Auto-Generating Weak Labels for Real & Synthetic Data to Improve Label-Scarce Medical Image Segmentation

• URL: http://arxiv.org/abs/2404.17033v1
• Date: Thu, 25 Apr 2024 20:47:08 GMT
• Title: Auto-Generating Weak Labels for Real & Synthetic Data to Improve Label-Scarce Medical Image Segmentation
• Authors: Tanvi Deshpande, Eva Prakash, Elsie Gyang Ross, Curtis Langlotz, Andrew Ng, Jeya Maria Jose Valanarasu,
• Abstract summary: We present a new approach to overcome the hurdle of costly medical image labeling by leveraging foundation models like Segment Anything Model (SAM) and its medical alternate MedSAM. Our pipeline has the ability to generate weak labels for any unlabeled medical image and subsequently use it to augment label-scarce datasets. This automation eliminates the manual prompting step in MedSAM, creating a streamlined process for generating labels for both real and synthetic images.
• Score: 5.368714143438489
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The high cost of creating pixel-by-pixel gold-standard labels, limited expert availability, and presence of diverse tasks make it challenging to generate segmentation labels to train deep learning models for medical imaging tasks. In this work, we present a new approach to overcome the hurdle of costly medical image labeling by leveraging foundation models like Segment Anything Model (SAM) and its medical alternate MedSAM. Our pipeline has the ability to generate weak labels for any unlabeled medical image and subsequently use it to augment label-scarce datasets. We perform this by leveraging a model trained on a few gold-standard labels and using it to intelligently prompt MedSAM for weak label generation. This automation eliminates the manual prompting step in MedSAM, creating a streamlined process for generating labels for both real and synthetic images, regardless of quantity. We conduct experiments on label-scarce settings for multiple tasks pertaining to modalities ranging from ultrasound, dermatology, and X-rays to demonstrate the usefulness of our pipeline. The code is available at https://github.com/stanfordmlgroup/Auto-Generate-WLs/.

Related papers

• Automatic Labels are as Effective as Manual Labels in Biomedical Images Classification with Deep Learning [32.71326861917332]
  One of the main limitations to train deep learning algorithms to perform a specific task is the need for medical experts to label data. This paper aims to investigate under which circumstances automatic labels can be adopted to train a DL model on the classification of Whole Slide Images (WSI) The application of the Semantic Knowledge Extractor Tool (SKET) algorithm to generate automatic labels leads to performance comparable to the one obtained with manual labels.
  arXiv  Detail & Related papers  (2024-06-20T14:20:50Z)
• INSITE: labelling medical images using submodular functions and semi-supervised data programming [19.88996560236578]
  Large amounts of labeled data to train deep models creates an implementation bottleneck in resource-constrained settings. We apply informed subset selection to identify a small number of most representative or diverse images from a huge pool of unlabelled data. The newly annotated images are then used as exemplars to develop several data programming-driven labeling functions.
  arXiv  Detail & Related papers  (2024-02-11T12:02:00Z)
• UniMOS: A Universal Framework For Multi-Organ Segmentation Over Label-Constrained Datasets [6.428456997507811]
  We present UniMOS, the first universal framework for achieving the utilization of fully and partially labeled images as well as unlabeled images. We incorporate a semi-supervised training module that combines consistent regularization and pseudolabeling techniques on unlabeled data. Experiments show that the framework exhibits excellent performance in several medical image segmentation tasks compared to other advanced methods.
  arXiv  Detail & Related papers  (2023-11-17T00:44:56Z)
• Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation [58.03255076119459]
  We address the task of weakly-supervised few-shot image classification and segmentation, by leveraging a Vision Transformer (ViT) Our proposed method takes token representations from the self-supervised ViT and leverages their correlations, via self-attention, to produce classification and segmentation predictions. Experiments on Pascal-5i and COCO-20i demonstrate significant performance gains in a variety of supervision settings.
  arXiv  Detail & Related papers  (2023-07-07T06:16:43Z)
• Losses over Labels: Weakly Supervised Learning via Direct Loss Construction [71.11337906077483]
  Programmable weak supervision is a growing paradigm within machine learning. We propose Losses over Labels (LoL) as it creates losses directly from ofs without going through the intermediate step of a label. We show that LoL improves upon existing weak supervision methods on several benchmark text and image classification tasks.
  arXiv  Detail & Related papers  (2022-12-13T22:29:14Z)
• BoMD: Bag of Multi-label Descriptors for Noisy Chest X-ray Classification [25.76256302330625]
  New medical imaging classification problems may need to rely on machine-generated noisy labels extracted from radiology reports. Current noisy-label learning methods designed for multi-class problems cannot be easily adapted. We propose a new method designed for the noisy multi-label CXR learning, which detects and smoothly re-labels samples from the dataset.
  arXiv  Detail & Related papers  (2022-03-03T08:04:59Z)
• Learning with less labels in Digital Pathology via Scribble Supervision from natural images [20.298424229156506]
  Cross-domain transfer learning from the natural image domain (NI) to the Digital Pathology domain is shown to be successful. We show that models trained with scribble labels yield the same performance boost as full pixel-wise segmentation labels.
  arXiv  Detail & Related papers  (2022-01-07T18:12:34Z)
• Structured Semantic Transfer for Multi-Label Recognition with Partial Labels [85.6967666661044]
  We propose a structured semantic transfer (SST) framework that enables training multi-label recognition models with partial labels. The framework consists of two complementary transfer modules that explore within-image and cross-image semantic correlations. Experiments on the Microsoft COCO, Visual Genome and Pascal VOC datasets show that the proposed SST framework obtains superior performance over current state-of-the-art algorithms.
  arXiv  Detail & Related papers  (2021-12-21T02:15:01Z)
• A Closer Look at Self-training for Zero-Label Semantic Segmentation [53.4488444382874]
  Being able to segment unseen classes not observed during training is an important technical challenge in deep learning. Prior zero-label semantic segmentation works approach this task by learning visual-semantic embeddings or generative models. We propose a consistency regularizer to filter out noisy pseudo-labels by taking the intersections of the pseudo-labels generated from different augmentations of the same image.
  arXiv  Detail & Related papers  (2021-04-21T14:34:33Z)
• Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization [112.68171734288237]
  We propose a novel framework for discriminative pixel-level tasks using a generative model of both images and labels. We learn a generative adversarial network that captures the joint image-label distribution and is trained efficiently using a large set of unlabeled images. We demonstrate strong in-domain performance compared to several baselines, and are the first to showcase extreme out-of-domain generalization.
  arXiv  Detail & Related papers  (2021-04-12T21:41:25Z)
• Unsupervised Person Re-identification via Multi-label Classification [55.65870468861157]
  This paper formulates unsupervised person ReID as a multi-label classification task to progressively seek true labels. Our method starts by assigning each person image with a single-class label, then evolves to multi-label classification by leveraging the updated ReID model for label prediction. To boost the ReID model training efficiency in multi-label classification, we propose the memory-based multi-label classification loss (MMCL)
  arXiv  Detail & Related papers  (2020-04-20T12:13:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.