ElixirNet: Relation-Aware Network Architecture Adaptation for Medical Lesion Detection

Jiang, Chonghe; Wang, Shaoju; Liang, Xiaodan; Xu, Hang; Xiao, Nong

doi:10.1609/aaai.v34i07.6765

Cited by 17 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar NAS-based MRI reconstruction network was introduced by EMR-NAS [29] where the search space contains eight different cells with the same kernel size 3 × 3 but different dilation rate and the connection between them Besides classification, segmentation, and reconstruction, lesion detection is another important task in medical analysis. In addition to TruncatedRPN balances positive and negative data for false-positive reduction; ElixirNet [33] proposed Auto-lesion Block (ALB) to locate the tiny-size lesion by dilated convolution with flexible receptive fields. The search space for ALB contains nine operators i.e.…”

Section: Nas For Other Medical Image Applicationsmentioning

confidence: 99%

Meta-Learning of NAS for Few-shot Learning in Medical Image Applications

Vo-Ho¹,

Yamazaki²,

Hoang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep learning methods have been successful in solving tasks in machine learning and have made breakthroughs in many sectors owing to their ability to automatically extract features from unstructured data. However, their performance relies on manual trialand-error processes for selecting an appropriate network architecture, hyperparameters for training, and pre-/post-procedures. Even though it has been shown that network architecture plays a critical role in learning feature representation feature from data and the final performance, searching for the best network architecture is computationally intensive and heavily relies on researchers' experience. Automated machine learning (AutoML) and its advanced techniques i.e. Neural Architecture Search (NAS) have been promoted to address those limitations. Not only in general computer vision tasks, but NAS has also motivated various applications in multiple areas including medical imaging. In medical imaging, NAS has significant progress in improving the accuracy of image classification, segmentation, reconstruction, and more. However, NAS requires the availability of large annotated data, considerable computation resources, and pre-defined tasks. To address such limitations, meta-learning has been adopted in the scenarios of few-shot learning and multiple tasks. In this book chapter, we first present a brief review of NAS by discussing well-known approaches in search space, search strategy, and evaluation strategy. We then introduce various NAS approaches in medical imaging with different applications such as classification, segmentation, detection, reconstruction, etc. Meta-learning in NAS for few-shot learning and multiple tasks is then explained. Finally, we describe several open problems in NAS.

show abstract

Section: Nas For Other Medical Image Applicationsmentioning

confidence: 99%

Meta-Learning of NAS for Few-shot Learning in Medical Image Applications

Vo-Ho¹,

Yamazaki²,

Hoang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…When IOU=1, it means that we have a completely correct bounding box and mask. In practice, to accept the predictions a threshold is set and if the IOU positions above it then the predictions are marked as correct and incorrect the other way around [22].…”

Section: Evaluation Criteriamentioning

confidence: 99%

DRDrV3: Complete Lesion Detection in Fundus Images Using Mask R-CNN, Transfer Learning, and LSTM

Shenavarmasouleh,

Mohammadi,

Amini

et al. 2021

Preprint

View full text Add to dashboard Cite

Medical Imaging is one of the growing fields in the world of computer vision. In this study, we aim to address the Diabetic Retinopathy (DR) problem as one of the open challenges in medical imaging. In this research, we propose a new lesion detection architecture, comprising of two sub-modules, which is an optimal solution to detect and find not only the type of lesions caused by DR, their corresponding bounding boxes, and their masks; but also the severity level of the overall case. Aside from traditional accuracy, we also use two popular evaluation criteria to evaluate the outputs of our models, which are intersection over union (IOU) and mean average precision (mAP). We hypothesize that this new solution enables specialists to detect lesions with high confidence and estimate the severity of the damage with high accuracy.

show abstract

“…Lesion detection is a fundamental task in medical imaging, as an end goal [6] or as a critical step for computer-aided diagnosis (CAD) [20]. This puts great impetus on developing powerful lesion detectors and there are many successful deep-learning efforts [25,27,10,12,15]. Because of the data-driven nature of deep learning, they rely on a large number of manually annotated images.…”

Section: Introductionmentioning

confidence: 99%

Hetero-Modal Learning and Expansive Consistency Constraints for Semi-Supervised Detection from Multi-Sequence Data

Lai¹,

Wu²,

Zhou³

et al. 2021

Preprint

View full text Add to dashboard Cite

Lesion detection serves a critical role in early diagnosis and has been well explored in recent years due to methodological advances and increased data availability. However, the high costs of annotations hinder the collection of large and completely labeled datasets, motivating semi-supervised detection approaches. In this paper, we introduce mean teacher hetero-modal detection (MTHD), which addresses two important gaps in current semi-supervised detection. First, it is not obvious how to enforce unlabeled consistency constraints across the very different outputs of various detectors, which has resulted in various compromises being used in the state of the art. Using an anchor-free framework, MTHD formulates a mean teacher approach without such compromises, enforcing consistency on the soft-output of object centers and size. Second, multi-sequence data is often critical, e.g., for abdominal lesion detection, but unlabeled data is often missing sequences. To deal with this, MTHD incorporates hetero-modal learning in its framework. Unlike prior art, MTHD is able to incorporate an expansive set of consistency constraints that include geometric transforms and random sequence combinations. We train and evaluate MTHD on liver lesion detection using the largest MR lesion dataset to date (1099 patients with > 5000 volumes). MTHD surpasses the best fully-supervised and semi-supervised competitors by 10.1% and 3.5%, respectively, in average sensitivity.

show abstract

ElixirNet: Relation-Aware Network Architecture Adaptation for Medical Lesion Detection

Cited by 17 publications

References 24 publications

Meta-Learning of NAS for Few-shot Learning in Medical Image Applications

Meta-Learning of NAS for Few-shot Learning in Medical Image Applications

DRDrV3: Complete Lesion Detection in Fundus Images Using Mask R-CNN, Transfer Learning, and LSTM

Hetero-Modal Learning and Expansive Consistency Constraints for Semi-Supervised Detection from Multi-Sequence Data

Contact Info

Product

Resources

About