Alexander Vucenovic, Osama Ali-Ozkan, Clifford Ekwempe, Ozgur Eren
{"title":"Explainable AI in Decision Support Systems : A Case Study: Predicting Hospital Readmission Within 30 Days of Discharge","authors":"Alexander Vucenovic, Osama Ali-Ozkan, Clifford Ekwempe, Ozgur Eren","doi":"10.1109/CCECE47787.2020.9255721","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255721","url":null,"abstract":"Explainable models are a critical requirement for predictive analytics applications in the healthcare domain. In this work we develop a hypothetical clinical decision support system for the classification task of predicting hospital readmission within 30 days of discharge. We compare a baseline logistic regression model with an implementation of the coordinate descent algorithm known as lasso. We choose lasso because it inherently performs variable selection during optimization which leads to an explainable model. Using model evaluation data we achieve an area under the ROC curve score of 0.795 improving on the baseline score of 0.683 without inflating the feature space.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115042954","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}
U. Waqas, Nimra Akram, S. Kim, Donghun Lee, Ji-Yeol Jeon
{"title":"Vehicle Damage Classification and Fraudulent Image Detection Including Moiré Effect Using Deep Learning","authors":"U. Waqas, Nimra Akram, S. Kim, Donghun Lee, Ji-Yeol Jeon","doi":"10.1109/CCECE47787.2020.9255806","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255806","url":null,"abstract":"Image-based vehicle insurance processing and loan management has large scope for automation in automotive industry. In this paper we consider the problem of car damage classification, where categories include medium damage, huge damage and no damage. Based on deep learning techniques, MobileNet model is proposed with transfer learning for classification. Moreover, moving towards automation also comes with diverse hurdles; users can upload fake images like screenshots or taking pictures from computer screens, etc. To tackle this problem a hybrid approach is proposed to provide only authentic images to algorithm for damage classification as input. In this regard, moiré effect detection and metadata analysis is performed to detect fraudulent images. For damage classification 95% and for moiré effect detection 99% accuracy is achieved.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115550346","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":"Measurement and Analysis of Small Cell Splitting in a Real-world LTE-A HetNet","authors":"Haijun Gao, Japjot Singh Bawa, R. Paranjape","doi":"10.1109/CCECE47787.2020.9255826","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255826","url":null,"abstract":"Network densification is an important topic which has been studied during the past decades in the 4G heterogeneous networks (HetNets). Deployment of small cells and cell-splitting technique are aimed to increase network capacity, cell coverage, and total cell throughput in HetNets. However, most published literature is about theoretical analysis. In this paper, extensive measurements are conducted in a real-world LTE-A HetNet environment. The cell-splitting strategy is applied in a real-world LTE-A HetNet. Four directional antennas operate as one cell and two cells respectively in an indoor gymnasium in the University of Regina. Optimization techniques such as ABS (Almost Blank Subframe) are utilized to mitigate interference and increase UE (user equipment) SINR inside the gymnasium. Users' (both static users and moving users) average SINR and system cell throughput are used to evaluate the performance of the tests. Our results show that operating the small cells from one cell to three cells for the whole building, the SINR inside the gymnasium decreased from 29 dB to 5 dB, and cell throughput decreased from 140 Mbps to 88Mbps. Even though the throughput performance of cells inside the gymnasium is slightly lowered, the overall network capacity of the building is enhanced.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123427802","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":"Agent-Based Model of Cell Signaling in Cancer","authors":"Y. Derbal","doi":"10.1109/CCECE47787.2020.9255675","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255675","url":null,"abstract":"Cancer is a genetic disease whose growth and proliferation is driven by the dysregulation of cell signaling and an aberrant metabolism. A better understanding of signaling dysregulation dynamics in cancer cells would inform the development of more effective therapies. In this respect, an agent-based model of cellular pathways is developed to study the dynamics of the cell signaling circuit in closed loop with cell metabolism. The model focuses on signaling pathways that involve frequently altered cancer genes. This would support explorations of therapeutic strategies aimed at derailing cancer proliferation through disruptions of major oncogenic pathways.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121586031","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":"Understanding Cyber-physical Resilience From A Power System Perspective","authors":"Nancy A. Mohamed, M. Salama","doi":"10.1109/CCECE47787.2020.9255805","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255805","url":null,"abstract":"Resilience, as a concept, has been recently become a strategic objective for power grids. However, there are still some conceptual unclarities. Resilience is often mistakenly used as a synonym for reliability. A power system can be reliable but not resilient. This paper clarifies this misconception. It discusses resilience definitions, cycle, and states to better understand resilience aspects. It also discusses power grids' new vulnerabilities and sheds the light on both cyber-physical resilience and cyber contingency concepts. Finally, it reviews the strategies used to enhance grid resilience.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125242853","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}
T. Carvalho, C. Costa, J. Mombach, Cristiane B. R. Ferreira, D. Fernandes, Fabrízzio Soares
