Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease
Abstract Objectives The literature regarding the use of diffusion-tensor imaging-derived metrics in the evaluation of Parkinson’s disease (PD) is controversial. This study attempted to assess the feasibility of a deep-learning-based method for detecting alterations in diffusion kurtosis measurements...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2021-09-01
|
| Series: | Brain Informatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40708-021-00139-z |
| _version_ | 1818407456209895424 |
|---|---|
| author | Junyan Sun Ruike Chen Qiqi Tong Jinghong Ma Linlin Gao Jiliang Fang Dongling Zhang Piu Chan Hongjian He Tao Wu |
| author_facet | Junyan Sun Ruike Chen Qiqi Tong Jinghong Ma Linlin Gao Jiliang Fang Dongling Zhang Piu Chan Hongjian He Tao Wu |
| author_sort | Junyan Sun |
| collection | DOAJ |
| description | Abstract Objectives The literature regarding the use of diffusion-tensor imaging-derived metrics in the evaluation of Parkinson’s disease (PD) is controversial. This study attempted to assess the feasibility of a deep-learning-based method for detecting alterations in diffusion kurtosis measurements associated with PD. Methods A total of 68 patients with PD and 77 healthy controls were scanned using scanner-A (3 T Skyra) (DATASET-1). Meanwhile, an additional five healthy volunteers were scanned with both scanner-A and an additional scanner-B (3 T Prisma) (DATASET-2). Diffusion kurtosis imaging (DKI) of DATASET-2 had an extra b shell compared to DATASET-1. In addition, a 3D-convolutional neural network (CNN) was trained from DATASET-2 to harmonize the quality of scalar measures of scanner-A to a similar level as scanner-B. Whole-brain unpaired t test and Tract-Based Spatial Statistics (TBSS) were performed to validate the differences between the PD and control groups using the model-fitting method and CNN-based method, respectively. We further clarified the correlation between clinical assessments and DKI results. Results An increase in mean diffusivity (MD) was found in the left substantia nigra (SN) in the PD group. In the right SN, fractional anisotropy (FA) and mean kurtosis (MK) values were negatively correlated with Hoehn and Yahr (H&Y) scales. In the putamen (Put), FA values were positively correlated with the H&Y scales. It is worth noting that these findings were only observed with the deep learning method. There was neither a group difference nor a correlation with clinical assessments in the SN or striatum exceeding the significance level using the conventional model-fitting method. Conclusions The CNN-based method improves the robustness of DKI and can help to explore PD-associated imaging features. |
| first_indexed | 2024-12-14T09:28:07Z |
| format | Article |
| id | doaj.art-edc8bb96da7942e181d52ed30dbcd8e8 |
| institution | Directory Open Access Journal |
| issn | 2198-4018 2198-4026 |
| language | English |
| last_indexed | 2024-12-14T09:28:07Z |
| publishDate | 2021-09-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Brain Informatics |
| spelling | doaj.art-edc8bb96da7942e181d52ed30dbcd8e82022-12-21T23:08:08ZengSpringerOpenBrain Informatics2198-40182198-40262021-09-018111210.1186/s40708-021-00139-zConvolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s diseaseJunyan Sun0Ruike Chen1Qiqi Tong2Jinghong Ma3Linlin Gao4Jiliang Fang5Dongling Zhang6Piu Chan7Hongjian He8Tao Wu9Department of Neurobiology, Neurology and Geriatrics, Xuanwu Hospital of Capital Medical University, National Clinical Research Center for Geriatric DiseaseCenter for Brain Imaging Science and Technology, College of Biomedical Engineering and Instrumental Science, Zhejiang UniversityCenter for Brain Imaging Science and Technology, College of Biomedical Engineering and Instrumental Science, Zhejiang UniversityDepartment of Neurology, Xuanwu Hospital of Capital Medical UniversityDepartment of Neurobiology, Neurology and Geriatrics, Xuanwu Hospital of Capital Medical University, National Clinical Research Center for Geriatric DiseaseDepartment of Radiology, Guang’anmen Hospital, China Academy of Chinese Medical SciencesDepartment of Neurobiology, Neurology and Geriatrics, Xuanwu Hospital of Capital Medical University, National Clinical Research Center for Geriatric DiseaseDepartment