Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation

By Nima Tajbakhsh, Laura Jeyaseelan, Q. Li, Jeffrey N. Chiang, Zhihao Wu, Xiaowei Ding. Medical Image Anal., 2019

1,007 citations recorded by the source corpus.

The medical imaging literature has witnessed remarkable progress in high-performing segmentation models based on convolutional neural networks. Despite the new performance highs, the recent advanced segmentation models still require large, representative, and high quality annotated datasets. However, rarely do we have a perfect training dataset, particularly in the field of medical imaging, where data and annotations are both expensive to acquire. Recently, a large body of research has studied the problem of medical image segmentation with imperfect datasets, tackling two major dataset limitations: scarce annotations where only limited annotated data is available for training, and weak annotations where the training data has only sparse annotations, noisy annotations, or image-level annotations. In this article, we provide a detailed review of the solutions above, summarizing both the technical novelties and empirical results. We further compare the benefits and requirements of the surveyed methodologies and provide our recommended solutions. We hope this survey article increases the community awareness of the techniques that are available to handle imperfect medical image segmentation datasets.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/374b4b7db9bc7cfff74dd37c8dadccbd67e16f87

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentation (2022, background): “well-annotated data where only experts can provide reliable and accurate annotations [17].”
Annotation-efficient deep learning for automatic medical image segmentation (2020, background): “Since various issues with the datasets may exist in reality, the effectiveness of these methods is compromised 22,26 .”
Medical Image Segmentation Using Deep Learning: A Survey (2020, background): “[26]reviewed solutions of medical image segmentation with imperfect datasets, including two major dataset limitations: scarce annotations and weak annotations.”
A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises (2020, background): “3) Medical image segmentation [15], which aims to assign labels to pixels so that the pixels with the same label form a segmented object.”
Advancing Medical Imaging Informatics by Deep Learning-Based Domain Adaptation (2020, background): “Single-modality DA adapts different image distributions within the same modality [8] (Figure 2).”
UTNet: A Hybrid Transformer Architecture for Medical Image Segmentation (2021, neutral): “Currently, the convolutional encoderdecoder architectures have made substantial progress in position-sensitive tasks, like semantic segmentation [14,11,20,17,6].”
Deep learning and medical image processing for coronavirus (COVID-19) pandemic: A survey (2020, positive): “It is an established fact that DL has gained immense momentum in the field of ML with its implementations across all sectors of human life (Tajbakhsh et al., 2020).”
CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection (2023, background): “Although a total of around 5,000 annotated abdominal CT scans are publicly available, an ingrained impression remains: medical imaging datasets are too small to develop robust AI models [13,45,55,61,78,79].”
Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectives (2022, background): “Image segmentation aims at pixel-wise classification (Tajbakhsh et al., 2020; Liu et al., 2022d).”
UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation (2019, methodology): “THE encoder-decoder networks are widely used in modern semantic and instance segmentation models [1]–[6].”
Inf-Net: Automatic COVID-19 Lung Infection Segmentation From CT Images (2020, background): “widely applied in medical segmentation task, where a frequent issue is the lack of pixel-level labeled data, even when large scale set of unlabeled image could be available [36], [42].”
Image Segmentation Using Deep Learning: A Survey (2020, neutral): “mentation has been biomedical imaging [195].”
Recent advances and clinical applications of deep learning in medical image analysis (2021, background): “Among all common image analysis tasks such as classification and detection, segmentation needs the strongest supervision (large amounts of high-quality annotations) (Tajbakhsh et al., 2020).”
AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem? (2020, methodology): “Recently, annotation-efficient methods, such as semi-supervised learning, weakly supervised learning, and continual learning, have received great attention in both computer vision and medical image analysis communities [6], [10], [11].”