operations of these procedures are of an erratic where due to delays and other factors of inefficiency management services can neither be properly availed nor be found online. Using this system, the tourism can avail services which are best suited for their need, based on different criteria including specialty, current location, queue and other aspects which are deemed to be of convenience.
An overview of prescription pattern of antibiotic among the different specialist doctors in Faridpur city
(Daffodil International University, 2018-08-03) Islam, Md. Saiful
An overview of prescribing pattern of antibiotics medication was completed for a period of up to 3 months in the territory of Faridpur city. The aims and objectives of this study were to observe the current pattern of antibiotics prescription, most prescribed antibiotics and find out the brand leader of antibiotics so that we can gain a better understanding of prescription pattern of antibiotics and the usage of these agents in different types of patients with different types of diseases that why the data was obtained from patients visiting in different private hospital and public hospital in Faridpur city. During the study period a total of 250 prescriptions were analyzed and out of 250 patients, 157 (62.80%) were male and 93 (37.20%) were female (including children and adults) where most of the patients were outpatients 228 (91.20%). From 250 prescriptions found that 1036 individual drugs were prescribed (an average of 4.14) and 124 prescriptions (49.60%) were prescribed the total four (4) drugs was found to be highest among 250 prescriptions. I found that the antibiotic containing prescription 184 that was 73.60% from total prescription and almost all prescription (100%) contained antibiotic drugs along with other group of drugs. From 250 prescriptions found that the four most commonly prescribed group of drugs were antiulcerants 197 (78.80%), NSAIDs 118 (47.20%), vitamin and minerals 96 (38.40%) and antihistamines 42 (16.80%). From this analysis I found that 41.85% prescription contained single antibiotic drug, 58.15% contained two antibiotic drugs and no prescription contained more than two antibiotic drugs in which maximum drugs (93.13%) were prescribed by their brand names. This study observed that percentages of different generic of antibiotics are Cefuroxime 23.02% is the highest area, and then prescribed commonly Cefixime 17.87%, Ciprofloxacin 16.49%, Levofloxacin 11.68%, Flucloxacillin 9.62%, Amoxicillin 7.90%, Cefpodoxime 2.75%, Cephradine 2.40% and others generic of antibiotics 8.25%. From the study I found that the percentage of the share of antibiotics among the different pharmaceutical company included Opsoninpharma Ltd 24.74% is the highest, Square pharmaceutical limited 19.24%, Popular pharmaceutical Ltd 18.21%, Beximco pharmaceutical limited 12.71%, Incepta pharmaceutical limited 8.59%, and the rest 16.50% antibiotics prescribed form others pharmaceutical company.
Analysing Facebook user risk using machine learning algorithm
(BRAC University, 2020-04) Barua, Arnab; Adnan, Fahim; Ghosh, Ananna; Arif, Hossain; Islam, Md. Saiful
Now-a-days people exchange their personal information and interact with companions and close relatives in a way which is revolutionized. In any case, the majority of them don’t have the foggiest idea how to utilize, where to click, where not to, where to remark, and where not to. A considerable lot of them are posting in Facebook anything they desire and wish. This posting, fellowship and so on once in a while brings shocking occasions like identity theft, phishing, Cyber-wrongdoing and so on. So, Social media security has captured a great concern among the public and authority. At present, many features have been added to reduce the risk of hacking information. It is widely acknowledged that these features have played an important role in the security system. The essential focus point of our paper is on the safety implications of consumers posting their own Facebook information. We have made a survey containing 44 inquiries dependent on Facebook clients’ propensity and different things. We have looked at the ongoing information security rupture on Facebook through certain data mining substances. We have targeted three questions about victim of malware, identity theft, and phishing. From, our dataset we will know how many were victim of the three target parameter. We have implemented machine learning algorithms like ANN, XGBoost, SVM, Random Forest, Decision Tree, Gaussian Naive Bayes, Logistic Regression to identify the percentage of how many Facebook accounts are in risk and safe. Moreover, we will compare the best possible approach and worst approach among the algorithms to find the result. Among the models, we see ANN providing us the best result for the three labels with 89.89%, 94.94% and 86.86%. This research illustrates how different machine learning algorithms predicts the risk of Facebook users and which algorithm is most and least suitable to use in this scenario.
