Browsing by Author "Rahman, Md. Habibur"

Now showing 1 - 20 of 53

A Deep Learning Approach to Learn Lip Sync from Audio
(Daffodil International University, 2021-01-14) Bhuiyan, Mazharul Islam; Mahato, Milon; Hassan, Md. Nazmul; Rahman, Md. Habibur
With accurate lip-sync of speaker independent, we synthesize several high-quality videos for sake of generating expected target video clip from the composite synthesized video. In our work we explore several related works of lip-syncing, out-of-sync, talking face generation, speaker independent target video content creation from input audio stream containing some limitation & failure and we also implement as better lip-sync by training our models which is not rely on specific speaker. Besides, in this work we detect key reason for the mentioned problems and improve the difficult factors with new evaluation strategies then solve as better output returning like Wav2lip model. By the way, more realistic matched lip-syncing appearance of any individual speaking video from any voices or input audio clip along with mapping RNN is an incredible outcome since it can generate proper mouth texture.
An Exploration Of The Igloo Ice-Cream and Milk Unit Internship
(Daffodil International University, 2024-02-12) Rahman, Md. Habibur
The internship was conducted at the Igloo Ice-Cream and Milk Unit in Shampur, Kadamtali, Dhaka, from 1 st to 31st August, 2023. The primary product of this plant is various kinds of ice cream. Fruit pulp, butter oil, coconut oil, sugar, stabilizer, emulsifier, water, food-grade taste, skim milk powder, glucose, and food-grade color are all needed to make ice cream. For quality control, they primarily examine physical, chemical, and microbiological testing. The creation of several certified, risk-free items is the foundation of my report. I've seen the many varieties of milk and ice cream. The report includes details about the company, a flowchart showing the production process, various industrial operations, concepts related to the manufacturing of machinery, and more.
An Investigation and Evaluation of N-Gram, TF-IDF and Ensemble Methods in Sentiment Classification
(Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST, 2020) Rahman, Sheikh Shah Mohammad Motiur; Biplob, Khalid Been Md. Badruzzaman; Rahman, Md. Habibur; Sarker, Kaushik; Islam, Takia
In the area of sentiment analysis and classification, the performance of the classification tasks can be varied based on the usage of text vectorization and feature extraction methods. This paper represents a detailed investigation and analysis of the impact on feature extraction methods to attain the highest classification accuracy of the sentiment from user reviews. Unigram, Bigram and Trigram are applied as n-gram vectorization models with TF-IDF features extraction method individually. Accuracy, misclassification rate, Receiver Operating Characteristics (ROC) and recall-precision are used in this study to evaluate which are counted as the most important performance measurement parameters in machine learning based approaches. Parameters are measured by the output obtained from Bagged Decision Tree (BDT), Random Forest (RF), Ada Boost (ADA), Gradient Boost (GB) and Extra Tree (ET). The outcomes of this study is to find out the best fitted combination of term frequency–inverse document frequency (TF-IDF) and n-grams for different data size.
An Overview of Antimicrobial Stewardship Optimization
(Daffodil International University, 2022-05-20) Rahman, Md. Mominur; Tumpa, Mst. Afroza Alam; Zehravi, Mehrukh; Sarker, Md. Taslim; Yamin, Md.; Islam, Md. Rezaul; Harun-Or-Rashid, Md.; Ahmed, Muniruddin; Ramproshad, Sarker; Mondal, Banani; Dey, Abhijit; Damiri, Fouad; Berrada, Mohammed; Rahman, Md. Habibur; Cavalu, Simona
Antimicrobial s are a type of agent widely used to prevent various microbial infections in humans and animals. Antimicrobial resistance is a major cause of clinical antimicrobial therapy failure, and it has become a major public health concern around the world. Increasing the development of multiple antimicrobial s has become available for humans and animals with no appropriate guidance. As a result, inappropriate use of antimicrobial s has significantly produced antimicrobial resistance. However, an increasing number of infections such as sepsis are untreatable due to this antimicrobial resistance. In either case, life-saving drugs are rendered ineffective in most cases. The actual causes of antimicrobial resistance are complex and versatile. A lack of adequate health services, um-optimized use of antimicrobial s in humans and animals, poor water and sanitation systems, wide gaps in access and research and development in healthcare technologies, and environmental pollution have vital impacts on antimicrobial resistance. This current review will highlight the natural history and basics of the development of antimicrobial s, the relationship between antimicrobial use in humans and antimicrobial use in animals, the simplistic pathways, and mechanisms of antimicrobial resistance, and how to control the spread of this resistance.
