2025 ASEE Annual Conference & Exposition

Undergraduate engineering students often struggle when presented with complex course projects that require critical thinking and integration of knowledge from past classes. In our junior level bioimaging course, we observed that students found it difficult to approach projects that demanded the application of coding skills learned in previous courses to solve image processing and analysis problems. This issue often leads to last-minute efforts and undue stress, which prevents students from truly understanding course material and gaining deeper insights into its content. To address this challenge, we introduce a scaffolded approach aimed at promoting self-directed learning and building productive work habits to help students solve large problems.

This work-in-progress research is being conducted at Texas A&M University university within the department of biomedical engineering. The teaching practice is being applied to coding projects in a bioimaging course where students were tasked with image processing problems designed to expose our students to active areas of research, image analysis, and to recognize the problems and shortcomings associated with the project. Specifically, the topics of the projects were utilizing medical imaging data for masking/pre-processing of the lungs, automated cell counting from histology slides, calculating ejection fraction from lung CT scans, and co-registration of image data from two imaging modalities. The projects were originally designed to each be completed within a two-week deadline. To enable this short turnaround time, the deliverables of each project report were designed so that the students would first familiarize themselves with the initial code (if provided) and data, research the topic, apply techniques from the literature, recognize shortcomings of the techniques, and provide a metric for the result with any relevant statistics and conclusions. The intent was not for the students to fully solve the problem within the two-week period. Rather, students’ abilities to approach large problems through applying their coding and statistics skills from prior classes, researching relevant coding tools, and discussing what could be done to improve shortcomings if given more time were the intended points of evaluation.

Through the first rollout of these projects, both instructors observed that students appeared to flounder with the size of the project, and as a result, they procrastinated working on the project until only a day or two before it was due. Initial check-ins regarding progress were met with little student enthusiasm or understanding. Much to the dismay of the instructors, initial polling of students about time spent on the projects was also far greater than desired, even with the delay in starting. This, along with the level of detail in the reports, indicated that students were not being productive and/or effectively utilizing their resources. Utilization of the first five minutes of each of the six class periods to discuss the “pain points” appeared to benefit the few who had started on the remaining projects, but this didn’t appear to improve early student involvement on the project for the bulk of the class. The use of a scaffolded approach with an initial deliverable three days into the last project by one of the classes resulted in better student engagement during the “pain points” discussions.

In the next iteration of the course three main modifications to the projects will include:
1. More scaffolded support designed to help students reduce each problem into tasks and progress steadily through the earlier projects
2. Student self-tracking of time spent on each project and short reflection about productivity
3. Maintenance of a Google Doc documenting their progress and for ease in reporting

The first two projects will be divided into several deliverables that feature corresponding learning modules. These modules will explore self-directed learning concepts, such as resource utilization, effective question-asking, trouble-shooting, and productive work. The final two projects will not have the learning modules but will require the reporting of time spent and shared documentation. Students will take pre- and post-course surveys about their confidence in approaching complex problems and effectively utilizing their time.

The goal is to create more independent learners with growth mindsets, giving them the skills and confidence needed to tackle complex coding challenges. By practicing these skills in the first projects, students are expected to gain confidence, continue these practices on the final projects with less instructor-provided support, identify practices that contribute to productive work, and self-reinforce time management practices. The surveys will be utilized to assess student confidence in approaching large problems in image processing and their reflections will further drive the development of the learning modules for the first two projects for future iterations.