Analysis of Self-Organizing Maps and Explainable Artificial Intelligence to Identify Hydrochemical Factors That Drive Drinking Water Quality in Haor Region
(Elsevier, 2023-12-04) Mia, Md. Yousuf; Haque, Md. Emdadul; Islam, Abu Reza Md Towfiqul; Jannat, Jannatun Nahar; Jion, Most. Mastura Munia Farjana; Islam, Md. Saiful; Siddique, Md. Abu Bakar; Idris, Abubakr M.; Senapathi, Venkatramanan; Talukdar, Swapan; Rahman, Atiqur
Water contamination undermines human survival and economic growth. Water resource protection and management require knowledge of water hydrochemistry and drinking water quality characteristics, mechanisms, and factors. Self-organizing maps (SOM) have been developed using quantization and topographic error approaches to cluster hydrochemistry datasets. The Piper diagram, saturation index (SI), and cation exchange method were used to determine the driving mechanism of hydrochemistry in both surface and groundwater, while the Gibbs diagram was used for surface water. In addition, redundancy analysis (RDA) and a generalized linear model (GLM) were used to determine the key drinking water quality parameters in the study area. Additionally, the study aimed to utilize Explainable Artificial Intelligence (XAI) techniques to gain insights into the relative importance and impact of different parameters on the entropy water quality index (EWQI). The SOM results showed that thirty neurons generated the hydrochemical properties of water and were organized into four clusters. The Piper diagram showed that the primary hydrochemical facies were HCO3−-Ca2+ (cluster 4), Cl---Na+ (all clusters), and mixed (clusters 1 and 4). Results from SI and cation exchange show that demineralization and ion exchange are the driving mechanisms of water hydrochemistry. About 45 % of the studied samples are classified as “medium quality”,” that could be suitable as drinking water with further refinement. Cl− may pose increased non-carcinogenic risk to adults, with children at double risk. Cluster 4 water is low-risk, supporting EWQI findings. The RDA and GLM observations agree in that Ca2+, Mg2+, Na+, Cl− and HCO3− all have a positive and significant effect on EWQI, with the exception of K+. TDS, EC, Na+, and Ca2+ have been identified as influencing factors based on bagging-based XAI analysis at global and local levels. The analysis also addressed the importance of SO4, HCO3, Cl, Mg2+, K+, and pH at specific locations.
Analyzing area-wise air pollution level using machine learning for a better future
(BRAC University, 2021-09) Sihan, Sk. Atik Tajwar; Rabbani, Maisha; Agarwala, Manish; Maliha, Sanjida Alam; Islam, Md. Saiful
Environment consists of nature and surroundings where all living beings co-exist. Harming the environment will in turn harm all living and non-living things alike. One of the major concerns of environment pollution is air pollution, which affects human health, vegetation and aquatic life. However, in developing countries like Bangladesh, air pollution is not considered a major issue. It is mostly caused by the release of harmful gases into the atmosphere. Our goal is to develop a model using machine learning which will determine the level of air pollution in a particular area, detect elements which cause air pollution and predict future pollution level. Algorithms such as Linear Regression, Facebook Prophet, RNN and ARIMA models have been used throughout the course of this study. From RNN we have used LSTM model for prediction which uses special units as well as standard units. With these models we have predicted the pollutant emission rate for analyzing the area-wise pollution rate. We have used different type of algorithms to successfully get the optimum result and to get the fi nal result with less error. This will help to analyze the overall air pollution condition which will help to take necessary steps accordingly.
