Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896267
Tushar Bhaskarwar, Rajan Chile, Sumit Aole, I. Elamvazuthi
This paper focuses on one of the important aspects in the industry 4.0 revolution which is remote monitoring of industrial process through the Industrial Internet of Things (IIoT). In this paper, the cascade control application of coupled level process plant is considered for remote monitoring with IIoT concept. The level of the tank was regulated using PLC-OPC-MATLAB configuration. In order to control the level of a Coupled tank process, two controllers for primary and secondary loops are used to achieve the desired set point. Proportional Integral Derivative-Proportional (PID-P) control strategy is implemented in primary and secondary loop respectively for tracking the set point with minimum settling time. The ThingSpeak platform, a cloud-based server is used for remote monitoring purpose. Finally, the results of the level response and its online observation shown in the result section.
{"title":"Remote Monitoring of Coupled Tank Accompanied by PLC-OPC-MATLAB Architecture","authors":"Tushar Bhaskarwar, Rajan Chile, Sumit Aole, I. Elamvazuthi","doi":"10.1109/SCORED.2019.8896267","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896267","url":null,"abstract":"This paper focuses on one of the important aspects in the industry 4.0 revolution which is remote monitoring of industrial process through the Industrial Internet of Things (IIoT). In this paper, the cascade control application of coupled level process plant is considered for remote monitoring with IIoT concept. The level of the tank was regulated using PLC-OPC-MATLAB configuration. In order to control the level of a Coupled tank process, two controllers for primary and secondary loops are used to achieve the desired set point. Proportional Integral Derivative-Proportional (PID-P) control strategy is implemented in primary and secondary loop respectively for tracking the set point with minimum settling time. The ThingSpeak platform, a cloud-based server is used for remote monitoring purpose. Finally, the results of the level response and its online observation shown in the result section.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130972334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896299
Indra Gandi Subramani, Ismail Arif bin Ahmad Fauzi, V. Perumal
Interdigitated electrode (IDE) as transducer in biosensor usually fabricated through tedious, time consuming photolithography technique by using rigid substrate such as silicon and glass. The material type and geometry of the IDE is one of the important factor for an enhanced sensitivity of the device. Flexible substrate, polyimide have been employed as alternatives to fabricate more sensitive, cheaper and robust small devices. In this paper, silver (Ag) and copper (Cu) IDEs was fabricated on polyimide substrate via one step radio frequency sputtering technique by transferring the pattern of IDE from Perspex hard mask onto polyimide substrate. Design of the IDE was drawn using AutoCAD 2018 software with appropriate geometry and Perspex hard mask was cut through laser cutting technique. IDE with two different surface area, circle shaped and square shaped was fabricated for each material type (silver and copper). High power microscope (HPM) was used to verify the dimensions of the fabricated IDE and to investigate optimum sputtering time. Voltage analysis demonstrate that Ag based circular shaped IDE with 0.8 mm gap dimension resulted 1.15 × 10–2 V, greater electrical behavior compares to copper based IDE.
{"title":"Expeditious Fabrication & Characterization of Metal Interdigitated Transducer on Polyimide film for Biosensing Application","authors":"Indra Gandi Subramani, Ismail Arif bin Ahmad Fauzi, V. Perumal","doi":"10.1109/SCORED.2019.8896299","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896299","url":null,"abstract":"Interdigitated electrode (IDE) as transducer in biosensor usually fabricated through tedious, time consuming photolithography technique by using rigid substrate such as silicon and glass. The material type and geometry of the IDE is one of the important factor for an enhanced sensitivity of the device. Flexible substrate, polyimide have been employed as alternatives to fabricate more sensitive, cheaper and robust small devices. In this paper, silver (Ag) and copper (Cu) IDEs was fabricated on polyimide substrate via one step radio frequency sputtering technique by transferring the pattern of IDE from Perspex hard mask onto polyimide substrate. Design of the IDE was drawn using AutoCAD 2018 software with appropriate geometry and Perspex hard mask was cut through laser cutting technique. IDE with two different surface area, circle shaped and square shaped was fabricated for each material type (silver and copper). High power microscope (HPM) was used to verify the dimensions of the fabricated IDE and to investigate optimum sputtering time. Voltage analysis demonstrate that Ag based circular shaped IDE with 0.8 mm gap dimension resulted 1.15 × 10–2 V, greater electrical behavior compares to copper based IDE.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128376903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896259
M. Atef, M. Abdullah, T. Khatib, M. Romlie
