Projects

Education Related Projects

NSF RED Spin-off project:

Our current assessment techniques struggle to facilitate adaptive teaching and self-directed learning, which acknowledge differences in students’ prior knowledge, skills, and attitudes. The struggle gets compounded as classroom diversity increases across educational domains, hence the need for more informative assessment practices that may better guide instructor teaching and individualized student learning. Towards this, introducing descriptive digital tags into assessment material is a useful method to effectively organize and analyze both group and individual student performance information. Furthermore, tagged assessment and instructional materials allow for categorical and chronological grouping of student data, thus allowing for automated determination of student and class strengths, learning deficiencies, and their respective progressions over time. This extracted information can be used by the instructors to verify and predict their intentioned learning objectives – empowering scalable adaptive teaching methods – and by students as a tool to support self-directed learning.

To address this gap, we have developed relevant skill and conceptual knowledge descriptive tags in tag-organized assessment (TOA). The approach leverages online assignment grading platform to provided data for creating interactive visuals of student and class performance that can be scaled to much larger class sizes. The analysis of TOA helps map performance indicators for each tag in the hierarchical network of concepts and skills in the course to teacher and student practices.

SIIP: Designing Early Interventions to Facilitate Student Study Skills in Introductory Problem-Solving Classes 

  • Yael Gartner, Computer Science
  • Benjamin Cosman, Computer Science
  • Juan Alvarez, Electrical and Computer Engineering
  • Jennifer Amos, Bioengineering 
  • Molly Goldstein (AE3 Mentor)

This team will devise and implement early intervention methods to help students improve their study skills with the underlying goal of improving retention and inclusion in engineering courses for undergraduates.

GIANT: Applying a Theoretical Understanding of Text-Based Learning Modalities To Develop New Course Modalities That Meet the Needs of Student With Disabilities

  • Hongye Liu, Computer Science
  • Lawrence Angrave, Computer Science
  • Chrystafis Vogiatzis, Industrial Engineering
  • David Dalpiaz, Statistics
  • Jennifer Amos, Bioengineering 

Our previous research has shown that students often wanted access to transcripted videos through ClassTranscribe (CT) and/or a course textbook, when these were unavailable. Building upon these findings, we propose an action-research approach to ebook development using CT based upon survey perspectives of students and faculty. This study will assess the impact of the intervention on students and their instructors with special focus on meeting the needs of students with disabilities.

Our research will study the preferred characteristics and use cases of text-based materials like textbooks from the student and instructor perspective and implement them into CT. By using an action-research theory of change, we will collaborate with a team of instructors, provisionally in ISE, Stats, CS, and BIOE, to develop technology and practices and evaluate interventions in multiple courses.

Healthcare Related Projects

ARCHES – Advanced Auscultation Audio Algorithmic Analysis (a5)

  • Adam Cross, MD, FAAP, Clinical Research Informaticist, OSF HealthCare
  • Jennifer Amos, PhD, University of Illinois Urbana-Champaign
  • Eliot Bethke, University of Illinois Urbana-Champaign

The proposed project will focus on advanced feature extraction and processing to improve analytical performance to enable end-to-end explainable output and avoid the “black box” problem so prevalent among current models in the literature. The project will use the publicly available ICBHI 2017 dataset of 920 lung audio recordings from 126 subjects. We will implement raters trained in auscultation to label adventitious lung sounds as well as inhalation, exhalation, and other notable lung sounds in our dataset, then use this data to build and test the algorithm. The output of the model will be critically reviewed by human experts, taking careful note of lung sound types and diagnoses that are challenging to label for human raters and for the algorithm.

CHA – Using a community-based approach to improve health literacy and cancer screening rates in minority populations

  • Sarah Donohue, PhD, Director, University of Illinois College of Medicine Peoria
  • Scott Barrows, Design Lab Lead, OSF HealthCare
  • Mary E. Stapel, MD, Medical Director, Community Care and Community Clinic, Medical Director, Hospital Outpatient Departments, Assistant Program Director, Internal Medicine-Pediatrics, GRowLocal (Global Rural Local Health Equity) Track Director, OSF HealthCare
  • Stephen B. Brown, MSW, LCSW, Senior Director, Social & Behavioral Health Transformation & Advocacy, Director of Preventive Emergency Medicine, University of Illinois Chicago
  • Michael G. Browne, PhD, Clinical Assistant Professor, University of Illinois Chicago
  • Jennifer R. Amos, PhD, Bioengineering University of Illinois Urbana-Champaign 

The lack of a comprehensive survey tool for specific diseases, as well as the information obtained from pilot focus groups, indicates that an intervention on barriers to screening, including health literacy, is necessary.

In this proposal, we seek to obtain baseline measures on health literacy for cancer screenings and to work on interventions to improve health literacy around these topics. Our target population is Black/African American and Hispanic/Latino individuals living in the South Side of Chicago.

ARCHES – Toward automated diagnosis and 3D visualization of seizure onset zones from SEEG, MRI and CT clinical data

  • Brad Sutton, PhD, University of Illinois Urbana-Champaign
  • Andres Maldonado, MD, OSF HealthCare
  • Jennifer Amos, PhD, University of Illinois Urbana-Champaign
  • Matthew Bramlet, MD, OSF HealthCare
  • Yogatheesan Varatharajah, PhD, University of Illinois Urbana-Champaign
  • Michael Xu, MD, OSF HealthCare

Using multimodal imaging combined with stereo electroencephalography (SEEG) to enable localization of the seizure foci and identification of nearby critical anatomy to plan the patient-specific intervention. This will transform epilepsy surgical decision making by translating existing medical data into 3D and 4D visuals; significantly increasing surgeon confidence, intervention planning, and improving patient outcomes.