Clemson University
Machine Learning Based Medical Ultrasound Image Classification and Grad-Cam Interpretation
Abstract
dc:description.abstractThis work takes a step in creating a diagnostic tool for the classification decision process of Achilles tendinopathy using ultrasound images. An attention-based multiple instance learning model is developed to classify the images. Typically, doctors capture multiple ultrasound images of the Achilles tendon during a study to determine a complete diagnosis. Multiple instance models adopt this behavior by providing a single label for a set of instances (images). The images are grouped into ”bags” at the study level and passed into the model. The MIL model then uses its attention property to assign an importance score to each image to make its final classification decision. While all bags of images are composed of ultrasound images of the Achilles tendon, its important to note that different ultrasound machines produce images with different qualities. In the diagnostic process, it is important to ensure the model can be used on ultrasound images captured from any machine. As a result, this work explores the robustness of a multiple instance learning model to different image qualities. To interpret the classification decision of the model, a post-processing step is explored to explain the decision-making process of the model. Using Gradient-weighted Class Activation Mapping, the classification decision of the model is visualized via a heatmap. Being able to interpret the decision-making process gives doctors confidence that the diagnostic classification system is accurate.
Degree
thesis:*- Name thesis:degree_name
- Master of Science (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Computer Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hemphill, Victoria C
- Contributors dc:contributor
-
- Kuang-Ching Wang
- Yongkai Wu
- Xialong Ma
Subjects
dc:subject × 9Identifiers
dc:identifier.*- Repository record dc:identifier
- https://open.clemson.edu/all_theses/4571