An improvement for a new hybrid power system model is presented. The improvement considers the most accurate model that gives the exact energy output from the Solar photovoltaic (SPV) system to give more accurate result about the perfect size of the PV in the hybrid photo-voltaic and gas turbine generator (GTG) (H-PVGTG) system. This result will affect the size of both the battery bank and the GTG units. The values must justify the technical requirements of the system reliability. This value is recommended to be 0.01 in Malaysia, and it is known as the Loss of Load Probability (LLP). The main goal of the research is to get the most accurate system size with the lowest Annualized Total Life-Cycle Cost (ATLCC). The mathematical model (Math-M) that has been used in the optimization algorithm saved more than 38 % from the operating cost of the power system that is used to supply the power to Universiti Teknologi PETRONAS (UTP). However, it has an error in the power output compared with the actual site power output. Due to the high operating cost of GTG system compared even with the grid supply in Malaysia, Tenaga Nasional Berhad (TNB). This paper proposed an Artificial Intelligent (AI) model to overcome the increase in the operating cost with lower power output error than the Math-M. The main challenge of the mathematical model was the low accuracy as it has +6.09% error than the actual power output of the SPV system and that is why a black box model (BB-M) has been trained to overcome this problem. A comparison between the BB-M, Math-M, GTG system, and TNB has been presented in this paper. The result concluded that BB-M has more accuracy than Math-M if compared with the actual power output of SPV system.
提出了一种新的混合动力系统模型的改进方法。该改进考虑了最精确的模型，该模型给出了太阳能光伏(SPV)系统的确切能量输出，从而给出了更准确的关于光伏和燃气涡轮发电机(H-PVGTG)混合系统中PV的完美尺寸的结果。这个结果将影响电池组和GTG单元的尺寸。这些值必须符合系统可靠性的技术要求。这个值在马来西亚被推荐为0.01，它被称为负载损失概率(LLP)。本研究的主要目标是以最低的年化总生命周期成本(ATLCC)获得最精确的系统尺寸。优化算法中使用的数学模型(Math-M)从用于向Universiti teknologii PETRONAS (UTP)供电的电力系统的运行成本中节省了38%以上。但输出功率与现场实际输出功率存在误差。由于GTG系统的运行成本即使与马来西亚的电网供应相比也很高，Tenaga Nasional Berhad (TNB)。本文提出了一种人工智能(AI)模型，以克服运行成本增加的问题，并具有比Math-M更小的功率输出误差。数学模型的主要挑战是精度低，因为它比SPV系统的实际功率输出有+6.09%的误差，这就是为什么训练黑箱模型(BB-M)来克服这个问题。本文对BB-M、Math-M、GTG系统和TNB系统进行了比较。结果表明，与SPV系统的实际输出功率相比，BB-M比Math-M精度更高。
{"title":"Utilization of Artificial Neural Networks to Improve the Accuracy of a Hybrid Power System Model","authors":"M. Atef, M. Abdullah, T. Khatib, M. Romlie","doi":"10.1109/SCORED.2019.8896259","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896259","url":null,"abstract":"An improvement for a new hybrid power system model is presented. The improvement considers the most accurate model that gives the exact energy output from the Solar photovoltaic (SPV) system to give more accurate result about the perfect size of the PV in the hybrid photo-voltaic and gas turbine generator (GTG) (H-PVGTG) system. This result will affect the size of both the battery bank and the GTG units. The values must justify the technical requirements of the system reliability. This value is recommended to be 0.01 in Malaysia, and it is known as the Loss of Load Probability (LLP). The main goal of the research is to get the most accurate system size with the lowest Annualized Total Life-Cycle Cost (ATLCC). The mathematical model (Math-M) that has been used in the optimization algorithm saved more than 38 % from the operating cost of the power system that is used to supply the power to Universiti Teknologi PETRONAS (UTP). However, it has an error in the power output compared with the actual site power output. Due to the high operating cost of GTG system compared even with the grid supply in Malaysia, Tenaga Nasional Berhad (TNB). This paper proposed an Artificial Intelligent (AI) model to overcome the increase in the operating cost with lower power output error than the Math-M. The main challenge of the mathematical model was the low accuracy as it has +6.09% error than the actual power output of the SPV system and that is why a black box model (BB-M) has been trained to overcome this problem. A comparison between the BB-M, Math-M, GTG system, and TNB has been presented in this paper. The result concluded that BB-M has more accuracy than Math-M if compared with the actual power output of SPV system.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128743687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896327
Soumya Mudgal, V. Mahajan
This paper presents wind energy as a possible alternative to the present conventional generation systems. Multi turbine wind systems situated in various farms are analyzed by incorporating the number of turbines, their repair and failure rates and, their forced outages. The impact of wind upon reliability indices and its effect on power flow are the main objectives of this paper. Discrete Markov Chains and Monte-Carlo simulations support the model to calculate Capacity Outages and Probability. Loss of Load Expectations (LOLE), Expected Energy Not Supplied (EENS) and Expected Demand Not Supplied (EDNS) indices are calculated for reliability assessment and comparison between the original and modified system. Load Flow analysis for daily estimation of active power losses reflects the reduced line losses when wind energy is incorporated in the system. The IEEE RTS-79 system of 24 bus is examined here, for evaluation of reliability indices.
