A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis

  title={A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis},
  author={Samuel Budd and Emma Claire Robinson and Bernhard Kainz},
  journal={Medical image analysis},

Figures from this paper

Enabling Autonomous Medical Image Data Annotation: A human-in-the-loop Reinforcement Learning Approach

Results show that an agent training with advice positively impacts obtaining new annotations from a data set with scarce labels, which opens up new possibilities for advancing the study and implementing autonomous approaches with human advice to create a cost-effective annotation in data sets for computer-aided medical image analysis.

Diminishing Uncertainty Within the Training Pool: Active Learning for Medical Image Segmentation

This work presents a query-by-committee approach for active learning where a joint optimizer is used for the committee to explore active learning for the task of segmentation of medical imaging data sets.

Joint Semi-supervised and Active Learning for Segmentation of Gigapixel Pathology Images with Cost-Effective Labeling

This work proposes a semi-supervised active learning framework with a region-based selection criterion that iteratively selects regions for an-notation query to quickly expand the diversity and volume of the labeled set.

Deep Active Learning for Effective Pulmonary Nodule Detection

This work proposes a novel active learning approach to improve the training efficiency for a deep network-based lung nodule detection framework as well as reduce the annotation cost for the low-dose computed tomography (CT) scans.

Active Learning for Abnormalities Detection on Videos of Endoscopic Capsule

This dissertation explored AL methods to reduce the CE videos’ annotation effort by compiling smaller datasets capable of representing their content, which could improve deep learning methods for its integration in clinical analysis.

Deep Learning for Orthopedic Disease Based on Medical Image Analysis: Present and Future

This paper provides orthopedic surgeons with an overall understanding of artificial intelligence-based image analysis and the information that medical data should be treated with low prejudice, providing developers and researchers with insight into the real-world context in which clinicians are embracing medical artificial intelligence.

Continual Deep Learning Framework for Medical Media Screening and Archival

Gradual self-improvement of the AI engine narrowing down the fuzzy zone between confident positive and confident negative of diagnosis would be the key achievement of proposed continuous deep learning framework.

Reducing Annotating Load: Active Learning with Synthetic Images in Surgical Instrument Segmentation

A general embeddable method to decrease the usage of labeled real images, using active generated synthetic images, and indicates a considerable improvement in performance, especially when the budget for annotation is small.

Continual Active Learning for Efficient Adaptation of Machine Learning Models to Changing Image Acquisition

A method for continual active learning on a data stream of medical images that recognizes shifts or additions of new imaging sources domains, adapts training accordingly, and selects optimal examples for labelling.



An overview of deep learning in medical imaging focusing on MRI

Overview of deep learning in medical imaging

  • Kenji Suzuki
  • Computer Science
    Radiological Physics and Technology
  • 2017
It is shown that ML with feature input (or feature-based ML) was dominant before the introduction of deep learning, and that the major and essential difference between ML before and after deep learning is the learning of image data directly without object segmentation or feature extraction; thus, it is the source of the power of deepLearning.

The Power of Ensembles for Active Learning in Image Classification

It is found that ensembles perform better and lead to more calibrated predictive uncertainties, which are the basis for many active learning algorithms, and Monte-Carlo Dropout uncertainties perform worse.

Deep Learning in Medical Image Analysis.

This review covers computer-assisted analysis of images in the field of medical imaging and introduces the fundamentals of deep learning methods and their successes in image registration, detection of anatomical and cellular structures, tissue segmentation, computer-aided disease diagnosis and prognosis, and so on.

A survey on deep learning in medical image analysis

Advancing Medical Imaging Informatics by Deep Learning-Based Domain Adaptation

DA has emerged as a promising solution to deal with the lack of annotated training data, especially for segmentation tasks and among various DA approaches, domain transformation (DT) and latent feature-space transformation (LFST) are discussed.

Suggestive Annotation: A Deep Active Learning Framework for Biomedical Image Segmentation

A deep active learning framework that combines fully convolutional network (FCN) and active learning to significantly reduce annotation effort by making judicious suggestions on the most effective annotation areas is presented.

Active Learning for Segmentation by Optimizing Content Information for Maximal Entropy

This work enforce domain-representativeness of selected samples using a proposed penalization scheme to maximize information at the network abstraction layer, and proposes a Borda-count based sample querying scheme for selecting samples for segmentation.

Human-Machine Collaboration for Medical Image Segmentation

This paper proposes a method based on conditional Generative Adversarial Network (cGAN) to address segmentation in semi-supervised setup and in a human-in-the-loop fashion, which uses the generator in the GAN to synthesize segmentations on unlabeled data and use the discriminator to identify unreliable slices for which expert annotation is required.

AFT*: Integrating Active Learning and Transfer Learning to Reduce Annotation Efforts

A novel method to naturally integrate active learning and transfer learning ( fine-tuning) into a single framework, called AFT*, which starts directly with a pre-trained CNN to seek "worthy" samples for annotation and gradually enhance the (fine-tuned) CNN via continuous fine- Tuning.