Android platform to find blood donors and predicting most suitable donors
(BRAC University, 2018-03) Ahmed, Nakib; Rashid, Safoan Bin; Moury, Sharmin Sultana; Nahar, Ankur; Tairin, Suraiya; Islam, Md. Saiful
The need for blood is constant in the medical field in order to save lives, as a number of medical procedures, including most surgical interventions require blood transfusion, but despite all medical and technological advancements, the only viable method available for acquiring blood is through blood donations. Blood donation is a complex time consuming process to find donors who have blood compatibility with patients. As procuring blood posthaste is crucial in life and death situations, our goal is to present an android mobile application that will allow recipients to bridge instant communication with the most apropriate donors forthwith. Our application will work to provide information regarding a requested blood type, and number of available donors around the location. Our aim is to provide an application that will benefit patients in critical situations where there is urgent requirement of a specific blood type. Our platform will connect blood donors and recipients in an effective manner. We intend to provide suggestions regarding the most suitable donors based on blood type, and location, through the implementation of Haversine algorithm. The main objective of this thesis was to create a search platform aimed to locate blood donors, and bring donors and recipient in dire need of blood, to a common platform, so that it could later be used as a medical assistance software.
Automated detection of Malignant Lesions in the ovary using deep learning models and XAI
(BRAC University, 2024-01) Ifty, Md. Hasin Sarwar; Nirjan, Nisharga; Diganta, M.A.; Islam, Labib; Ornate, Reeyad Ahmed; Islam, Md. Saiful; Tasnim, Anika
Cancer is a complex and highly invasive disease that forms due to the abnormal growth of cells in any part of the body. A majority of cancers are unraveled and treated by incorporating advanced technology. However, ovarian cancer remains a dilemma as it has inaccurate non-invasive detection and a time consuming and invasive procedure for accurate detection. Medical professionals are constantly acquiring enhanced diagnostic and treatment abilities by implementing deep learning models to analyze medical data for better clinical decision, disease diagnosis and drug discovery. Thus, in this research, several Convolutional Neural Networks such as LeNet-5, ResNet, VGGNet and GoogLeNet/Inception have been utilized to develop a model that accurately detects and identifies ovarian cancer. For effective model training, the dataset OvarianCancer&SubtypesDatasetHistopathology from Mendeley has been used. After selecting a base model, we utilized XAI models such as LIME, Integrated Gradients and SHAP to explain the black box outcome of the selected model. For evaluating the performance of the base model, Accuracy, Precision, Recall, F1-Score and ROC Curve/AUC have been used. From the evaluation, it was seen that the slightly compact InceptionV3 model with ReLu had the overall best result achieving an average score of 94% across the performance metrics in the augmented dataset. Lastly for XAI, the three aforementioned XAI have been used for an overall comparative analysis. It is the aim of this research that the contributions of the study will help in achieving a better detection method for ovarian cancer.
Autonomous tractor to plough and harvest within a laser perimeter
(BRAC University, 2017-08) Ekram, S.M.A.; Kader, Md. Mobin; Shahera, Fatema tuz; Shirmin, Nusrat; Islam, Md. Saiful
The world as we know is rapidly changing, the world population is ever growing and with this increased population growth the never ending need for food is also growing like never before. But on the contrary, our careers are getting more industrial or corporate based whereas it should be more agriculture based to cope up with the demand of food all over the world. One of the main reason towards this negligence towards agriculture is because farming requires a lot of manual labor and hard work, moreover the profit is also very limited compared to the hard work done, especially in Bangladesh. One of the most difficult task for the farmers in the agriculture field is the ploughing of the land. A lot of manual labor and hard work is needed to do this. Harvesting crops also takes a lot of time and energy. Thus to reduce this huge work load of the farmers we have introduced an autonomous ploughing and harvesting system for the farmers. The plough will be attached at the rear side of the tractor which can be lowered or raised with the help of an actuator. Two blades are attached at the front side of the tractor which rotates with the power of DC motors attached with the blades. It will help to cut down the crops for harvesting. To set a perimeter of the field we have used Laser diodes. To ensure the tractor stays inside the perimeter we added LDRs to the tractor so that it can turn itself when the laser rays hits it at each end of the field. In order to stop the tractor when it is done ploughing the field we used sonar sensor at the end point of the field and RF module to communicate with the tractor, in order to stop it when it reaches the end point and comes in range of the sonar sensor. The tractor is powered by rechargeable Lead Acid batteries. In order to ensure that all the components gets their required powers to complete their function, we used relays and switches. To automatically drive the tractor around the whole field and take the turns at the edges of the field smoothly, we installed glass motors to power the tractor in rotating its wheel in rough surface and IR sensors embedded with encoders in the wheels of the tractor which will help it to turn with precision at both ends of the field. With all these equipment combined our tractor can plough and harvest a whole field automatically without any human intervention, thus reducing the work load of the farmers for ploughing and crop harvesting to a great extent.