{"title":"Reliability and Active Power Loss Assessment Of Power System Network With Wind Energy","authors":"Soumya Mudgal, V. Mahajan","doi":"10.1109/SCORED.2019.8896327","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896327","url":null,"abstract":"This paper presents wind energy as a possible alternative to the present conventional generation systems. Multi turbine wind systems situated in various farms are analyzed by incorporating the number of turbines, their repair and failure rates and, their forced outages. The impact of wind upon reliability indices and its effect on power flow are the main objectives of this paper. Discrete Markov Chains and Monte-Carlo simulations support the model to calculate Capacity Outages and Probability. Loss of Load Expectations (LOLE), Expected Energy Not Supplied (EENS) and Expected Demand Not Supplied (EDNS) indices are calculated for reliability assessment and comparison between the original and modified system. Load Flow analysis for daily estimation of active power losses reflects the reduced line losses when wind energy is incorporated in the system. The IEEE RTS-79 system of 24 bus is examined here, for evaluation of reliability indices.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131216600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896339
Mohamad Nasrun bin Mohd Nasir, A. Sabo, N. Wahab
The Phasor Measurement Unit (PMU) is the heart of smart grid system as it provides the data such as voltage and phase angle measurements of all buses of the system and thereby maintaining the system observability. In this context, this paper summarizes the various research based on PMU for complete observability and monitoring of integrated power system. The survey indicates that most of the recent researches are focusing on optimal PMU placement (OPP) rather than design and modeling of PMU considering various cases. Moreover, the state estimation using synchrophasor technology are also presented as addition objective to obtain the optimal number of PMU that need to be installed in the system for power system analysis and economic benefits of the system. The trend of research based on synchrophasor technology are evolving for real-time power system monitoring application where it also covers for dynamic power system assessment.
{"title":"A Review on Synchrophasor Technology for Power System Monitoring","authors":"Mohamad Nasrun bin Mohd Nasir, A. Sabo, N. Wahab","doi":"10.1109/SCORED.2019.8896339","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896339","url":null,"abstract":"The Phasor Measurement Unit (PMU) is the heart of smart grid system as it provides the data such as voltage and phase angle measurements of all buses of the system and thereby maintaining the system observability. In this context, this paper summarizes the various research based on PMU for complete observability and monitoring of integrated power system. The survey indicates that most of the recent researches are focusing on optimal PMU placement (OPP) rather than design and modeling of PMU considering various cases. Moreover, the state estimation using synchrophasor technology are also presented as addition objective to obtain the optimal number of PMU that need to be installed in the system for power system analysis and economic benefits of the system. The trend of research based on synchrophasor technology are evolving for real-time power system monitoring application where it also covers for dynamic power system assessment.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123756860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896347
Upasana Lakhina, I. Elamvazuthi, N. Badruddin, F. Meriaudeau, G. Ramasamy, A. Jangra
Excessive growth in electricity consumption has been experienced over the past few years due to an increase in population around the world.This tends to increase the use of renewable energy and randomness of the load. So.it is important to improve the traditional methodologies and techniques applied on microgrid to make it more intelligent. In this paper, multi agent system is employed over autonomous microgrid framework to endorse its intelligence. The Multi-Agent system is simulated in Java Agent Development Environment (JADE) environment and matlab toolbox Simulink is used for the implementation of the microgrid model. Further, MACSimJX is used to communicate between the micro grid and agent system. This paper shows the communication between the agents and the microgrid model and how they process the data through MACSimJX to make intelligent decisions.
