Two graduate students will present their research at the next Beckman Institute Graduate Student Seminar: Jialu Li, electrical and computer engineering; and Edgar Mejia, materials science and engineering.
The hybrid seminar will take place at noon on Wednesday, Dec. 7 in 1005 Beckman and on Zoom. Lunch will be provided to in-person attendees.
Register in advance to attend.
Towards automatic understanding parent-child interaction patterns from family audio to monitor child mental health
Jialu LiJialu Li is a senior Ph.D. student in the department of electrical and computer engineering working with Professor Mark Hasegawa-Johnson. She was awarded the Beckman Institute Graduate Fellowship and the ECE Rambus Fellowship for the 2022-2023 academic year. Her work is about interdisciplinary research of psychology and automatic audio analysis of at-home parent/infant vocalizations for monitoring children's mental health. She also studies automatic audio analysis of rapid conversations between clinicians and children for quickly diagnosing autism spectrum disorder in children. She earned her bachelor’s degree in electrical engineering from the University of Illinois Urbana-Champaign in May 2017.
In the U.S., about 15-17% of children 2-8 years of age are estimated to have at least one diagnosed mental, behavioral, or developmental disorder. However, such disorders often go undiagnosed. Thus, more fundamental understanding of how those mental health issues reach a clinically significant level from daily emotional and behavioral disturbance is required. Child mental health problems emerge from daily interactions with family members that are repeated and reinforced over time, and previous psychology studies have discovered the attributes of parent-child interaction that are correlated with later mental health problems among children. Based on these findings, our research goal aims to develop sophisticated machine learning models that automatically predict and analyze critical home-life vocalization interactions to better support child mental health outcomes. The goal is very challenging because typical home recordings contain very few examples of events relevant to child mental health, whereas training robust ML models requires many labeled examples. In this talk, I will demonstrate how to use cutting-edge speech-processing technology by leveraging thousands of hours of unlabeled family audio data to improve performances on a small amount of labeled audio for both parent/infant speaker diarization and vocalization classifications. The state-of-the-art unsupervised speech-processing model significantly alleviates the data sparsity problems and thus advances family audio analysis to next level.
Direct-write 3D printing and upcycling of frontally polymerized thermosets
Edgar MejiaEdgar Mejia is a Graduate College Fellow in the Department of Materials Science and Engineering working with Professor Nancy Sottos. His current research focuses on developing thermoset polymers for additive manufacturing with recyclability properties. In 2018, Edgar received a Fulbright scholarship to investigate the upcycling of plastic waste into material for 3D printing in Abu Dhabi. In 2019, Edgar presented this innovative process at the Youth Climate Summit at the United Nations in New York. His interest in sustainable polymers sprouted from research exposure during his undergraduate studies at Beckman Institute for Advanced Science and Technology.
Thermoset polymers possess the necessary chemical and mechanical properties critical for achieving lightweight, durable structures in the energy, aerospace, and transportation industries. However, these materials are extremely difficult to reprocess due to their chemical crosslinks, leading to a lack of end-of-life strategies. Current methods for managing thermoset waste are usually limited to either incineration or landfill disposal. Given increasing environmental concerns surrounding both the production and disposal of thermoset polymers, transformative strategies are required for sustainable low-energy manufacturing and end-of-life management. This project explores a different end-of-life strategy for 3D-printed thermosets through chemical recycling.
Learn more about Beckman's Graduate Student Seminar Series.
Read Q&As with student researchers on Beckman's Student Researcher Spotlight page.
