CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images
Abstract Background Accurate segmentation of stroke lesions on MRI images is very important for neurologists in the planning of post-stroke care. Segmentation helps clinicians to better diagnose and evaluation of any treatment risks. However, manual segmentation of brain lesions relies on the experi...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-09-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-023-02289-y |
| _version_ | 1827634076181331968 |
|---|---|
| author | Yousef Gheibi Kimia Shirini Seyed Naser Razavi Mehdi Farhoudi Taha Samad-Soltani |
| author_facet | Yousef Gheibi Kimia Shirini Seyed Naser Razavi Mehdi Farhoudi Taha Samad-Soltani |
| author_sort | Yousef Gheibi |
| collection | DOAJ |
| description | Abstract Background Accurate segmentation of stroke lesions on MRI images is very important for neurologists in the planning of post-stroke care. Segmentation helps clinicians to better diagnose and evaluation of any treatment risks. However, manual segmentation of brain lesions relies on the experience of neurologists and is also a very tedious and time-consuming process. So, in this study, we proposed a novel deep convolutional neural network (CNN-Res) that automatically performs the segmentation of ischemic stroke lesions from multimodal MRIs. Methods CNN-Res used a U-shaped structure, so the network has encryption and decryption paths. The residual units are embedded in the encoder path. In this model, to reduce gradient descent, the residual units were used, and to extract more complex information in images, multimodal MRI data were applied. In the link between the encryption and decryption subnets, the bottleneck strategy was used, which reduced the number of parameters and training time compared to similar research. Results CNN-Res was evaluated on two distinct datasets. First, it was examined on a dataset collected from the Neuroscience Center of Tabriz University of Medical Sciences, where the average Dice coefficient was equal to 85.43%. Then, to compare the efficiency and performance of the model with other similar works, CNN-Res was evaluated on the popular SPES 2015 competition dataset where the average Dice coefficient was 79.23%. Conclusion This study presented a new and accurate method for the segmentation of MRI medical images using a deep convolutional neural network called CNN-Res, which directly predicts segment maps from raw input pixels. |
| first_indexed | 2024-03-09T15:08:22Z |
| format | Article |
| id | doaj.art-ed6d571787d243e395b05cee15dcab90 |
| institution | Directory Open Access Journal |
| issn | 1472-6947 |
| language | English |
| last_indexed | 2024-03-09T15:08:22Z |
| publishDate | 2023-09-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj.art-ed6d571787d243e395b05cee15dcab902023-11-26T13:32:32ZengBMCBMC Medical Informatics and Decision Making1472-69472023-09-0123111410.1186/s12911-023-02289-yCNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI imagesYousef Gheibi0Kimia Shirini1Seyed Naser Razavi2Mehdi Farhoudi3Taha Samad-Soltani4Department of Software Engineering, Faculty of Electrical and Computer Engineering, University of TabrizDepartment of Software Engineering, Faculty of Electrical and Computer Engineering, University of TabrizDepartment of Software Engineering, Faculty of Electrical and Computer Engineering, University of TabrizNeurosciences Research Center (NSRC), Tabriz University of Medical SciencesDepartment of Health Information Technology, School of Management and Medical Informatics, Tabriz University of Medical SciencesAbstract Background Accurate segmentation of stroke lesions on MRI images is very important for neurologists in the planning of post-stroke care. Segmentation helps clinicians to better diagnose and evaluation of any treatment risks. However, manual segmentation of brain lesions relies on the experience of neurologists and is also a very tedious and time-consuming process. So, in this study, we proposed a novel deep convolutional neural network (CNN-Res) that automatically performs the segmentation of ischemic stroke lesions from multimodal MRIs. Methods CNN-Res used a U-shaped structure, so the network has encryption and decryption paths. The residual units are embedded in the encoder path. In this model, to reduce gradient descent, the residual units were used, and to extract more complex information in images, multimodal MRI data were applied. In the link between the encryption and decryption subnets, the bottleneck strategy was used, which reduced the number of parameters and training time compared to similar research. Results CNN-Res was evaluated on two distinct datasets. First, it was examined on a dataset collected from the Neuroscience Center of Tabriz University of Medical Sciences, where the average Dice coefficient was equal to 85.43%. Then, to compare the efficiency and performance of the model with other similar works, CNN-Res was evaluated on the popular SPES 2015 competition dataset where the average Dice coefficient was 79.23%. Conclusion This study presented a new and accurate method for the segmentation of MRI medical images using a deep convolutional neural network called CNN-Res, which directly predicts segment maps from raw input pixels.https://doi.org/10.1186/s12911-023-02289-yIschemic strokeConvolutional networkLesion segmentationMRIInformaticsDeep learning |
| spellingShingle | Yousef Gheibi Kimia Shirini Seyed Naser Razavi Mehdi Farhoudi Taha Samad-Soltani CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images BMC Medical Informatics and Decision Making Ischemic stroke Convolutional network Lesion segmentation MRI Informatics Deep learning |
| title | CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images |
| title_full | CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images |
| title_fullStr | CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images |
| title_full_unstemmed | CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images |
| title_short | CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images |
| title_sort | cnn res deep learning framework for segmentation of acute ischemic stroke lesions on multimodal mri images |
| topic | Ischemic stroke Convolutional network Lesion segmentation MRI Informatics Deep learning |
| url | https://doi.org/10.1186/s12911-023-02289-y |
| work_keys_str_mv | AT yousefgheibi cnnresdeeplearningframeworkforsegmentationofacuteischemicstrokelesionsonmultimodalmriimages AT kimiashirini cnnresdeeplearningframeworkforsegmentationofacuteischemicstrokelesionsonmultimodalmriimages AT seyednaserrazavi cnnresdeeplearningframeworkforsegmentationofacuteischemicstrokelesionsonmultimodalmriimages AT mehdifarhoudi cnnresdeeplearningframeworkforsegmentationofacuteischemicstrokelesionsonmultimodalmriimages AT tahasamadsoltani cnnresdeeplearningframeworkforsegmentationofacuteischemicstrokelesionsonmultimodalmriimages |