An Overview of Diabetic Foot Ulcers and Associated Problems with Special Emphasis on Treatments with Antimicrobials
(Daffodil International University, 2022-07-14) Baig, Mirza Shahed; Banu, Ahmadi; Zehravi, Mehrukh; Rana, Ritesh; Burle, Sushil S.; Khan, Sharuk L.; Islam, Fahadul; Siddiqui, Falak A.; Massoud, Ehab El Sayed; Rahman, Md. Habibur; Cavalu, Simona
One of the most significant challenges of diabetes health care is diabetic foot ulcers (DFU). DFUs are more challenging to cure, and this is particularly true for people who already have a compromised immune system. Pathogenic bacteria and fungi are becoming more resistant to antibiotics, so they may be unable to fight microbial infections at the wound site with the antibiotics we have now. This article discusses the dressings, topical antibacterial treatment, medications and debridement techniques used for DFU and provides a deep discussion of DFU and its associated problems. English-language publications on DFU were gathered from many different databases, such as Scopus, Web of Science, Science Direct, Springer Nature, and Google Scholar. For the treatment of DFU, a multidisciplinary approach involving the use of diagnostic equipment, skills, and experience is required. Preventing amputations starts with patient education and the implementation of new categorization systems. The microbiota involved in DFU can be better understood using novel diagnostic techniques, such as the 16S-ribosomal DNA sequence in bacteria. This could be achieved by using new biological and molecular treatments that have been shown to help prevent infections, to control local inflammation, and to improve the healing process.
Analytical Approach for Modeling and Simulation of Photonic Crystal Fiber Based on Low Effective Material Loss
(Springer, 2023-03-06) Rahman, Md. Habibur; Rana, Md. Masud; Hossain, Md. Selim; Sen, Shuvo; Al‑Amin, Md.
COMSOL Multiphysics simulation software has been used to create a hexagonal photonic crystal fiber (H-PCF) with hexagonal cladding and a rotating hexa elliptical shape core. The hexagonal photonic crystal fiber (H-PCF) fiber is built on five layers of circular air holes, and it is suitable for telecommunication applications especially optical fiber communication in the terahertz (THz) frequency range. The hexagonal photonic crystal fiber (H-PCF) is designed to have an ultra-low effective material loss (EML), a higher core power fraction, a bigger effective area, and reduced confinement loss. The smallest effective material loss from the proposed hexagonal photonic crystal fiber (H-PCF) is 0.00689 cm−1, with a better core power fraction of 82%, less confinement loss of 3.45 × 10–14 cm−1 and a better effective area of 3.65 × 10–4 m2 is achieved at one terahertz (THz) waveguide region. Furthermore, using the features of the V-Parameter, our developed hexagonal photonic crystal fiber (H-PCF) fiber reveals an optical waveguide with one mode throughout a frequency range of terahertz (THz) wave area. So, it has been said that our hexagonal photonic crystal fiber (H-PCF) structure will be highly beneficial for optical fiber communications applications in the THz frequency range.
