2021 IEEE Integrated STEM Education Conference (ISEC)最新文献

{"title":"Classification of Skin Phenotype: Melanoma Skin Cancer","authors":"Ayushi Kumar, Ari Kapelyan, Avimanyou K. Vatsa","doi":"10.1109/ISEC52395.2021.9763999","DOIUrl":"https://doi.org/10.1109/ISEC52395.2021.9763999","url":null,"abstract":"Skin cancer (skin phenotype) is most common cancer in United State of America (USA). Skin cancer can affect anyone, regardless of skin color, race, gender, and age. The characteristics of skin phenotype of melanoma lesion has an arbitrary shape, size, uneven and rough edge, and cannot be divided in half. Further, it is a leading cause of deaths worldwide. Every year, more than 5 million patients are newly diagnosed in USA. The deadliest and serious form of skin cancer is called melanoma. The diagnosis of melanoma has been done by visual examination and manual techniques by skilled doctors. It is time consuming process and highly prone to error. The skin images captured by dermoscopy eliminates the surface reflection of skin and gives better visualization of deeper levels of skin. In spite of these, image of skin lesion has many artifacts, noises, complex nature of lesion structure. Due to these complex natures of images, the border detection, feature extraction, and classification process is a complex problem. In order to identify and predict melanoma in early stage, there is need to classify images using better classification and prediction algorithms. Therefore, there is need to make an efficient, effective, and accurate melanoma identification, classification, and prediction such that it may be identified and classified in very early stage. The goal of this poster is to review and analyze the various classification deep learning algorithms - Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) - on images of skin lesions on each one of those and test with publicly available International Skin Imaging Collaboration (ISIC) archive large data sets. Also, ISIC raw datasets will be preprocessed and resized to make the data compatible to algorithms. Moreover, the performance of these algorithms will be measures and compared using five parameters including ROC.","PeriodicalId":329844,"journal":{"name":"2021 IEEE Integrated STEM Education Conference (ISEC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134101104","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":"Sensitivity of Voter Turnouts in Presidential Elections – A Retrospective Analysis","authors":"Kavin S Sankar","doi":"10.1109/ISEC52395.2021.9763955","DOIUrl":"https://doi.org/10.1109/ISEC52395.2021.9763955","url":null,"abstract":"Voter turnout is a major swaying factor in presidential elections. One of the main missions of presidential campaigns is to rile up their political base and independents to come to the voting booths and vote for them. An important comparison is the 2016 and 2020 elections. In the 2016 election, the democrats (Clinton) won the popular vote but the republicans (Trump) managed to win more key states and won the Electoral College, and the presidency. However, in 2020, the democrats energized were smarter about campaigning and put lots of effort in increasing voter turnout in key swing states. In fact, the republicans voter turnout in 2020 increased by 17.84% in relation to 2016, but democrats increased their voter turnout by 23.43%, which allowed the democrats to win the presidency. This indicates understanding sensitivity of voter turnouts and how it affects the Electoral College is an integral part to predicting which candidate will win the presidential election. There are many minor and major factors that can significantly alter voter turnout for both parties. The objective of this research project is to understand what affects voter turnout and by how much it affects the outcome in key battleground states. Towards this, I have analyzed the 2016 election in R to understand which states had the closest elections. My analysis of percentage difference between the 2 parties’ votes at national scale shows the strategies by both parties at the county level. The republicans campaigned for the more rural areas and won many more counties than the democrats in key battleground states. On the contrary, the democrats campaigned primarily in urban and populous areas, thereby winning the popular vote but not the Electoral College. Another big factor behind the republicans’ win was that the republicans won most of the battleground states (Michigan, Florida, and North Carolina) by a close margin. All of these states had the closest margins in 2016 with Michigan being the closest state that year. Out of the top 10 closest state electoral colleges the republicans won 6 of them (102 electoral colleges) and the democrats only won 4 of them (23 electoral colleges). This analysis shows how important it is to focus campaigning in key counties relevant to their base and also sway independents towards their candidates. I intend on continuing this analysis of voter sensitivity by going through all of the elections in the 2000s. I plan to develop an analysis interface which can take user inputs to analyze the past elections. These user inputs can be a list of past close state Electoral College outcomes or it can be a change in voter turnout indicated by percentage increase/decrease towards a party in key battleground states. I also