Machine Learning for Detection of Heart Arrhythmia
Dr. Shadnaz Asgari, Associate Professor, CSULB
Arrhythmia (irregular heart rhythm) is a common heart disorder that may cause no symptoms but can be life-threatening. Diagnosis of arrhythmia mainly relies on visual inspection of electrocardiogram (ECG) by the cardiologist. To facilitate earlier detection of arrhythmia, in this project students will learn how to design and implement an algorithm using machine learning techniques to automatically detect the most common form of arrhythmia. The students will also have the option of collecting their own ECGs and running their program on it.
User Experience and Interaction Design
Dr. Bo Fu, Assistant Professor, CSULB
Traditional approaches to designing and developing computing systems typically emphasize rigorous testing of correct system behavior, e.g. verifying that a set of system functionalities work to their specification, while interaction design is often ignored, e.g. a lack of usability and user experience testing to evaluate how well users are interacting with a system. In reality, it is often a usability or user experience issue that actually prevents a user from successfully completing a task rather than a critical system error. This workshop introduces students to designing and running experiments on user interfaces of their choice for the purpose of identifying and resolving usability issues in user-centered design and development.
Analyzing acoustic telemetry data to track sharks
Dr. Alvaro Monge, Professor, CSULB
The SharkLab at CSULB uses acoustic telemetry to study the movement of animals along the coastal waters of Southern California. They study several species, including, of course, sharks. In this project, students will visit the Shark Lab, interact with researchers in the lab, learn about acoustic telemetry and the data that has been collected using these techniques, and the research questions they’re trying to answer about the animals they’re tracking. Students will be given a data set and be asked to propose their own research questions and ways to analyze and visualize the data.
Resilient Smart Gardens
Dr. Birgit Penzenstadler, Assistant Professor, CSULB
Despite rising interest in being able to grow food ourselves for educational as well as sustainability purposes, in drought-prone Southern California, it is challenging to grow herbs and vegetables. In this project, students learn how to build a mobile app that will help design and monitor a garden according to permaculture principles. To expand on that scope, students can tie the system in with automated watering using an Arduino board fueled by solar energy.
Computer Science Education Research for Diversity and Inclusion
Dr. Debra Richardson, Professor, UCI
It is well known that women and students from some racial and ethnic groups are underrepresented in computing. While some of the reasons for this underrepresentation are fairly well-understood--for example, students from these groups tend to have less pre-college experience and less confidence in their technical skills--less is known about the experience for students in these underrepresented groups who do choose to major in computing in college. This workshop introduces students to education research projects at the college level that aim to both introduce and study educational interventions (pedagogies, curriculum, programs) aimed to help improve the experience for students from groups underrepresented in computer science at the college level. Students will be introduced to methods for designing educational interventions, as well as data analysis techniques for studying their effects.
Deep Learning for 3D Neuron Segmentation of Microscopy Images
Dr. Wenlu Zhang, Assistant Professor, CSULB
Deep learning is recently considered as one of the most breakthrough technologies in computer science. Deep learning models have been successfully applied to a variety of real-world applications such as natural image classification, generation and detection. However, in the biomedical field, deep learning is still very challenging due to the multi-modality of patterns and the incomplete medical or biological data. This workshop will introduce students to implementing and designing deep learning models in 3-dimensional (3D) microscopy images. A central challenge in neuroscience is to identify the 3D morphology of neurons from microscopy images. Students will first understand the basic concepts and techniques in deep learning, such as convolutional neural network (CNN), recurrent neural network (RNN) and generative adversarial network (GAN), etc. They will also learn how to develop 3D encoder-decoder neural network architectures to train and predict using large microscopy images in an end-to-end manner.