Imbalance-Aware Self-supervised Learning for 3D Radiomic Representations

Li, Hongwei; Xue, Fei-Fei; Chaitanya, Krishna; Luo, Shengda; Ezhov, Ivan; Wiestler, Benedikt; Zhang, Jianguo; Menze, Bjoern

doi:10.1007/978-3-030-87196-3_4

Cited by 19 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Azizi et al 19 demonstrate that grouping multiple images attributed to the same medical condition along with combining natural and medical images for contrastive SSL training improves performance. Specifically for deep radiomics applications, Li et al 32 propose targeting data imbalance in existing data and present a combined approach of traditional radiomic features and self-supervised learning representations, improving performance for discriminating tumor grade and tumor staging tasks. Li et al 33 proposed a novel selfsupervised collaborative approach for creating latent representations from radiomic features.…”

Section: Discussionmentioning

confidence: 99%

Foundation Models for Quantitative Biomarker Discovery in Cancer Imaging

Pai,

Bontempi,

Hadzic

et al. 2023

Preprint

View full text Add to dashboard Cite

Foundation models represent a recent paradigm shift in deep learning, where a single large-scale model trained on vast amounts of data can serve as the foundation for various downstream tasks. Foundation models are generally trained using self-supervised learning and excel in reducing the demand for training samples in downstream applications. This is especially important in medicine, where large labeled datasets are often scarce. Here, we developed a foundation model for imaging biomarker discovery by training a convolutional encoder through self-supervised learning using a comprehensive dataset of 11,467 radiographic lesions. The foundation model was evaluated in distinct and clinically relevant applications of imaging-based biomarkers. We found that they facilitated better and more efficient learning of imaging biomarkers and yielded task-specific models that significantly outperformed their conventional supervised counterparts on downstream tasks. The performance gain was most prominent when training dataset sizes were very limited. Furthermore, foundation models were more stable to input and inter-reader variations and showed stronger associations with underlying biology. Our results demonstrate the tremendous potential of foundation models in discovering novel imaging biomarkers that may extend to other clinical use cases and can accelerate the widespread translation of imaging biomarkers into clinical settings.

show abstract

Section: Discussionmentioning

confidence: 99%

Foundation Models for Quantitative Biomarker Discovery in Cancer Imaging

Pai,

Bontempi,

Hadzic

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, Siamese networks and teacher-student networks have become popular structures applied in medical areas. Siamese network learning for medical applications -Spitzer et al [150] utilized a Siamese network to calculate spatial distances between image patches sampled randomly from the cor-tex in random sections of the same brain. Learning to discriminate several cortical brain areas through their model implicitly indicated that the designed pretext task was suitable for high-resolution cytoarchitectonic mapping.…”

Section: Instance-instance Contrastive Learning For Medical Image Ana...mentioning

confidence: 99%

A Review of Predictive and Contrastive Self-supervised Learning for Medical Images

et al. 2023

View full text Add to dashboard Cite

Over the last decade, supervised deep learning on manually annotated big data has been progressing significantly on computer vision tasks. But, the application of deep learning in medical image analysis is limited by the scarcity of high-quality annotated medical imaging data. An emerging solution is self-supervised learning (SSL), among which contrastive SSL is the most successful approach to rivalling or outperforming supervised learning. This review investigates several state-of-the-art contrastive SSL algorithms originally on natural images as well as their adaptations for medical images, and concludes by discussing recent advances, current limitations, and future directions in applying contrastive SSL in the medical domain.

show abstract

“…However, these contrastive approaches require generating effective positive and negative pairs which is not feasible in every task such as nuclei segmentation or skin lesion segmentation, where the input samples are almost related and it is relatively hard to generate negative pairs of the samples (Li et al, 2021). Following this context, a redundancy reduction based strategy is adopted that does not require generation of positive and negative pairs for pre-training.…”

Section: Related Workmentioning

confidence: 99%