intend to analyze correlation patterns between voter turnouts and key socio-economic indicators (e.g., employment, economy and crisis). This way we can analyze the change of the close battleground states and use recent events to determine what is","PeriodicalId":329844,"journal":{"name":"2021 IEEE Integrated STEM Education Conference (ISEC)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134172908","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":"STEM-Coding using Drones","authors":"M. Roopaei, Justine Horst","doi":"10.1109/ISEC52395.2021.9764109","DOIUrl":"https://doi.org/10.1109/ISEC52395.2021.9764109","url":null,"abstract":"STEM learning from an early age is essential to confirm that students are prepared to meet the needs of the world they will inherit. Currently, there are incredible and inexpensive technologies to keep students engaged and passionate about their learning. A drone is a handson apparatus that could be incorporated in STEM for several learning applications such as class projects, experiments, and exploration. In this paper, the STEMcoding framework is designed for kids to learn coding using a drone and Scratch programming. This platform attempts to visualize their coding by drone while Scratch makes a gaming environment to enhance their programming using block-based learning. The platform has applied in a STEM event at the University of Wisconsin Platteville and the results show the benefit of utilizing drone in STEM-coding as follows: i) transforms abstract concepts into concrete learning, ii) delivers an entertaining and motivating learning environment, iii) combines STEM theory with practice, and iv) provides a hand-on skill which is dynamic and attractive.","PeriodicalId":329844,"journal":{"name":"2021 IEEE Integrated STEM Education Conference (ISEC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124930919","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 Design of the Extrusion System for Chocolate 3D Printing","authors":"Hongyi Jiang","doi":"10.1109/ISEC52395.2021.9764074","DOIUrl":"https://doi.org/10.1109/ISEC52395.2021.9764074","url":null,"abstract":"Food 3D printing is one of the newest developments in food design and manufacturing with great potential in both food recipe and industrial processing. Chocolate 3D printing, especially, has received investment from big companies like Hersey and 3D System. Unlike the traditional production of customized food, which requires extensive skilled labor and a long process of molding, 3D printing food allows users to design the shape by editing the digital model file. However, there are problems with the current chocolate printers that need to improve to make them popular. The first problem is that price is too high for individual users and small stores. Most people could not afford a printer that is about several thousand dollars, not to mention the expensive printer-specific material. Another problem is the ability of the printer. Some printers can only produce few default shapes set by the producer, so users do not have much freedom to print the shapes they want. This limits the ability that is supposed to be the biggest advantage of the 3D printer. And one of the biggest weaknesses of the current chocolate 3D printers is that they cannot perform tempering, a critical process in chocolate production. Without chocolate tempering, the final product chocolate will not have a smooth, glossy texture that is preferred for desserts. So the printer cannot be used for high-end dessert production. The goal of this research is meant to design an extrusion system of the chocolate printer to solve these problems mentioned above. The goal of the printer is that it can take chocolate chips, temper the chocolate, extrude it out and form shape according to the design. It is designed to be constructed with cheap materials while having accurate control of temperature during the printing process. This research is planned to last for two years long and now I am halfway through the first year. My current plan is to work on a commercial 3D printer and replace its extrusion system with my design. The general frame of the printer is kept because that is not the focus of this research. Right now, I have finished my design of the extruder in 3D models and start building and testing prototypes. In future research, I would expect to have the extrusion system assembled on the 3D printer and investigating the optimal working condition for chocolate printing.","PeriodicalId":329844,"journal":{"name":"2021 IEEE Integrated STEM Education Conference (ISEC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122841599","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":"Diagnosing Skin Cancer Using Artificial Intelligence and Machine Learning","authors":"Riya J. Roy","doi":"10.1109/ISEC52395.2021.9763919","DOIUrl":"https://doi.org/10.1109/ISEC52395.2021.9763919","url":null,"abstract":"Artificial Intelligence (AI) and Machine Learning (ML) have many applications in the healthcare field. A lifesaving way in which these futuristic tools can be used is to diagnose skin cancer. I developed AI & ML models that can diagnose 7 different forms of skin cancer just from a skin lesion image. My goal was to enable people to upload an image of their skin lesion, and the model will generate a diagnosis for them and a percentage for the diagnosis’ accuracy. Thus, anyone can quickly and easily receive a precise diagnosis. I first researched about skin cancer and found that 1 in 3 cancers are skin cancers, indicating its prevalence. In certain communities, access to professional healthcare is scarce, depriving patients of quality care. To alleviate this, I created a website using AI & ML where people upload a picture of a skin lesion and obtain a diagnosis. Since people upload their own pictures, I had to consider that these images are likely not professional images, but the model should still diagnose it. Hence, I performed data augmentation on my dataset of skin lesion images. I made duplicates of my dataset and manipulated them by flipping, blurring, resizing, or zooming them using OpenCV. I then created numerous machine learning models such as K Nearest Neighbors (KNN), Convolutional Neural Network (CNN), Grid Search CNN, and Transfer Learning, to determine which model best diagnoses the different skin cancers. I evaluated the models using the Receiver Operator Curve (ROC), which shows the relationship between a model’s true positive and true negative rate. To interpret the curve, I used the Area Under the Curve (AUC) metric, which compares the model to one which randomly guesses. Additionally, I plotted Confusion Matrices to view a detailed configuration of each model’s performance. After evaluating, I found that my Transfer Model performed the best. For my Transfer Learning model, I used Keras’ VGG16 Machine Learning model as the base, and added my own layers of neurons to it. I further improved this model by considering different skin tones, since a bias in the dataset and model can be dangerous to those who use it. Thus, I trained my model on images of multiple skin tones. I then deployed this model to a website I created using JavaScript and HTML. My design allows people to visit my website, upload an image of their skin lesion, and receive a diagnosis in seconds. My design met my goal of creating a model which could output a diagnosis based off of just a skin lesion image. I also created a user-friendly website that makes the process of receiving a diagnosis easy and efficient. Going forward, I want to improve my Transfer Learning model as well as explore additional machine learning models. This way I can improve the diagnostic accuracy of my model. This is crucial as a false diagnosis in the medical field can be detrimental. By continuously improving my project, I hope to help those who struggle with skin cancer.","PeriodicalId":329844,"journal":{"name":"2021 IEEE Integrated STEM Education Conference (ISEC)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122913777","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 Testing of Solar Power Heating","authors":"Victor Robila","doi":"10.1109/ISEC52395.2021.9763922","DOIUrl":"https://doi.org/10.1109/ISEC52395.2021.9763922","url":null,"abstract":"The use of non-renewable energy sources continues to impact the society’s ability to minimize its environmental footprint and impedes the efforts to slow down climate change with stark implications to the global economy, humanity’s own well-being and even existence. The importance of this topic is illustrated by the constant presence in the news as well as support for broad efforts to identify and perfect alternative sources of energy. The use of renewable energy sources, such as water, wind, or solar continues to grow, however challenges in efficient energy production and storage remain. The sun’s emanated energy constitutes one of the most readily and widely available renewable resources. Yet, converting it to usable energy still requires expensive equipment such as solar panels. This poster investigates the design of solar heating of liquids as a means for improving the energy transfer. While solar water heaters have been used for decades, improvements in design have not kept up. Various aspects, such as the heater design and coating as well as the addition of salt to the water were considered as part of the experiments for this project. In my work, I posed the hypothesis that heating salty water is more efficient than regular water because the salt would make it harder for the heat to escape. The solar heater followed a classic box design with the liquid circulating through a transparent tube inside the box and connected with a container in a closed loop system. As it heated, the liquid would travel upwards through the tube towards the container. When the water travelled into the box, it would heat up and rise outside of the box. Experiments that controlled both the light intensity and other environmental parameters (such as outside temperature) showed that the use of salt in water results in speeding up the heating process. The experiment showed, not only that the heater was successful, but also that adding a saline quality to the water would in fact make the water heat up faster and be more practical. This could be immensely practical in use at a larger size and while the increase may seem small, this could be a big part in helping eliminate the use of fossil fuels in at least one of the many sectors involving them.","PeriodicalId":329844,"journal":{"name":"2021 IEEE Integrated STEM Education Conference (ISEC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131667138","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}