{"title":"Diabetic Retinopathy Detection: Improving Accuracy Using Multiple Transfer Learning Features from Pre-trained Deep Learning Networks","authors":"Kelvin Ka Yung Tiong, W. Wong, F. Juwono, I. Chew","doi":"10.1109/GECOST55694.2022.10010468","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010468","url":null,"abstract":"Diabetic Retinopathy (DR) is a type of complications caused by diabetes. Patients with DR may experience worsening vision, blindness, and eye pain. To effectively address this disorder, DR must be identified and classified according to its severity. Therefore, automated diagnosis of fundus lesions is of great interest for DR early detection. The development of deep learning technology has provided a strong foundation for effective implementation of the automated detection system. In particular, transfer learning techniques have greatly benefited the research community to reduce computation and reuse trained features. In this paper, the outputs from the ”average pooling” and ”fully connected” layers are used as the features to the Support Vector Machine (SVM) classifier with Error Correction Output Code (ECOC). The proposed method outperforms the fine-tuned pre-trained networks in predicting the severity classes with an accuracy of 80.1%. This means that multiple features extracted from the pre-trained networks contribute to a better recognition process.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"335 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134511388","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}
{"title":"Development of Iot Monitoring System For Aquaculture Application","authors":"H. F. Hawari, Mohammad Arief Hazwan","doi":"10.1109/GECOST55694.2022.10010661","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010661","url":null,"abstract":"Malaysia is blessed with abundant water resources such as sea and inland freshwaters which has enormous potential in providing with more healthful and nutritious food through aquaculture. Aquaculture has also been identified as a priority sector in the development of Malaysia’s economy in contribution to nation ‘food security”. With forecasted population expansion, demand is anticipated to continue to increase. This research is about developing an Internet of Things (IoT) system to monitor the quality of freshwater for aquatic species in Malaysia. Dramatic changes in local weather and climate have also significantly affect in aquaculture water quality resulted in many fish dying and fish farmers suffering production losses. The traditional method in checking water quality is time-consuming and expensive since it requires manual sample to be send to lab. However, with the advent of the IoT system, user can real time and observe the current state of water quality anywhere by using its own website or application. The IoT system will consists of several parts, namely a database to store data, node red application apps, sensors, components, and circuits. This IoT system is capable to detect the rate of water turbidity, the quantity of air in the water and the quantity of pH. The system has been tested thoroughly in terms of several lake locations in UTP and 2 fish farms located in Pahang. The sensors were calibrated accurately with the amount of standard deviation below 2. This developed system will enable fish farmers to monitor their aquaculture products real time and effectively.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115191976","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}
F. Sia, Lim Hong Yao, Tuen Yong Hao, Peter Chin, Lim Min Xuan
{"title":"Predicting Study or Work From Home Satisfaction using Data Mining Technique: A Case Study in Malaysia","authors":"F. Sia, Lim Hong Yao, Tuen Yong Hao, Peter Chin, Lim Min Xuan","doi":"10.1109/GECOST55694.2022.10010647","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010647","url":null,"abstract":"The COVID-19 pandemic has urged the government of Malaysia to implement Movement Control Order (MCO) which forces working people to work from home while students to study from home. People’s satisfaction on work from home is crucial in determining their work productivity and efficiency whereas student’s satisfaction on study from home is important for their learning effectiveness. There is no work has been done yet for exploring data mining techniques to build a model for predicting work or study from home satisfaction using Malaysia as a case study. This paper aimed to identify the best data mining model for predicting the work or study from home satisfaction. The prediction model is learned by analyzing the demographic, the personality traits, and the work from home experience collected from a group of Malaysia people. This study attempts to investigate four data mining techniques that are the decision tree, linear kernel support vector machine, polynomial support vector machine, and radial basis support vector machine. Experiment results show that the radial basis support vector machine outperformed other techniques in predicting the work or study from home satisfaction of Malaysia’s community.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132271715","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}
{"title":"A Preliminary Performance Analysis of A Bio-Inspired Antenna on a Glossy Paper Substrate","authors":"N. Affendi, S. Saudin","doi":"10.1109/GECOST55694.2022.10010477","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010477","url":null,"abstract":"A bio-inspired antenna is designed on a flexible substrate in the context of this project. The rectenna, which combines an antenna and a rectifier, is a tool for gathering wireless energy in the air. The analysis is the focus of this report. Here the Bio-inspired antenna on glossy paper substrate for 2-5 GHZ application in Malaysia. This antenna is used a glossy paper substrate with a er value of 3.6. The antenna has been designed and optimized for its gain and efficiency with an improved performance than other antennas. The antenna has been designed within 1.6- 2.00mm of thickness, 26.6mm of width, and 27.6mm of length. This design is a frontline component for enhancing 5G energy harvesting is an antenna as well.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116654392","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}
{"title":"Investigation on Grid-Scale BESS Capacity for Frequency Support under various System Inertia","authors":"Hoe Yin Chan, R. Tan, Ousama Ajami","doi":"10.1109/GECOST55694.2022.10010379","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010379","url":null,"abstract":"Rapid growth in the renewable energy sector supplying to the grid including solar PV and wind has caused the system inertia to reduce. Consequently, instability problems and supply interruption will happen more frequently during power imbalance in a low inertia power system. Hence, the solution of implementing a grid-scale battery energy storage system (BESS) is proposed to provide frequency support to the grid. This paper presents the modelling of the 565 MW Eastern Region of the Peninsular Malaysia power system in MATLAB/Simulink, where the frequency response of the system under power disturbances is studied and the minimum BESS capacity required for frequency support is investigated. The effect of frequency support with and without BESS implementation is also investigated under various system inertia. Comparing the frequency nadir of the system with and without BESS, the results show that under the largest disturbance of 300 MW, which is 53% of the 565 MW system and the lowest inertia of 2s, the 300 kWh BESS supported the frequency by increasing the frequency nadir up to 0.024 Hz. The results show trends in which the BESS will contribute more to systems with lower inertia value and larger disturbance during frequency support.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"201 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133749321","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}