Bengali hand sign language recognition using convolutional neural networks
(BRAC University, 2019-04) Rumi, Roisul Islam; Hossain, Syed Moazzim; Shahriar, Ahmed; Islam, Ekhwan; Arif, Hossain; Islam, Md. Saiful
Throughout the world the number of deaf and mute population is rising ever so increasingly. In particular Bangladesh has around 2.6 million individuals who aren't able to communicate with society using spoken language. Countries such as Bangladesh tend to ostracize these individuals very harshly thus creating a system that can allow them the opportunity to communicate with anyone regardless of the fact that they might know sign language is something we should pursue. Our system makes use of convolutional neural networks (CNN) to learn from the images in our dataset and detect hand signs from input images. We have made use of inception v3 and vgg16 as image recognition models to train our system with and without imagenet weights to the images. Due to the poor accuracy we saved the best weights after running the model by setting a checkpoint. It resulted in a improved accuracy. The inputs are taken from live video feed and images are extracted to be used for recognition. The system then separates the hand sign from the image and gets predicted by the model to get a Bangla alphabet as the result. After running the model on our dataset and testing it, we received an average accuracy of 99%. We wish to improve upon it as much as possible in the hopes to make deaf/mute communication with the rest of the society as e ortless as possible.
Bengali Home Assistant
(Daffodil International University, 2019-04) Ahad, Farhan; Hasan, Rakib; Islam, Md. Saiful
Right now we are standing at the edge of the 4th industrial revolution. It been said that this revolution will lead us in a path of where we will let technologies integrate in our day to day life. An intelligent system for home assistant is a desired technology in the 21st century. The main attraction of any home assistant system is reducing human labor, effort, time on their day to day life. The goal of this project is to design a voice control intelligent system based on the only language for which people of Bangladesh gave their lives & blood to gain the rights to talk also the 7th largest language in the world having more than 250 million native speakers. Using remote control system via World Wide Web or Internet gives the ability to control home appliances from anywhere in the world. Various sensor based control, facial recognition, speech pattern recognition can be added to this prototype to improve the intelligence and to improve the ability to make more accurate decision.