Anti-Inflammatory and Analgesic Activities of Ethanolic Extract of Gynura Nepalensis (Leaf)
(PharmacologyOnLine, 2018-12-30) Rahman, Md Mizanur; Rahman, Md. Habibur; Bassar, Abul; Oliullah, Md
The present communication attempts to evaluate the Anti-inflammatory and Analgesic activities of ethanolic extract of Glochidion Thomsonii (Bark) belongs to the family: Euphorbiaceae, which is widely distributed in Bangladesh and India. Methods: Ethanolic extract of Glochidion Thomsonii at doses of 250mg/kg and 500mg/kg were tested for their anti-inflammatory activity against Carrageenan induced paw edema & Xylene-Induced Ear Edema test in rats and analgesic activity was tested against Formalin-induced Paw licking test & acetic acid induced writhing (in mice) models and the two phases of formalin induced paw licking models of pain. Results: The results indicate that the ethanolic extracts at the doses of 250 and 500mg/kg body weight significantly inhibited, the carrageenan induced paw edema and Xylene-induced ear edema models of inflammation, similarly the extract at the doses of 250 and 500mg/kg significantly inhibited the acetic acid induced writhing (in mice) models and the two phases of formalin induced paw licking models of pain. Conclusion: This study confirms that the ethanolic extract of Glochidion Thomsonii has good analgesic and anti-inflammatory property. Further studies are necessary to isolate the compounds from the ethanolic extract of Glochidion Thomsonii responsible for the activity. PhOL Rahman, et al. 318 (page 317-323)
Automated License Plate Detection and Recognition
(Daffodil International University, 22-09-12) Rahman, Md. Habibur; Islam, Md. Rakibul
Automated license plate recognition is important in many contexts like security and law enforcement, monitoring vehicles, automated parking control, etc. To enable these automated services, we are reviewing and combining several established methods in this paper. There were three steps involved in reading car license plates: plate detection, plate extraction, and character recognition. Each stage has many sub-steps. For every sub-step, we have reviewed many methods, and chosen the one that proved to be the best solution after thorough testing and observation. The main objective of this research is to gain high accuracy using as less CPU Time as possible, keeping into consideration the facts like- lighting conditions, vehicle motion, noisy plates, and segmented words in the input image. Our primary target of this thesis is to extract a clean image of license plates of private or community vehicles. Although we target our system to be able to detect standard license plates, we also tested our methods on non-standard plates.
Break-Down Help Assistance: A Helping Assistance App for Vehicles on the Road
(Daffodil International University, 2025-01-13) Rahman, Md. Habibur
The primary objective of the app is to facilitate immediate roadside assistance for drivers experiencing vehicle breakdowns by connecting them with the nearest available garages and mechanics. It aims to ensure quick, efficient, and reliable support, reducing driver stress and minimizing delays caused by mechanical issues. The app leverages location-based services to match drivers with nearby service providers, enhancing safety and convenience. The Android based application named Breakdown Help Assistance to provide immediate service for the drivers when they fall in danger. This Android app is designed to provide immediate roadside assistance for vehicle drivers experiencing breakdowns. The app connects drivers in distress with nearby garages and mechanics, ensuring rapid response and support. When a breakdown occurs, the driver can use the app to request emergency help by sending a service request to the nearest available garage. The request includes the driver’s location and vehicle details, allowing mechanics to assess the situation and respond efficiently. Mechanics can view incoming requests, accept jobs, and navigate to the driver's location for on-site assistance or towing services. The app aims to enhance safety, minimize downtime, and provide a seamless experience for both drivers and service providers by leveraging real-time GPS tracking and a user-friendly interface.
