Contrastive Learning vs. Self-Learning vs. Deformable Data Augmentation in Semantic Segmentation of Medical Images: Self-Learning vs. Data Augmentation in Semantic Segmentation.
To develop a robust segmentation model, encoding the underlying features/structures of the input data is essential to discriminate the target structure from the background. To enrich the extracted feature maps, contrastive learning and self-learning techniques are employed, particularly when the siz...
| Publicado en: | Journal of Digital Imaging Vol. 37; no. 6; pp. 3217 - 3231 |
|---|---|
| Autores principales: | , |
| Formato: | diagnostic images equations & formulas pictorial tables/charts Journal Article |
| Publicado: |
Springer Nature
Dec2024
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=182283979&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 182283979 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 08971889 DOQ jtl: Journal of Digital Imaging issn: 08971889 maglogo: N pubinfo: dt: Dec2024 vid: 37 iid: 6 pid: 237 pub: Springer Nature place: New York, New York artinfo: ui: 182283979 182283979 182283979 10.1007/s10278-024-01159-x 182283979 ppf: 3217 ppct: 14 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Contrastive Learning vs. Self-Learning vs. Deformable Data Augmentation in Semantic Segmentation of Medical Images: Self-Learning vs. Data Augmentation in Semantic Segmentation. aug: au: Arabi, Hossein Zaidi, Habib affil: Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland sug: subj: Semantics Utilization Deep Learning Methods Diagnostic Imaging Education Self-Directed Learning Human Magnetic Resonance Imaging Hippocampus Radiography Descriptive Statistics Tomography, X-Ray Computed Confidence Intervals Analysis of Variance Post Hoc Analysis T-Tests ab: To develop a robust segmentation model, encoding the underlying features/structures of the input data is essential to discriminate the target structure from the background. To enrich the extracted feature maps, contrastive learning and self-learning techniques are employed, particularly when the size of the training dataset is limited. In this work, we set out to investigate the impact of contrastive learning and self-learning on the performance of the deep learning-based semantic segmentation. To this end, three different datasets were employed used for brain tumor and hippocampus delineation from MR images (BraTS and Decathlon datasets, respectively) and kidney segmentation from CT images (Decathlon dataset). Since data augmentation techniques are also aimed at enhancing the performance of deep learning methods, a deformable data augmentation technique was proposed and compared with contrastive learning and self-learning frameworks. The segmentation accuracy for the three datasets was assessed with and without applying data augmentation, contrastive learning, and self-learning to individually investigate the impact of these techniques. The self-learning and deformable data augmentation techniques exhibited comparable performance with Dice indices of 0.913 ± 0.030 and 0.920 ± 0.022 for kidney segmentation, 0.890 ± 0.035 and 0.898 ± 0.027 for hippocampus segmentation, and 0.891 ± 0.045 and 0.897 ± 0.040 for lesion segmentation, respectively. These two approaches significantly outperformed the contrastive learning and the original model with Dice indices of 0.871 ± 0.039 and 0.868 ± 0.042 for kidney segmentation, 0.872 ± 0.045 and 0.865 ± 0.048 for hippocampus segmentation, and 0.870 ± 0.049 and 0.860 ± 0.058 for lesion segmentation, respectively. The combination of self-learning with deformable data augmentation led to a robust segmentation model with no outliers in the outcomes. This work demonstrated the beneficial impact of self-learning and deformable data augmentation on organ and lesion segmentation, where no additional training datasets are needed. pubtype: Academic Journal doctype: diagnostic images equations & formulas pictorial tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|