-
A deep learning approach for brain tumor detection using magnetic resonance imaging
Authors:
Al-Akhir Nayan,
Ahamad Nokib Mozumder,
Md. Rakibul Haque,
Fahim Hossain Sifat,
Khan Raqib Mahmud,
Abul Kalam Al Azad,
Muhammad Golam Kibria
Abstract:
The growth of abnormal cells in the brain's tissue causes brain tumors. Brain tumors are considered one of the most dangerous disorders in children and adults. It develops quickly, and the patient's survival prospects are slim if not appropriately treated. Proper treatment planning and precise diagnoses are essential to improving a patient's life expectancy. Brain tumors are mainly diagnosed using…
▽ More
The growth of abnormal cells in the brain's tissue causes brain tumors. Brain tumors are considered one of the most dangerous disorders in children and adults. It develops quickly, and the patient's survival prospects are slim if not appropriately treated. Proper treatment planning and precise diagnoses are essential to improving a patient's life expectancy. Brain tumors are mainly diagnosed using magnetic resonance imaging (MRI). As part of a convolution neural network (CNN)-based illustration, an architecture containing five convolution layers, five max-pooling layers, a Flatten layer, and two dense layers has been proposed for detecting brain tumors from MRI images. The proposed model includes an automatic feature extractor, modified hidden layer architecture, and activation function. Several test cases were performed, and the proposed model achieved 98.6% accuracy and 97.8% precision score with a low cross-entropy rate. Compared with other approaches such as adjacent feature propagation network (AFPNet), mask region-based CNN (mask RCNN), YOLOv5, and Fourier CNN (FCNN), the proposed model has performed better in detecting brain tumors.
△ Less
Submitted 25 October, 2022;
originally announced October 2022.
-
IRHA: An Intelligent RSSI based Home automation System
Authors:
Samsil Arefin Mozumder,
A S M Sharifuzzaman Sagar
Abstract:
Human existence is getting more sophisticated and better in many areas due to remarkable advances in the fields of automation. Automated systems are favored over manual ones in the current environment. Home Automation is becoming more popular in this scenario, as people are drawn to the concept of a home environment that can automatically satisfy users' requirements. The key challenges in an intel…
▽ More
Human existence is getting more sophisticated and better in many areas due to remarkable advances in the fields of automation. Automated systems are favored over manual ones in the current environment. Home Automation is becoming more popular in this scenario, as people are drawn to the concept of a home environment that can automatically satisfy users' requirements. The key challenges in an intelligent home are intelligent decision making, location-aware service, and compatibility for all users of different ages and physical conditions. Existing solutions address just one or two of these challenges, but smart home automation that is robust, intelligent, location-aware, and predictive is needed to satisfy the user's demand. This paper presents a location-aware intelligent RSSI-based home automation system (IRHA) that uses Wi-Fi signals to detect the user's location and control the appliances automatically. The fingerprinting method is used to map the Wi-Fi signals for different rooms, and the machine learning method, such as Decision Tree, is used to classify the signals for different rooms. The machine learning models are then implemented in the ESP32 microcontroller board to classify the rooms based on the real-time Wi-Fi signal, and then the result is sent to the main control board through the ESP32 MAC communication protocol to control the appliances automatically. The proposed method has achieved 97% accuracy in classifying the users' location.
△ Less
Submitted 16 January, 2022;
originally announced January 2022.
-
Smart IoT-Biofloc water management system using Decision regression tree
Authors:
Samsil Arefin Mozumder,
A S M Sharifuzzaman Sagar
Abstract:
The conventional fishing industry has several difficulties: water contamination, temperature instability, nutrition, area, expense, etc. In fish farming, Biofloc technology turns traditional farming into a sophisticated infrastructure that enables the utilization of leftover food by turning it into bacterial biomass. The purpose of our study is to propose an intelligent IoT Biofloc system that imp…
▽ More
The conventional fishing industry has several difficulties: water contamination, temperature instability, nutrition, area, expense, etc. In fish farming, Biofloc technology turns traditional farming into a sophisticated infrastructure that enables the utilization of leftover food by turning it into bacterial biomass. The purpose of our study is to propose an intelligent IoT Biofloc system that improves efficiency and production. This article introduced a system that gathers data from sensors, store data in the cloud, analyses it using a machine learning model such as a Decision regression tree model to predict the water condition, and provides real-time monitoring through an android app. The proposed system has achieved a satisfactory accuracy of 79% during the experiment.
△ Less
Submitted 5 December, 2021;
originally announced December 2021.