COVID-19 Diagnostic Methods in Developing Countries
(Daffodil International University, 2022-05-27) Maniruzzaman, Md.; Islam, Md. Missile; Ali, Md. Hazrat; Mukerjee, Nobendu; Maitra, Swastika; Kamal, Mohammad Amjad; Ghosh, Arabinda; Castrosanto, Melvin A.; Alexiou, Athanasios; Ashraf, Ghulam Md.; Tagde, Priti; Rahman, Md. Habibur
COVID-19 has become one of the few leading causes of death and has evolved into a pandemic that disrupts everyone’s routine, and balanced way of life worldwide, and will continue to do so. To bring an end to this pandemic, scientists had put their all effort into discovering the vaccine for SARS-CoV-2 infection. For their dedication, now, we have a handful of COVID-19 vaccines. Worldwide, millions of people are at risk due to the current pandemic of coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2). Despite the lack of clinically authorized antiviral medications and vaccines for COVID-19, clinical trials of many recognized antiviral agents, their combination, and vaccine development in patients with confirmed COVID-19 are still ongoing. This discovery gave us a chance to get immune to this disease worldwide and end the pandemic. However, the unexpected capacity of mutation of the SARS-CoV-2 virus makes it difficult, like the recent SAS-CoV-2 Omicron variant. Therefore, there is a great necessity to spread the vaccination programs and prevent the spread of this dreadful epidemic by identifying and isolating afflicted patients. Furthermore, several COVID-19 tests are thought to be expensive, time-consuming, and require the use of adequately qualified persons to be carried out efficiently. In addition, we also conversed about how the various COVID-19 testing methods can be implemented for the first time in a developing country and their cost-effectiveness, accuracy, human resources requirements, and laboratory facilities.
Design and Development of Bi Directional Power Meter Using Microcontroller
(Indonesian Journal of Electrical Engineering and Computer Science, Institute of Advanced Engineering and Science, 2020-03-03) Rahman, Tasnim Ikra; Rashid, Anika Fariha; Rahman, Md. Habibur
The significance of power or energy measurement has been very acute in recent times with the increase of Grid-Tied Solar PV system that needs Net-Metering. Power can be flown in both directions, from grid to load or from load (PV- Panel) to grid line. If power flow from both side is not calculated and surplus power which is not being used at the time instant is not stored, accurate billing is not possible. That is why a bidirectional power meter is required which can measure both the positive and negative power at any instant. The aim of this paper is to develop a bidirectional power meter using microcontroller. The work of this paper constitutes a microcontroller, a current sensor, bulbs as load and a display. The rms values of voltage and current is determined and showed in the display. Then the calibration is done inside the developed program. After that the power is calculated and showed in the display. Then the direction of the power is determined depending on the direction (whether positive or negative) of the power which is also dependent on the phase difference between the voltage and current. If there is no phase difference between the voltage and current, then the power is positive and leaving from the grid. If there is 180 degree phase difference between the voltage and current, then the power is negative and entering to the grid. The direction of the power flow is very important for efficiently of the proposed meter. For some defined load the accuracy of the developed meter is calculated as 94%.
Design of an Off-grid Rooftop Solar Pv System for Younus Khan Scholars Garden-2, Daffodil Internation University
(Daffodil International University, 2021-06) Ahmmed, Md. Raju; Rahman, Md. Habibur
Development of Renewable Energy in Bangladesh is an alternative solution for an increasing of fuel-based energy demand problem. Higher educational institutes may contribute to reducing the energy consumption by implementing green campus policies which include creating a medium-scale solar power installation. This work presents the feasibility analysis of photovoltaic power generation plants development in a Younus khan scholars garden-02 Daffodil International University hostel. In this paper, different off-grid solar rooftop systems were considered in one of the campuses located in rural area. The methodology applied provides a useful and simple approach for sizing and analyzing the hybrid systems using HOMER, an optimization model for renewable energies. This study assesses the potential of a solar system to electrify a Hostel in Daffodil International University. The solar Optimization of Multiple Electric Renewables model is used to assess primary data, develop a load profile and identify the optimal least cost system option for the Hostel. A sensitivity analysis was performed to determine the effect of variations in solar radiation, diesel price on optimal system configurations. The results show that a PV system with a combination of photovoltaic array, converter, battery and diesel generator is the best option from an economic point of view. To meet the Younus khan scholars garden-02 Daffodil International University daily peak demand of 151.60 kW, energy generation cost is estimated at 32.21 BDT ($0.380) per kilowatt hour and net present cost at $1.4M. The optimal system allows for a reduction of 298,450 kg/yr of carbon dioxide emissions per year compared with diesel-only electricity generation. The technical viability of the proposed solar energy system is analyzed using HOMER software. The economic aspects of the plant are analyzed based on standard parameters, the NPC (Net Present Cost) and IRR (Internal Rate of Return) methods. The simulation studies are carried out to identify technical and cost-effective configuration. Feasibility study of the Photovoltaic energy generation was conducted by revealing their potential contributions and applicability’s. This study gives emphasis to the techno-economic analysis of renovating the energy supply system of an off grid-connected Daffodil International University hostel building through a Renewable Energy System. Finally, results of the research can be used as reference for the relevant stakeholders and policymakers in developing the off-grid- solar PV systems for power generation in Bangladesh.
