Hora de Carvalho, Gonçalo (2021) Evaluating the quality of semantic segmentation in multi-source MRI images of the heart. Bachelor's Thesis, Artificial Intelligence.
|
Text
Bachelor_thesis_Evaluating_the_quality_of_semantic_segmentation_in_multi_source_MRI_images_of_the_heart_s3450295_.pdf Download (2MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (147kB) |
Abstract
Recent research has shown advances in the analysis of cardiovascular MRI images using deep learning. However, two problems are apparent: How to measure the quality of the result of semantic segmentations and how to expose dependencies on the actual MRI apparatus used in obtaining the image data sets. The proposed method is based on traditional evaluations at the pixel level. Admittedly, it would be convenient to judge incoming samples on their fa- miliarity in relation to the training data. This would allow for filtering out inadequate samples. In order to solve this conveniently, it is proposed to compare incoming samples to prototypical centroid vectors in an embedding (sub space), by using dimensionality reduction. MRI images used for this experiment are fed through a fully connected network model trained on short-axis MRI’s of left ventricles. The machine learning model was tested using two different data sets collected from two different MRI devices, one generating the UK Biobank data and another, UMCG’s data. An optimal measurement is discovered among three standard distance calculations (SAD, SSD and mean correlation), that is, SAD. This was the best measurement of similarity in raw MRIs (non-segmented) as well as serving as a predictor of segmentation quality, as verified by the Dice metric.
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Supervisor name: | Yeung, M.W. and Benjamins, J.W. and Schomaker, L.R.B. |
| Degree programme: | Artificial Intelligence |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 22 Sep 2021 13:50 |
| Last Modified: | 22 Sep 2021 13:50 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/26114 |
Actions (login required)
 |
View Item |