of Neurobiology, Neurology and Geriatrics, Xuanwu Hospital of Capital Medical University, National Clinical Research Center for Geriatric DiseaseCenter for Brain Imaging Science and Technology, College of Biomedical Engineering and Instrumental Science, Zhejiang UniversityDepartment of Neurobiology, Neurology and Geriatrics, Xuanwu Hospital of Capital Medical University, National Clinical Research Center for Geriatric DiseaseAbstract Objectives The literature regarding the use of diffusion-tensor imaging-derived metrics in the evaluation of Parkinson’s disease (PD) is controversial. This study attempted to assess the feasibility of a deep-learning-based method for detecting alterations in diffusion kurtosis measurements associated with PD. Methods A total of 68 patients with PD and 77 healthy controls were scanned using scanner-A (3 T Skyra) (DATASET-1). Meanwhile, an additional five healthy volunteers were scanned with both scanner-A and an additional scanner-B (3 T Prisma) (DATASET-2). Diffusion kurtosis imaging (DKI) of DATASET-2 had an extra b shell compared to DATASET-1. In addition, a 3D-convolutional neural network (CNN) was trained from DATASET-2 to harmonize the quality of scalar measures of scanner-A to a similar level as scanner-B. Whole-brain unpaired t test and Tract-Based Spatial Statistics (TBSS) were performed to validate the differences between the PD and control groups using the model-fitting method and CNN-based method, respectively. We further clarified the correlation between clinical assessments and DKI results. Results An increase in mean diffusivity (MD) was found in the left substantia nigra (SN) in the PD group. In the right SN, fractional anisotropy (FA) and mean kurtosis (MK) values were negatively correlated with Hoehn and Yahr (H&Y) scales. In the putamen (Put), FA values were positively correlated with the H&Y scales. It is worth noting that these findings were only observed with the deep learning method. There was neither a group difference nor a correlation with clinical assessments in the SN or striatum exceeding the significance level using the conventional model-fitting method. Conclusions The CNN-based method improves the robustness of DKI and can help to explore PD-associated imaging features.https://doi.org/10.1186/s40708-021-00139-zParkinson’s diseaseDiffusion kurtosis imagingConvolutional neural networkMean kurtosisKurtosis fractional anisotropyMean diffusivity |
| spellingShingle | Junyan Sun Ruike Chen Qiqi Tong Jinghong Ma Linlin Gao Jiliang Fang Dongling Zhang Piu Chan Hongjian He Tao Wu Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease Brain Informatics Parkinson’s disease Diffusion kurtosis imaging Convolutional neural network Mean kurtosis Kurtosis fractional anisotropy Mean diffusivity |
| title | Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease |
| title_full | Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease |
| title_fullStr | Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease |
| title_full_unstemmed | Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease |
| title_short | Convolutional neural network optimizes the application of diffusion kurtosis imaging in Parkinson’s disease |
| title_sort | convolutional neural network optimizes the application of diffusion kurtosis imaging in parkinson s disease |
| topic | Parkinson’s disease Diffusion kurtosis imaging Convolutional neural network Mean kurtosis Kurtosis fractional anisotropy Mean diffusivity |
| url | https://doi.org/10.1186/s40708-021-00139-z |
| work_keys_str_mv | AT junyansun convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT ruikechen convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT qiqitong convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT jinghongma convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT linlingao convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT jiliangfang convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT donglingzhang convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT piuchan convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT hongjianhe convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease AT taowu convolutionalneuralnetworkoptimizestheapplicationofdiffusionkurtosisimaginginparkinsonsdisease |