Detection of Molecular Signatures and Pathways Shared in Inflammatory Bowel Disease and Colorectal Cancer
(Genomics, Elsevier Inc, 2020-09-01) Al-Mustanjid, Md.; Mahmud, S.M. Hasan; Royel, Md. Rejaul Islam; Rahman, Md. Habibur; Islam, Tania; Rahman, Md. Rezanur; Ali Mon, Mohammad; Mahmud, S.M. Hasan
Emerging evidence indicates IBD is a risk factor for the increasing incidence of colorectal cancer (CRC) development. We used a system biology approach to identify common molecular signatures and pathways that interact between IBD and CRC and the indispensable pathological mechanisms. First, we identified 177 common differentially expressed genes (DEGs) between IBD and CRC. Gene set enrichment, protein-protein, DEGs-transcription factors, DEGs-microRNAs, protein-drug interaction, gene-disease association, Gene Ontology, pathway enrichment analyses were conducted to these common genes. The inclusion of common DEGs with bimolecular networks disclosed hub proteins (LYN, PLCB1, NPSR1, WNT5A, CDC25B, CD44, RIPK2, ASAP1), transcription factors (SCD, SLC7A5, IKZF3, SLC16A1, SLC7A11) and miRNAs (mir-335-5p, mir-26b-5p, mir-124-3p, mir-16-5p, mir-192-5p, mir-548c-3p, mir-29b-3p, mir-155-5p, mir-21–5p, mir-15a-5p). Analysis of the interaction between protein and drug discovered ASAP1 interacts with cysteine sulfonic acid and double oxidized cysteine drug compounds. Gene-disease association analysis retrieved ASAP1 also associated with pulmonary and bladder neoplasm diseases
Development of Strategic Fit Model of Manufacturing Unit for Garments Industry
(Khulna University of Engineering & Technology (KUET), Khulna, Bangladesh, 2019-04) Rahman, Md. Habibur; Rahman, Dr. Azizur
Though Bangladesh is one of the largest garments manufacturing country in the world and there happened revolutionary changes more than four decades ago, they couldn’t achieve sustainable platform yet. The failure to achieve up to the requirement level for the competitive capabilities/manufacturing metrics is the common phenomenon for the manufacturers. Even there is an alarming issue that the manufactures yet don’t know how they are affected by these failures and also can’t measure how much they are statistically fit. The manufacturers fail to compete with their competitors, since they can’t achieve their manufacturing metrics up to the requirements. To proceed towards world class manufacturing and to create a sustainable platform considering highly competitive business market, the manufacturer should aware about their metrics capabilities, competitive capabilities and reasons of metrics failure. By being motivated from manufacturer’s failure, we worked on a manufacturing unit of a garments industry where we aggregated manufacturing metrics and determined how the manufacturers are affected by the failure of these metrics. This research work will conclude by proposing few models with their mathematical and graphical explanation. By using these models, the manufacturers will be able to determine their strategic fitness, security level and associated loss/penalty.