{"title":"PETA-System – A Piano Expression Teaching Aid System","authors":"T. Carvalho, C. Costa, J. Mombach, Cristiane B. R. Ferreira, D. Fernandes, Fabrízzio Soares","doi":"10.1109/CCECE47787.2020.9255820","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255820","url":null,"abstract":"In order to provide good musical performance to the piano, the wrist movement is very important during the performance of experienced players. However, to acquire proficiency in this technique is not a simple task since it is not simple to find formal representation, such as sheet music, to provide information about body movements. Moreover, piano teachers have to spend lots of effort to demonstrate an exercise to a single student, while many times they have to deal with many students in a classroom. Therefore, learning this approach requires not only observing experienced players but also, try to reproduce their execution to get practice. In this work, we propose PETA-System: a computer aid tool for teaching wrist movements on piano playing. The tool provides an interactive interface on which a tutor can record video musical excerpts to be performed by a learner. To evaluate the system, we carried out tests with actual learners to verify the improvement of the learning experience of wrist movement. A result has shown that the tool provided a stimulating learning environment for the student while reducing teacher efforts since they can share a time monitoring many students at the same time rather than a single one.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130096063","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 0.3V 15.6MHz 7T SRAM with Boosted Write and Read Worldlines","authors":"M. Al-Fayyad, K. Abugharbieh","doi":"10.1109/CCECE47787.2020.9255734","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255734","url":null,"abstract":"An ultra-low power 7T -based SRAM system is proposed. The seven-transistor cells are used with write and read wordlines boost assist circuits: WWLB and RWLB. A low power switching PMOS sense amplifier (SPSA) is also presented. The read and write assist circuits utilize charge pumps that generate voltages above VDD and below ground to improve speed of operation. The proposed system works properly at a very low supply voltage equal to 0.3 V. For a 32 Kb system, typical power and energy consumption are 0.147 mW and 3.82 pJ, respectively. The operating frequency is 15.6 MHz and the static noise margin, SNM, is 55mV. All circuits were simulated in Hspice using 28nm CMOS technology devices.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125985794","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}
Soumava Bhattacharjee, Sukanta Halder, Animesh Kundu, K. Iyer, N. Kar
{"title":"Artificial Neural Network Based Improved Modulation Strategy for GaN–based Inverter in EV","authors":"Soumava Bhattacharjee, Sukanta Halder, Animesh Kundu, K. Iyer, N. Kar","doi":"10.1109/CCECE47787.2020.9255829","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255829","url":null,"abstract":"Wide–bandgap (WBG) device based high–frequency inverters using Gallium Nitride (GaN) switches are gaining significant research attention in the field of electric vehicles (EVs) due to their potential to operate at higher switching frequencies with improved efficiency as compared to the available power devices. However, the computation time of the control algorithm plays a significant role in the effective control and operation of such high–frequency converters. This paper presents an advanced artificial neural network (ANN) based improved space vector pulse width modulation (SVPWM) control for Gallium Nitride based inverter in EV application. The proposed neural–network (NN) based control technique has two core objectives: the first objective is to overcome the processing speed of the complex algorithm during high switching frequency operation and hence reduce the computation time which is the major challenge in WBG device–based inverter control. The second objective is to minimize the GaN inverter switching losses and to improve the overall performance of the inverter. The NN based SVPWM is trained using the reference voltage to get the modulated signal for pulse generation, thereby reducing the computation time and improving the performance of the inverter. The proposed ANN–based improved switching strategy has been validated experimentally using a GaN inverter and a comparative performance analysis with a conventional SVPWM technique is presented in this paper.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127747632","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}
Bamdad Vafaie, M. Shamsi, M. S. Javan, K. El-Khatib
{"title":"A New Statistical Method for Anomaly Detection in Distributed Systems","authors":"Bamdad Vafaie, M. Shamsi, M. S. Javan, K. El-Khatib","doi":"10.1109/CCECE47787.2020.9255700","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255700","url":null,"abstract":"Distributed computing systems are increasing in popularity and being widely used as a new way of large-scale data processing. However, to achieve a reliable and efficient performance in a distributed environment, it is important to deal with system anomalies as soon as they are encountered. In this paper, two novel anomaly detection algorithms will be introduced and compared with previous anomaly detection algorithms. These novel algorithms are devised based on data summarization and error prediction in comparison with previously extracted data. The result of our experiments show that the proposed methods exhibit higher performance in terms of precision and accuracy.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132648215","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":"Low Power Data Acquisition System for Noise Pollution Monitoring","authors":"Mark Lipski, M. James, P. Spachos, S. Gregori","doi":"10.1109/CCECE47787.2020.9255780","DOIUrl":"https://doi.org/10.1109/CCECE47787.2020.9255780","url":null,"abstract":"Low-power data-acquisition systems are instrumental in meeting the growing demand for Internet-of-things applications. Activity-aware wake-up circuits reduce power consumption by detecting activity in the analog domain and intelligently feeding that information to digital control systems. This paper investigates implementations of low-power audio systems with activity-aware wake-up and discrete components. Experiments are run to demonstrate the functionality of the wake-up function and estimate the power savings.","PeriodicalId":296506,"journal":{"name":"2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133895550","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}