Bioaccumulation and sources of metal(loid)s in fish species from a subtropical river in Bangladesh: a public health concern
(Scopus, 2024-12-07) Ali, Mir Mohammad; Kubra, Khadijatul; Alam, Edris; Mondol, Anwar Hossain; Islam, Md. Saiful; Karim, Ehsanul; Ahmed, A. S. Shafiuddin; Siddique, Md. Abu Bakar; Malafaia, Guilherme; Ahmed, A. S. Shafiuddin; Siddique, Md. Abu Bakar
Toxic metals and freshwater fish’s metalloid contamination are significant environmental concerns for overall public health. However, the bioaccumulation and sources of metal(loids) in freshwater fishes from Bangladesh still remain unknown. Thus, the As, Pb, Cd, and Cr concentrations in various freshwater fish species from the Rupsha River basin were measured, including Tenualosa ilisha, Gudusia chapra, Otolithoides pama, Setipinna phasa, Mystus vittatus, Glossogobius giuris, and Pseudeutropius atherinoides. An atomic absorption spectrophotometer was used to determine metal concentrations. The mean concentrations of metal(loids) in the fish muscle (mg/kg) were found to be As (1.53) > Pb (1.25) > Cr (0.51) > Cd (0.39) in summer and As (1.72) > Pb (1.51) > Cr (0.65) > Cd (0.49) in winter. The analyzed fish species had considerably different metal(loid) concentrations with seasonal variation, and the distribution of the metals (loids) was consistent with the normal distribution. The demersal species, M. vittatus, displayed the highest bio-accumulative value over the summer. However, in both seasons, none of the species were bio-accumulative. According to multivariate statistical findings, the research area’s potential sources of metal(loid) were anthropogenic activities linked to geogenic processes. Estimated daily intake, target hazard quotient (THQ), and carcinogenic risk (CR) were used to assess the influence of the risk on human health. The consumers’ THQs values were < 1, indicating that there were no non-carcinogenic concerns for local consumers. Both categories of customers had CRs that fell below the permissible range of 1E − 6 to 1E − 4, meaning they were not at any increased risk of developing cancer. The children’s group was more vulnerable to both carcinogenic and non-carcinogenic hazards. Therefore, the entry of metal(loids) must be regulated, and appropriate laws must be used by policymakers
Brain hemorrhage detection using hybrid machine learning algorithm
(BRAC University, 2022-01) Iqbal, Khondoker Nazia; Azad, Istinub; Emon, Md. Imdadul Haque; Amlan, Nibraj Safwan; Aporna, Amena Akter; Islam, Md. Saiful; Rahman, Rafeed
Machine learning (ML) helps computers learn and program data without humans’ help. According to data scientists, machine learning can extract 60% high-quality information, reduce the cost up to 46%, and increase operation speed by approximately 48% [1]. Recently, there has been successful implementation of machine learning in data analysis, computer vision, computer-aided diseases (CAD), and many more fields. Machine learning is broadly used in the medical industry because of its processing power for image data and pattern recognition quality. The image processing power of machine learning can be used in medical images to classify the brain images automatically. Segmentation and classification of brain image can provide valuable information and quantitative assessment of lesions which can be used for treatment strategies and predicting patient condition (Kamnitsas et al., 2017). According to research [2], an estimated 64-74 million people in the world are affected by traumatic brain injury every year. It affects the lives of nearly every one out of six persons. In our proposed system, we will use a hybrid approach of multiple machine learning algorithms together for the classification of CT brain images and diagnose brain disorders and diseases like brain hemorrhage. Some ML algorithms such as different 3D Convolutional Neural Networks (CNN) , AlexNet, DenseNet121, GoogleNet and some other models like Multilayer Perceptron Model (MLP), Support Vector Machine (SVM) and Random Forest (RF) have been applied successfully in this field in the past. Modifying previous methods, we want to build a hybrid machine learning algorithm by combining different CNN models like VGG-16, VGG-19, Random forest and Multilayer Perceptron (MLP) classifiers for detecting brain hemorrhage. We have used the VGG-16 and VGG-19 model to derive image features from the CT brain images and Random forest classifier and MLP classifier for testing the accuracy of our model. To test the efficiency of our system, we have used CT brain image datasets from Kaggle. The CT brain imaging data will be the input of our model and our model will detect brain hemorrhage and classify them into one of six classes: Epidural, Intraparenchymal, Intraventricular, Subarachnoid, Subdural and No Hemorrhage. Using our hybrid approach the best accuracy we achieved was around 97.24% using a combined approach of VGG-16 and Multilayer Perceptron classifier. Also we used Explainable AI to explain the prediction of the hemorrhagic classes.