Effect of Crop Establishment and Weed Control Method on Productivity of Transplanted aman Rice
(Weed & Turfgrass Science, 2018) Ali, Mohammad; Bir, Md. Shahidul Haque; Rahman, Md. Habibur; Ayesha, Sultana Kaniz; Hoque, Aminul; Harun-Ar-Rashid, Md.; Islam, Md. Rashidul; Park, Kee Woong
This experiment was conducted to find the most suitable crop establishment method and weed management practices for transplanted aman rice in Bangladesh. Rice variety Bangladesh Rice Research Institute (BRRI) dhan44 was used as planting materials where three crop establishment methods (T1: direct wet seeding by drum seeder; T2: hand broadcasting; T3: transplanting) and four weeding options (W1: Hand weeding (HW); W2: BRRI weeder+HW; W3: Herbicide+HW; and W4: no weeding) were tested. Among the crop establishment methods, the highest grain yield (5.12 t ha-1) was obtained with the T3, while the highest benefit cost ratio (BCR) of 2.08 was found in T2. In case of the weed management method, W1 showed superior results on the plant density/m2 (139.66) at 60 days after transplanting (DAT), grain yield (4.97 t ha-1), and BCR (2.03). On the other hand, the highest plant dry matter (36.20 g) at 60 DAT and the highest yield (6.10 t ha-1) were obtained in a T3W1 combination. The results of this study show that the productivity of rice during aman season could be most significantly increased with the use of transplanting (T3) alone, hand weeding (W1) alone, or a combination of the two methods (T3W1).
Effect of crop establishment method and N management practice on productivity of transplanted aman rice
(Bioscience Research, 2018) Bir, Md. Shahidul Haque; Ali, Mohammad; Rahman, Md. Habibur; Ayesha, Sultana Kaniz; Hoque, Aminul; Rashid, Md. Harun-Ar; Islam, Md. Rashidul; Cho, Jin-Woong; Park, Kee Woong
An experiment was conducted at the Bangladesh Rice Research Institute Regional Station, Rajshahi, Bangladesh to find the most suitable crop establishment method and best application of N management practice for transplanted aman rice. The monsoon-season rainfed rice is aman, which is planted in two ways: direct seeding in March or April and transplantation between July and August. Both types are harvested from November through December. T. aman rice variety, BRRI dhan44 was used as planting material where three crop establishment methods (T1: direct wet seeding by drum seeder, T2: hand broadcasting and T3: transplanting) and four N management practices (N1: ⅓ at 15 DAT+ ⅓ at 30 DAT +⅓ at panicle initiation (PI), N2: ¼ at basal + ¼ at 15 DAT + ¼ at 30 DAT + ¼ at PI, N3: leaf color chart (LCC) based N application, N4: no nitrogen) were tested. Among crop establishment methods, the transplanting method (T3) produced the highest plant density m-2 (143.16) and yield (5.05 t ha–1 ). Among N management practices, N2 gave the greatest yield (4.97 t ha–1 ) and the highest benefit cost ratio (1.97). Treatment combination of T3N2 at 60 DAT produced the highest plant density m-2 (152.33). Therefore, from the current study, it can be concluded that the transplanting method (T3) and N split fertilization of ¼ at basal + ¼ at 15 DAT + ¼ at 30 DAT + ¼ at PI (N2), singly or together, would be the most productive application treatments during aman season.
Energy Consumption of DIU Buildings: A Study for Future
(Daffodil International University, 23-01-18) Salauddin, Md.; Rahman, Md. Habibur; Seum, Md. Minhaz Uddin
In this paper, we propose a framework to characterize the energy consumption of DIU Buildings and their subcomponents (e.g., connected light, fan, air condition (AC), lift, street light, campus light, power socket, water pump, IT accessories, CL light, etc). This framework serves as a key component in describing the environmental impacts of decisions in manufacturing operations design and control, especially at the sub-cellular level. Furthermore, the suggested framework may utilize to investigate the inter-relationship of operational decisions and environmental impacts within a manufacturing plant. The proposed framework for gathering energy data is illustrated in the paper's conclusion. Keywords: Data Analysis, Energy Benchmarking, Load Curve, Energy Consumption, Energy Monitoring System.