在过去的几年里，由于世界各地人口的增加，电力消费出现了过度增长。这往往会增加可再生能源的使用和负荷的随机性。所以。改进传统的微电网方法和技术，提高微电网的智能化水平是十分重要的。本文将多智能体系统应用于自主微网框架，以验证其智能。在Java Agent Development Environment (JADE)环境下对多Agent系统进行仿真，并利用matlab工具箱Simulink实现微电网模型。在此基础上，利用MACSimJX实现微网与代理系统之间的通信。本文展示了智能体与微电网模型之间的通信，以及它们如何通过MACSimJX处理数据来进行智能决策。
{"title":"Multi-Agent Based Energy Management in Microgrids Using MACSimJX","authors":"Upasana Lakhina, I. Elamvazuthi, N. Badruddin, F. Meriaudeau, G. Ramasamy, A. Jangra","doi":"10.1109/SCORED.2019.8896347","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896347","url":null,"abstract":"Excessive growth in electricity consumption has been experienced over the past few years due to an increase in population around the world.This tends to increase the use of renewable energy and randomness of the load. So.it is important to improve the traditional methodologies and techniques applied on microgrid to make it more intelligent. In this paper, multi agent system is employed over autonomous microgrid framework to endorse its intelligence. The Multi-Agent system is simulated in Java Agent Development Environment (JADE) environment and matlab toolbox Simulink is used for the implementation of the microgrid model. Further, MACSimJX is used to communicate between the micro grid and agent system. This paper shows the communication between the agents and the microgrid model and how they process the data through MACSimJX to make intelligent decisions.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"91 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126700398","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896222
Y. Hafeez, S. Ali, Syed Faraz, M. Moinuddin, Syed Hasan Adil
The EEG-neurofeedback modality has direct implication on alpha asymmetry. The efficacy of EEG-neurofeedback may be affected by the stimulus contents. This research investigated the effectiveness of 2D and 3D game stimulus content (GSC) on stress mitigation during neurofeedback training (NFT). The effectiveness is compared between stimulus contents by measuring the mean prefrontal alpha asymmetry using quantitative Electroencephalogram (qEEG) analysis. For this, ten healthy participants among university students were recruited and performed twenty-minutes neurofeedback training (NFT) on Fp1-Fp2 within a period of sixty days to record forty sessions of data. The statistical analysis of the data after the neurofeedback training showed an effect of game contents on alpha asymmetry. The graphical analysis of alpha power showed that the 3D game content was more effective than the 2D game content. The outcome of 3D game stimulus content showed effect on the prefrontal alpha asymmetry and improve the treatment efficacy of neurofeedback for stress mitigation.
{"title":"Effect Of Neurofeedback 2D and 3D Stimulus Content On Stress Mitigation","authors":"Y. Hafeez, S. Ali, Syed Faraz, M. Moinuddin, Syed Hasan Adil","doi":"10.1109/SCORED.2019.8896222","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896222","url":null,"abstract":"The EEG-neurofeedback modality has direct implication on alpha asymmetry. The efficacy of EEG-neurofeedback may be affected by the stimulus contents. This research investigated the effectiveness of 2D and 3D game stimulus content (GSC) on stress mitigation during neurofeedback training (NFT). The effectiveness is compared between stimulus contents by measuring the mean prefrontal alpha asymmetry using quantitative Electroencephalogram (qEEG) analysis. For this, ten healthy participants among university students were recruited and performed twenty-minutes neurofeedback training (NFT) on Fp1-Fp2 within a period of sixty days to record forty sessions of data. The statistical analysis of the data after the neurofeedback training showed an effect of game contents on alpha asymmetry. The graphical analysis of alpha power showed that the 3D game content was more effective than the 2D game content. The outcome of 3D game stimulus content showed effect on the prefrontal alpha asymmetry and improve the treatment efficacy of neurofeedback for stress mitigation.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114778473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896248
Zia Khan, N. Yahya, K. Alsaih, F. Mériaudeau
The number of prostate cancer cases is steadily increasing especially with rising number of ageing population. It is reported that 5-year relative survival rate for man with stage 1 prostate cancer is almost 99% hence, early detection will significantly improve treatment planning and increase survival rate. Magnetic resonance imaging (MRI) technique is a common imaging modality for diagnosis of prostate cancer. MRI provide good visualization of soft tissue and enable better lesion detection and staging of prostate cancer. The main challenge of prostate whole gland segmentation is due to blurry boundary of central gland (CG) and peripheral zone (PZ) which lead to differential diagnosis. Since there is substantial difference in occurance and characteristic of cancer in both zones. So to enhance the diagnosis of prostate gland, we implemented DeeplabV3+ semantic segmentation approach to segment the prostate into zones. DeepLabV3+ achieved significant results in segmentation of prostate MRI by applying several parallel atrous convolution with different rates. The CNN-based semantic segmentation approach is trained and tested on NCI-ISBI 1.5T and 3T MRI dataset consist of 40 patients. Performance evaluation based on Dice similarity coefficient (DSC) of the Deeplab-based segmentation is compared with two other CNN-based semantic segmentation techniques: FCN and PSNet. Results shows that prostate segmentation using DeepLabV3+ can perform better than FCN and PSNet with average DSC of 70.3% in PZ and 88% in CG zone. This indicates the significant contribution made by the atrous convolution layer, in producing better prostate segmentation result.
