Dynamic memory to alleviate catastrophic forgetting in continual learning with medical imaging

Perkonigg, Matthias; Hofmanninger, Johannes; Herold, Christian; Brink, James A.; Pianykh, Oleg S.; Prosch, Helmut; Langs, Georg

doi:10.1038/s41467-021-25858-z

Cited by 43 publications

(28 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This similarity opens different avenues for further investigation and practical opportunities. For example, studies with multi-centre, multi-vendor data sets 34 can be effortlessly scaled up and automated.…”

Section: Resultsmentioning

confidence: 99%

Assessing radiomics feature stability with simulated CT acquisitions

Flouris

Jiménez–del–Toro

Aberle

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Medical imaging quantitative features had once disputable usefulness in clinical studies. Nowadays, advancements in analysis techniques, for instance through machine learning, have enabled quantitative features to be progressively useful in diagnosis and research. Tissue characterisation is improved via the “radiomics” features, whose extraction can be automated. Despite the advances, stability of quantitative features remains an important open problem. As features can be highly sensitive to variations of acquisition details, it is not trivial to quantify stability and efficiently select stable features. In this work, we develop and validate a Computed Tomography (CT) simulator environment based on the publicly available ASTRA toolbox (www.astra-toolbox.com). We show that the variability, stability and discriminative power of the radiomics features extracted from the virtual phantom images generated by the simulator are similar to those observed in a tandem phantom study. Additionally, we show that the variability is matched between a multi-center phantom study and simulated results. Consequently, we demonstrate that the simulator can be utilised to assess radiomics features’ stability and discriminative power.

show abstract

Section: Resultsmentioning

confidence: 99%

Assessing radiomics feature stability with simulated CT acquisitions

Flouris

Jiménez–del–Toro

Aberle

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Inspired by previous research on OOD detection for semantic segmentation [9], we detect data shifts by calculating the Mahalanobis distance D M (z; µ, Σ) to the training distribution. In contrast to other methods for assessing similarity, such as the Gram distance popular in rehearsal-based continual learning [21,22], the Mahalanobis distance requires storing only µ and Σ.…”

Section: Methodsmentioning

confidence: 99%

“…Methods for task-agnostic continual learning are overwhelmingly rehearsalbased [1,2,12,21,27], i.e. store a subset of past images or features in a memory buffer, which is not admissible in many diagnostic settings due to patient privacy considerations.…”

Section: Memory Performancementioning

confidence: 99%

“…Most research follows regularization-based strategies that calculate the importance of parameters and penalize their deviation [19,30]. Approaches have also been proposed for active learning [31], others allow the storage of previous samples [21,28]. Some methods leverage feature disentanglement to alleviate forgetting [16,18] or maintain task-dependent batch normalization layers [13].…”

Section: Memory Performancementioning

confidence: 99%

See 1 more Smart Citation

Task-agnostic Continual Hippocampus Segmentation for Smooth Population Shifts

González¹,

Ranem²,

Othman³

et al. 2022

Preprint

View full text Add to dashboard Cite

Most continual learning methods are validated in settings where task boundaries are clearly defined and task identity information is available during training and testing. We explore how such methods perform in a task-agnostic setting that more closely resembles dynamic clinical environments with gradual population shifts. We propose ODEx, a holistic solution that combines out-of-distribution detection with continual learning techniques. Validation on two scenarios of hippocampus segmentation shows that our proposed method reliably maintains performance on earlier tasks without losing plasticity.

show abstract

“…Short training time allow to test more hyperparameters and optimizes the overall training process leaving more time for increasing and labelling the amount of training data available. Shorter training times also allows better model training and adaption strategies when new scanners are implemented and more images are added to the training dataset 26. …”

mentioning

confidence: 99%

Detection of metallic objects on digital radiographs with convolutional neural networks: A MRI screening tool

Lie

Lysdahlgaard

2022

Radiography

View full text Add to dashboard Cite

Introduction: Screening for metallic implants and foreign bodies before magnetic resonance imaging (MRI) examinations, are crucial for patient safety. History of health are supplied by the patient, a family member, screening of electronic health records or the picture and archive systems (PACS). PACS securely store and transmits digital radiographs (DR) and related reports with patient information. Convolutional neural networks (CNN) can be used to detect metallic objects in DRs stored in PACS. This study evaluates the accuracy of CNNs in the detection of metallic objects on DRs as an MRI screening tool. Methods: The musculoskeletal radiographs (MURA) dataset consisting of 14.863 upper extremity studies were stratified into datasets with and without metal. For each anatomical region: Elbow, finger, hand, humerus, forearm, shoulder and wrist we trained and validated CNN algorithms to classify radiographs with and without metal. Algorithm performance was evaluated with area under the receiver-operating curve (AUC), sensitivity, specificity, predictive values and accuracies compared with a reference standard of manually labelling. Results: Sensitivities, specificities and area under the ROC-curves (AUC) for the six anatomic regions ranged from 85.33% (95% CI: 78.64%e90.57%) to 100.00% (95% CI: 98.16%e100.00%), 75.44% (95% CI: 62.24%e85.87%) to 93.57% (95% CI: 88.78%e96.75%) and 0.95 to 0.99, respectively. Conclusion: CNN algorithms classify DRs with metallic objects for six different anatomic regions with near-perfect accuracy. The rapid and iterative capability of the algorithms allows for scalable expansion and as a substitute MRI screening tool for metallic objects. Implications for practice: All CNNs would be able to assist in metal detection of digital radiographs prior to MRI, an substantially decrease screening time.

show abstract

Dynamic memory to alleviate catastrophic forgetting in continual learning with medical imaging

Cited by 43 publications

References 38 publications

Assessing radiomics feature stability with simulated CT acquisitions

Assessing radiomics feature stability with simulated CT acquisitions

Task-agnostic Continual Hippocampus Segmentation for Smooth Population Shifts

Detection of metallic objects on digital radiographs with convolutional neural networks: A MRI screening tool

Contact Info

Product

Resources

About