{"title":"Review on Machine Learning Methods for Remaining Useful Lifetime Prediction of Lithium-ion Batteries","authors":"Nicholas Su, F. Juwono, W. Wong, I. Chew","doi":"10.1109/GECOST55694.2022.10010569","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010569","url":null,"abstract":"Electric cars are considered as the most ecologically friendly and low-cost means of transportation in the future. As a result, battery technology advancement is of interest for many researchers. Lithium-ion batteries are mostly used for electric vehicles. However, if the Remaining Useful Lifetime (RUL) drops below capacity degradation, devastating device failure will occur. Hence, it is important to predict the RUL to prevent such problems. Data-driven methods are demonstrated to be superior to model-based methods for this reason. This paper provides a review on Machine Learning (ML), one of the data-driven methods, and summarizes various approaches that have been used in lithium-ion (Li-ion) batteries RUL prediction. In addition, comparison of model-based and ML methods are discussed. In particular, the comparison of three ML methods,i.e., Support Vector Machine (SVM), Neural Networks (NN), and Deep Learning(DL) are also presented. Simulation results show that SVM is able to provide higher RUL accuracy than LSTM and ANN.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134440685","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}
Alex Zi Neng Tan, Saaveethya Sivakumar, C. W. Chiong
{"title":"Joint Angle Analysis with IMU Sensors for Gait","authors":"Alex Zi Neng Tan, Saaveethya Sivakumar, C. W. Chiong","doi":"10.1109/GECOST55694.2022.10010542","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010542","url":null,"abstract":"Applications in Inertial Measurement Unit (IMU) has gradually increased in gait analysis. IMUs provide gait data acquisition in outdoor environments for long term gait monitoring in daily basis. The recent studies confirm that the data is reliable low error and high correlations when compared to the gold standard systems such as camera based laboratory setups. However, IMU sensors are affected with drift and undesired noise, thus pre-processing is needed. This review presents the recent advancements of gait analysis with respect to IMU sensors in joint motion computation. In recent years, IMU sensor based applications are combined with deep learning methods to enhance the accuracy and to reduce the number of sensors utilized.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132280683","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}
Zahraa S. Ismail, Eman F. Sawires, F. Amer, S. Abdellatif
{"title":"Study the C-V behavior of cesium-lead halides perovskite solar cells under various simulation parameters","authors":"Zahraa S. Ismail, Eman F. Sawires, F. Amer, S. Abdellatif","doi":"10.1109/GECOST55694.2022.10010474","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010474","url":null,"abstract":"Capacitance response of perovskite solar cells (PSCs) can be oppressed to deduce underlying physical mechanisms, both in the materials at external interfaces and in bulk materials. Accordingly, this paper investigates the Capacitance-Voltage (C-V) characteristic curves of cesium lead halides (CsPbX3:X = I, Br, or Cl) used as an active layer in PSCs. The SCAPS-1D simulator harnessed the actual device (CsPbX3:X = I, Br, or Cl) with material parameters from previous experimental work. Three main simulation parameters were investigated: the thickness of the active layer, the doping, and the defects impacts.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132144425","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}
{"title":"Design and Development of Deep Learning Approach for Dental Implant Planning","authors":"Rushikesh Bodhe, Saaveethya Sivakumar, Ayush Raghuwanshi","doi":"10.1109/GECOST55694.2022.10010527","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010527","url":null,"abstract":"Over the past two decades, the area of artificial intelligence (AI) has developed and grown remarkably. Recent advancements in machine learning, digital data collecting, and computer infrastructure have allowed AI applications to spread into fields traditionally regarded as the domain of human expertise. AI has enormous potential to alter the healthcare industry and enhance patient care when used in medicine and dentistry. AI is being researched in dentistry for a number of objectives, including the precise identification of healthy and unhealthy structures, disease detection, and treatment result prediction. In this research, we develop an approach for emphasizing its assistive and supplementary role for medical professionals in implant planning and maxillofacial surgery. We plan a dental implant operation by developing an AI-based system for detecting implant type and position, the exact location of the mandibular canal, a canal located on both sides of the lower jaw that contains the alveolar nerve, and the total number of missing teeth precisely enough to implement the correct methodology for operating over missing teeth.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114802424","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}
{"title":"Power Disaggregation Considering Seasonality of Unmonitored Building Facilities","authors":"Fuyuki Sato, N. Yamaguchi","doi":"10.1109/GECOST55694.2022.10010564","DOIUrl":"https://doi.org/10.1109/GECOST55694.2022.10010564","url":null,"abstract":"Estimating building energy efficiency and conservation effects remains challenging owing to a lack of detailed energy metering. To address unknown power consumption of facilities in commercial building energy management, disaggregation has been proposed to separate total power into multiple elements. However, assumptions are required when applying disaggregation to an actual building because not all facilities’ operating status can be monitored. Therefore, this study proposes that the power of the entire building can be accurately disaggregated by assuming that the power of unmonitored facilities is seasonal with daily periodicity. The proposed method is applied to a commercial building with 36 monitored outdoor air-conditioning units and unmonitored facilities, including indoor air-conditioning units, lighting, and other plug loads. Experimental results demonstrate that the proposed method can more precisely decompose the facility’s total power consumption than the conventional method. The proposed method is widely applicable to general commercial buildings and can contribute to energy-saving simulations and decision-making support for energy managers.","PeriodicalId":424601,"journal":{"name":"2022 International Conference on Green Energy, Computing and Sustainable Technology (GECOST)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115101013","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}