{"title":"Zonal Segmentation of Prostate T2W-MRI using Atrous Convolutional Neural Network","authors":"Zia Khan, N. Yahya, K. Alsaih, F. Mériaudeau","doi":"10.1109/SCORED.2019.8896248","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896248","url":null,"abstract":"The number of prostate cancer cases is steadily increasing especially with rising number of ageing population. It is reported that 5-year relative survival rate for man with stage 1 prostate cancer is almost 99% hence, early detection will significantly improve treatment planning and increase survival rate. Magnetic resonance imaging (MRI) technique is a common imaging modality for diagnosis of prostate cancer. MRI provide good visualization of soft tissue and enable better lesion detection and staging of prostate cancer. The main challenge of prostate whole gland segmentation is due to blurry boundary of central gland (CG) and peripheral zone (PZ) which lead to differential diagnosis. Since there is substantial difference in occurance and characteristic of cancer in both zones. So to enhance the diagnosis of prostate gland, we implemented DeeplabV3+ semantic segmentation approach to segment the prostate into zones. DeepLabV3+ achieved significant results in segmentation of prostate MRI by applying several parallel atrous convolution with different rates. The CNN-based semantic segmentation approach is trained and tested on NCI-ISBI 1.5T and 3T MRI dataset consist of 40 patients. Performance evaluation based on Dice similarity coefficient (DSC) of the Deeplab-based segmentation is compared with two other CNN-based semantic segmentation techniques: FCN and PSNet. Results shows that prostate segmentation using DeepLabV3+ can perform better than FCN and PSNet with average DSC of 70.3% in PZ and 88% in CG zone. This indicates the significant contribution made by the atrous convolution layer, in producing better prostate segmentation result.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125009448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896277
D. O. Alebiosu, Fermi Pasha Muhammad
Medical image classification is an important step in the effective and accurate retrieval of medical images from large digital database where they are stored. This paper examines the effectiveness of using domain transferred neural networks (DCNNs) for classification of medical X-ray images. We employed two different convolutional neural network (CNN) architectures. VGGNet-16 and AlexNet pre-trained on ImageNet, a non- medical image database consisting of over 1.2 million scenery images were used for the classification task. The pre-trained networks served both as feature extractors and as fine-tuned networks. The extracted feature vector was used to train a linear support vector machine (SVM) to generate a model for the classification task. The fine-tuning process was done by replacing and retraining the last fully connected layers through backward propagation. Our method was evaluated on ImageCLEF2007 medical database. The database consist of 11,000 medical X-ray images (training dataset) and 1,000 images (testing dataset) classified into 116 categories. We compared the performance of the two networks both as feature generators and as fine-tuned networks on our dataset. The overall classification accuracy across all the 116 image classes shows that VGGNet-16 + SVM produced 79.6% and 85.77% as fine-tuned network. AlexNet + SVM produced a total classification accuracy of 84.27% and as a fine-tuned network produced a total of 86.47% which is the highest among the four techniques across all the 116 image classes. This study shows that the employment of a shallower pre-trained neural network such as AlexNet learn features that are more generalizable compared to deeper networkers such as VGGNet-16 and has a greater capability of increasing classification accuracy of medical image database. Though the pre-trained AlexNet outperformed VGGNet-16 in both ways, it can be noted that some image classes from the same sub-body region are difficult to classify accurately. This is as a result of inter-class similarity that exists among the images.
