Automatic detection of multiple sclerosis lesions in brain magnetic resonance imaging using BIANCA
The aim of this work was to design and optimize a workflow to apply the Machine Learning classifier BIANCA (Brain Intensity AbNormalities Classification Algorithm) to detect lesions characterized by white matter T2 hyperintensity in clinical Magnetic Resonance Multiple Sclerosis datasets. The design...
| Main Author: | |
|---|---|
| Format: | masterThesis |
| Language: | eng |
| Published: |
2021
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10773/30878 |
| Country: | Portugal |
| Oai: | oai:ria.ua.pt:10773/30878 |
| Summary: | The aim of this work was to design and optimize a workflow to apply the Machine Learning classifier BIANCA (Brain Intensity AbNormalities Classification Algorithm) to detect lesions characterized by white matter T2 hyperintensity in clinical Magnetic Resonance Multiple Sclerosis datasets. The designed pipeline includes pre-processing, lesion identification and optimization of BIANCA options. The classifier has been trained and tuned on 15 cases making up the training dataset of the MICCAI 2016 (Medical Image Computing and Computer Assisted Interventions) challenge and then tested on 30 cases from the Lesjak et al. public dataset. The results obtained are in good agreement with those reported by the 13 teams concluding the MICCAI 2016 challenge, thus confirming that this algorithm can be a reliable tool to detect and classify Multiple Sclerosis lesions in Magnetic Resonance studies. |
|---|