Enhancing Quality of Service in SDN Based on Multi-path Routing Optimization with DFS
(2019 International Conference on Sustainable Technologies for Industry 4.0 (STI), IEEE, 2019-12-25) Hossen, Md. Sajid; Rahman, Md. Habibur; Al-Mustanjid, Md.; Nobin, Md. Arif Shakil; Habib, Md. Ahsan
Remarkable growth of network traffic has been noticed recently due to the use of online applications and cloud services. Thus, multiple routing problems need to solve without increasing latency to optimize the traffic. Based on the responses from industry, Software Defined Networking (SDN) has been developed to meet this challenge. SDN allows the programmable features of network and responds dynamically. In this paper, multipath routing problem in SDN is addressed and resolved under OpenFlow protocol where Ryu controllers are used to implement SDN using the Mininet WiFi emulator. Besides, in addition to a graphical user interface, a module has been developed using python programming along with OpenFlow based Ryu controller. In the simulation results, depth first search (DFS) algorithm outperforms over breadth first search (BFS) in terms of round trip time (RTT), delay and throughput. Hence, this article enhanced the quality of service (QoS) in SDN.
Helping Healthcare Providers To Differentiate COVID-19 Pneumonia by Analyzing Digital Chest X-Rays
(Daffodil International University, 22) Ullah, Abu Naser Zafar; Rahman, Md. Habibur; Allayear, Shaikh Muhammad; Khan, Mohammed Liakwat Ali; Sheikh Md. Faysal; Chowdhury, ABM Alauddin; Uddin, Md. Nasir; Khan, Hafiz T. A.
Background: Detecting COVID-19 pneumonia and differentiating it from community acquired pneumonia (CAP) has been a challenging task for healthcare providers since the pandemic began. We therefore aim to develop and evaluate a simple, noninvasive tool to accurately detect COVID-19 by using digital chest X-ray (CXR). Methods and Results: We performed a retrospective, multi-center study in which deep learning frameworks were used to develop the system architecture of the diagnostic tool. The tool was trained and validated by using data from the GitHub database and two hospitals in Bangladesh. Python programming was used to calculate all statistical estimates. Our study revealed that the artificial intelligence (AI)-based diagnostic tool was able to detect COVID-19 accurately by examining chest X-ray (CXR). During the testing phase, the tool could interpret CXR with precision of 0.98, recall/sensitivity of 0.97 and F1 score of 0.97 for COVID-19. The evaluation results showed high sensitivity (90%) and specificity (92%) in detecting COVID-19. The AUC values for COVID-19 and pneumonia were 0.91 and 0.87, respectively. Conclusion: The developed AI-based diagnostic tool can offer the healthcare providers an effective means of detecting and differentiating COVID-19 from other types of pneumonia, thus contributing to reducing the long-term impact of this deadly disease.(International Journal of Biomedicine. 2022;12(3):459-465.) Key Points • The major strength of this study is that it has led to the development of a technology-based tool that can precisely detect COVID-19 at an early stage and immediately isolate infected patients from the healthy population. • This next generation test can be accessed by healthcare providers remotely and therefore provide an added convenience. • The main limitation is with this tool COVID-19 pneumonia may be confused with other types of pneumonia if the quality of the chest image is too low. Moreover, it focuses only on detecting whether or not the disease is COVID-19, but not its severity. Keywords: COVID-19 • chest X-Ray • artificial intelligence • Bangladesh
Implementation of Low Cost Real-time Attendance Management System
(ICRITO 2020 - IEEE 8th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions), IEEE, 2020-09-15) Hasan, Rakib; Islam, Sajedul; Rahman, Md. Habibur; Saifuzzaman, Mohd.; Shetu, Syeda Farjana; Moon, Nazmun Nessa
In this research, a systematic approach was adopted to expand the level of service to the real-time attendance system. The test was performed to assess the efficiency of the attendance system, which would be quite cost-effective but accurate. To that the expense of linking our biometric attendance system to a central server where the data is changed instantly. So, the possibility of data loss is nothing at all. Our proposed prototype may cost around 30$ while the available existing devices are high priced. Moreover, most devices generate abundant data in favor of an individual user where our proposed model provides well organized data to track down the attendance. Furthermore, a very low amount of energy is needed to run the entire system. The entire system can be installed anywhere with a very low expense