{"title":"Medical Image Classification: A Comparison of Deep Pre-trained Neural Networks","authors":"D. O. Alebiosu, Fermi Pasha Muhammad","doi":"10.1109/SCORED.2019.8896277","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896277","url":null,"abstract":"Medical image classification is an important step in the effective and accurate retrieval of medical images from large digital database where they are stored. This paper examines the effectiveness of using domain transferred neural networks (DCNNs) for classification of medical X-ray images. We employed two different convolutional neural network (CNN) architectures. VGGNet-16 and AlexNet pre-trained on ImageNet, a non- medical image database consisting of over 1.2 million scenery images were used for the classification task. The pre-trained networks served both as feature extractors and as fine-tuned networks. The extracted feature vector was used to train a linear support vector machine (SVM) to generate a model for the classification task. The fine-tuning process was done by replacing and retraining the last fully connected layers through backward propagation. Our method was evaluated on ImageCLEF2007 medical database. The database consist of 11,000 medical X-ray images (training dataset) and 1,000 images (testing dataset) classified into 116 categories. We compared the performance of the two networks both as feature generators and as fine-tuned networks on our dataset. The overall classification accuracy across all the 116 image classes shows that VGGNet-16 + SVM produced 79.6% and 85.77% as fine-tuned network. AlexNet + SVM produced a total classification accuracy of 84.27% and as a fine-tuned network produced a total of 86.47% which is the highest among the four techniques across all the 116 image classes. This study shows that the employment of a shallower pre-trained neural network such as AlexNet learn features that are more generalizable compared to deeper networkers such as VGGNet-16 and has a greater capability of increasing classification accuracy of medical image database. Though the pre-trained AlexNet outperformed VGGNet-16 in both ways, it can be noted that some image classes from the same sub-body region are difficult to classify accurately. This is as a result of inter-class similarity that exists among the images.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"43 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120921355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/SCORED.2019.8896323
Devi R Krishnan, C. Maddipati, Gayathri P Menakath, A. Radhakrishnan, Yarrangangu Himavarshini, A. A, K. Mukundan, Rahul Krishnan Pathinarupothi, Bithin Alangot, Sirisha Mahankali
Diabetes Mellitus (DM) is one of the major global health challenges of the 21st century. It is a chronic disease leading to multiple complications bearing a lot of social, physical and financial impact on individuals and society. Gestational diabetes mellitus (GDM) is a type of DM that is developed in a few pregnant women although it usually reverts back to normalcy after delivery. However, it is well established that the risk in developing DM at a later stage in their lives increases with GDM. Very few works done in this area explore the possibility of using prognostic Machine Learning algorithms to predict occurrence of DM after GDM. In this paper, we conduct a methodical review of current practices, and then analyze GDM data from our University hospital to identify predisposing factors that could be used as inputs to different ML techniques.
{"title":"Evaluation of predisposing factors of Diabetes Mellitus post Gestational Diabetes Mellitus using Machine Learning Techniques","authors":"Devi R Krishnan, C. Maddipati, Gayathri P Menakath, A. Radhakrishnan, Yarrangangu Himavarshini, A. A, K. Mukundan, Rahul Krishnan Pathinarupothi, Bithin Alangot, Sirisha Mahankali","doi":"10.1109/SCORED.2019.8896323","DOIUrl":"https://doi.org/10.1109/SCORED.2019.8896323","url":null,"abstract":"Diabetes Mellitus (DM) is one of the major global health challenges of the 21st century. It is a chronic disease leading to multiple complications bearing a lot of social, physical and financial impact on individuals and society. Gestational diabetes mellitus (GDM) is a type of DM that is developed in a few pregnant women although it usually reverts back to normalcy after delivery. However, it is well established that the risk in developing DM at a later stage in their lives increases with GDM. Very few works done in this area explore the possibility of using prognostic Machine Learning algorithms to predict occurrence of DM after GDM. In this paper, we conduct a methodical review of current practices, and then analyze GDM data from our University hospital to identify predisposing factors that could be used as inputs to different ML techniques.","PeriodicalId":231004,"journal":{"name":"2019 IEEE Student Conference on Research and Development (SCOReD)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134618951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：