-
A Machine Learning Approach for Early Detection of Fish Diseases by Analyzing Water Quality
Authors:
Al-Akhir Nayan,
Ahamad Nokib Mozumder,
Joyeta Saha,
Khan Raqib Mahmud,
Abul Kalam Al Azad,
Muhammad Golam Kibria
Abstract:
Early detection of fish diseases and identifying the underlying causes are crucial for farmers to take necessary steps to mitigate the potential outbreak and thus to avert financial losses with apparent negative implications to the national economy. Typically, fish diseases are caused by viruses and bacteria; according to biochemical studies, the presence of certain bacteria and viruses may affect…
▽ More
Early detection of fish diseases and identifying the underlying causes are crucial for farmers to take necessary steps to mitigate the potential outbreak and thus to avert financial losses with apparent negative implications to the national economy. Typically, fish diseases are caused by viruses and bacteria; according to biochemical studies, the presence of certain bacteria and viruses may affect the level of pH, DO, BOD, COD, TSS, TDS, EC, PO43-, NO3-N, and NH3-N in water, resulting in the death of fishes. Besides, natural processes, e.g., photosynthesis, respiration, and decomposition, also contribute to the alteration of water quality that adversely affects fish health. Being motivated by the recent successes of machine learning techniques, a state-of-art machine learning algorithm has been adopted in this paper to detect and predict the degradation of water quality timely and accurately. Thus, it helps to take preemptive steps against potential fish diseases. The experimental results show high accuracy in detecting fish diseases specific to water quality based on the algorithm with real datasets.
△ Less
Submitted 11 November, 2021; v1 submitted 15 February, 2021;
originally announced February 2021.
-
Real Time Multi-Class Object Detection and Recognition Using Vision Augmentation Algorithm
Authors:
Al-Akhir Nayan,
Joyeta Saha,
Ahamad Nokib Mozumder,
Khan Raqib Mahmud,
Abul Kalam Al Azad
Abstract:
The aim of this research is to detect small objects with low resolution and noise. The existing real time object detection algorithm is based on the deep neural network of convolution need to perform multilevel convolution and pooling operations on the entire image to extract a deep semantic characteristic of the image. The detection models perform better for large objects. The features of existin…
▽ More
The aim of this research is to detect small objects with low resolution and noise. The existing real time object detection algorithm is based on the deep neural network of convolution need to perform multilevel convolution and pooling operations on the entire image to extract a deep semantic characteristic of the image. The detection models perform better for large objects. The features of existing models do not fully represent the essential features of small objects after repeated convolution operations. We have introduced a novel real time detection algorithm which employs upsampling and skip connection to extract multiscale features at different convolution levels in a learning task resulting a remarkable performance in detecting small objects. The detection precision of the model is shown to be higher and faster than that of the state-of-the-art models.
△ Less
Submitted 11 November, 2020; v1 submitted 16 March, 2020;
originally announced March 2020.
-
Energy deposition on nuclear emulsion by slow recoil ions for directional dark matter searches
Authors:
Akira Hitachi,
A. Mozumder,
Kiseki D. Nakamura
Abstract:
The electronic energy deposited on nuclear emulsions due to C ions of 5 -- 200~keV and Kr ions of 5 -- 600~keV are evaluated and compared with those due to fast ions for design and construction of fine grain nuclear emulsion for directional dark matter searches. Nuclear quenching factors and the electronic LET (linear energy transfer), the specific electronic energy deposited along the ion track,…
▽ More
The electronic energy deposited on nuclear emulsions due to C ions of 5 -- 200~keV and Kr ions of 5 -- 600~keV are evaluated and compared with those due to fast ions for design and construction of fine grain nuclear emulsion for directional dark matter searches. Nuclear quenching factors and the electronic LET (linear energy transfer), the specific electronic energy deposited along the ion track, are evaluated. The so-called core and penumbra of heavy-ion track structure is modified for understanding the track due to recoil ions produced by dark matter candidate, WIMPs, striking nucleus in the AgBr crystal of nuclear emulsion. The very heavy recoil ions, 100 -- 180~keV Pb ions, produced in $α$-decay are also studied. In addition, the track structures due to proton ions of 25 -- 80~keV are evaluated to consider the influence of background neutrons in underground laboratories.
△ Less
Submitted 16 May, 2019;
originally announced May 2019.
-
Properties for liquid argon scintillation for dark matter searches
Authors:
Akira Hitachi,
A. Mozumder
Abstract:
The scintillation yield for recoil Ar ions of 5 to 250 keV energy in liquid argon have been evaluated for direct dark matter searches. Lindhard theory is taken for estimating nuclear quenching. A theoretical model based on a biexcitonic diffusion-reaction mechanism is performed for electronic (scintillation) quenching. The electronic LET (linear energy transfer) is evaluated and used to obtain the…
▽ More
The scintillation yield for recoil Ar ions of 5 to 250 keV energy in liquid argon have been evaluated for direct dark matter searches. Lindhard theory is taken for estimating nuclear quenching. A theoretical model based on a biexcitonic diffusion-reaction mechanism is performed for electronic (scintillation) quenching. The electronic LET (linear energy transfer) is evaluated and used to obtain the initial track structure due to recoil Ar ions. The results are compared with experimental values reported for nuclear recoils from neutrons. The behavior of scintillation and ionization on the field are discussed. with notes on radiation physics and chemistry for dark matter searches.
△ Less
Submitted 14 March, 2019;
originally announced March 2019.
