Synthesising missing modalities for multimodal MRI segmentation
Multiple MRI images modalities are extensively utilised in medical imaging tasks such as tumour segmentation as they account for information variability and image diversity. However, in practice, it is frequent that some modalities are missing in patients' data sources, and there is data imbala...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/148195 |
| _version_ | 1826114524027027456 |
|---|---|
| author | Rajasekara Pandian Akshaya Muthu |
| author2 | Jagath C Rajapakse |
| author_facet | Jagath C Rajapakse Rajasekara Pandian Akshaya Muthu |
| author_sort | Rajasekara Pandian Akshaya Muthu |
| collection | NTU |
| description | Multiple MRI images modalities are extensively utilised in medical imaging tasks such as tumour segmentation as they account for information variability and image diversity. However, in practice, it is frequent that some modalities are missing in patients' data sources, and there is data imbalance due to varying imaging protocols and image corruption.
Rather than re-acquiring all patient modality images as a complete set, it is more feasible to use the patients' existing modalities to synthesise the missing modalities and use this improved data for tumour segmentation. Therefore, we propose a generative adversarial network (GAN) that carries out missing modality synthesis for data completion and tumour segmentation. Our experiments were carried out using the Brain Tumour Image Segmentation Benchmark 2019 (BraTS ’19) dataset and our experiments support that the synthesis of the missing modality benefitted the tumour segmentation results and produced better results compared to other experiments with missing modality. Because of an impending technical disclosure being written, some details about the proposed model has been omitted from this report. |
| first_indexed | 2024-10-01T03:40:52Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/148195 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:40:52Z |
| publishDate | 2021 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1481952021-07-23T05:30:21Z Synthesising missing modalities for multimodal MRI segmentation Rajasekara Pandian Akshaya Muthu Jagath C Rajapakse School of Computer Science and Engineering ASJagath@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Multiple MRI images modalities are extensively utilised in medical imaging tasks such as tumour segmentation as they account for information variability and image diversity. However, in practice, it is frequent that some modalities are missing in patients' data sources, and there is data imbalance due to varying imaging protocols and image corruption. Rather than re-acquiring all patient modality images as a complete set, it is more feasible to use the patients' existing modalities to synthesise the missing modalities and use this improved data for tumour segmentation. Therefore, we propose a generative adversarial network (GAN) that carries out missing modality synthesis for data completion and tumour segmentation. Our experiments were carried out using the Brain Tumour Image Segmentation Benchmark 2019 (BraTS ’19) dataset and our experiments support that the synthesis of the missing modality benefitted the tumour segmentation results and produced better results compared to other experiments with missing modality. Because of an impending technical disclosure being written, some details about the proposed model has been omitted from this report. Bachelor of Engineering (Computer Science) 2021-04-26T07:27:16Z 2021-04-26T07:27:16Z 2021 Final Year Project (FYP) Rajasekara Pandian Akshaya Muthu (2021). Synthesising missing modalities for multimodal MRI segmentation. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/148195 https://hdl.handle.net/10356/148195 en application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Rajasekara Pandian Akshaya Muthu Synthesising missing modalities for multimodal MRI segmentation |
| title | Synthesising missing modalities for multimodal MRI segmentation |
| title_full | Synthesising missing modalities for multimodal MRI segmentation |
| title_fullStr | Synthesising missing modalities for multimodal MRI segmentation |
| title_full_unstemmed | Synthesising missing modalities for multimodal MRI segmentation |
| title_short | Synthesising missing modalities for multimodal MRI segmentation |
| title_sort | synthesising missing modalities for multimodal mri segmentation |
| topic | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence |
| url | https://hdl.handle.net/10356/148195 |
| work_keys_str_mv | AT rajasekarapandianakshayamuthu synthesisingmissingmodalitiesformultimodalmrisegmentation |