IP Library › Granted Patent US 12,277,687
Granted Patent B2
US 12,277,687 · App. 17/246,032 · Granted Apr 15, 2025

Systems, methods, and apparatuses for the use of transferable visual words for AI models through self-supervised learning in the absence of manual labeling for the processing of medical imaging

Inventors: Fatemeh Haghighi (Tempe, AZ); Mohammad Reza Hosseinzadeh Taher (Tempe, AZ); Zongwei Zhou (Tempe, AZ); Jianming Liang (Tempe, AZ)
Assignee: Arizona Board of Regents on Behalf of Arizona State University
G06T5/77G06T3/04G06T7/0014G06V10/25G06V10/764G06V10/82G06T2207/20081G06T2207/20084
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,277,687
App. No.
17/246,032
Granted
Apr 15, 2025
Kind
B2
Abstract

Described herein are means for the generation of semantic genesis models through self-supervised learning in the absence of manual labeling, in which the trained semantic genesis models are then utilized for the processing of medical imaging. For instance, an exemplary system is specially configured with means for performing a self-discovery operation which crops 2D patches or crops 3D cubes from similar patient scans received at the system as input; means for transforming each anatomical pattern represented within the cropped 2D patches or the cropped 3D cubes to generate transformed 2D anatomical patterns or transformed 3D anatomical patterns; means for performing a self-classification operation of the transformed anatomical patterns by formulating a C-way multi-class classification task for representation learning; means for performing a self-restoration operation by recovering original anatomical patterns from the transformed 2D patches or transformed 3D cubes having transformed anatomical patterns embedded therein to learn different sets of visual representation; and means for providing a semantics-enriched pre-trained AI model having a trained encoder-decoder structure with skip connections in between based on the performance of the self-discovery operation, the self-classification operation, and the self-restoration operation. Other related embodiments are disclosed.

Claims (48)

1. A system comprising:

a memory to store instructions;

a processor to execute the instructions stored in the memory;

a receive interface to receive a plurality of unlabeled medical images obtained from a plurality of human patients;

wherein the system is specially configured to perform self-supervised learning for an artificial intelligence (AI) model having a trained encoder-decoder structure with skip connections in between and a classification head at an output of the encoder portion, and preceding a decoder portion, of the trained encoder-decoder structure, by executing the instructions via the processor for:

performing a self-discovery operation that crops two-dimensional (2D) patches or crops three-dimensional (3D) cubes representing a plurality of unique anatomical patterns each reoccurring at a respective one of a plurality of unique fixed coordinates across the plurality of unlabeled medical images, and assigns one of a plurality of pseudo labels to each of the cropped 2D patches or 3D cubes based on their respective unique fixed coordinates across the plurality of unlabeled medical images;

transforming each of the cropped 2D patches or the cropped 3D cubes to generate transformed 2D anatomical patterns or transformed 3D anatomical patterns (hereinafter “the transformed anatomical patterns”);

performing a self-classification operation on the transformed anatomical patterns to learn semantically enriched visual representations of a human body derived from the plurality of unique anatomical patterns each reoccurring at the respective one of the plurality of unique fixed coordinates across the plurality of unlabeled medical images by formulating a multi-class classification task on the plurality of pseudo labels;

performing a self-restoration operation by recovering anatomical patterns from the transformed anatomical patterns to learn different sets of the semantically enriched visual representations of the human body.

2. The system of claim 1 , wherein transforming each of the cropped 2D patches or the cropped 3D cubes to generate the transformed anatomical patterns comprises applying one or more of the following transformations to each of the cropped 2D patches or the cropped 3D cubes:

a non-linear transformation;

a local-shuffling transformation;

an out-painting transformation; and

an in-painting transformation.

3. The system of claim 1 , wherein performing the self-discovery operation further comprises:

randomly anchoring one of the plurality of human patients as a reference in the plurality of unlabeled medical images; and

searching the plurality of unlabeled medical images for nearest neighbors to the randomly anchored human patient by computing an distance of latent representation vectors to determine a set of human patients similar in appearance.

4. The system of claim 1 , wherein performing the self-discovery operation that assigns one of the plurality of pseudo labels to each of the cropped 2D patches or 3D cubes based on their respective one of a plurality of unique fixed coordinates across the plurality of unlabeled medical images, results in a new dataset, in which each cropped 2D patch and cropped 3D cube is associated with one of a plurality of classes of visual words.

5. The system of claim 1 , wherein the system further comprises an auto-encoder network which is first trained with training data to extract deep features of each of the plurality of unlabeled medical images to learn an identical mapping from each unlabeled medical image to itself.

6. The system of claim 1 , wherein the self-classification operation causes the AI model to learn from the unique anatomical patterns each reoccurring at the respective one of the plurality of fixed coordinates across the plurality of unlabeled medical images characterized by a deep semantically enriched visual representation of the human body.

7. The system of claim 1 :

wherein a classification branch of the encoder-decoder structure encodes the plurality of unique anatomical patterns into a latent space followed by a sequence of fully-connected (fc) layers; and

wherein the classification branch predicts the pseudo labels assigned to the cropped 2D patches and cropped 3D cubes representing the plurality of unique anatomical patterns.

8. The system of claim 1 :

wherein a restoration branch of the encoder-decoder structure encodes the transformed anatomical patterns generated in the transforming operation into a latent space; and

wherein the restoration branch decodes the transformed anatomical patterns to an original resolution from the latent space to recover each of the plurality of unique anatomical patterns from the corresponding transformed anatomical pattern.

9. The system of claim 1 , wherein a restoration branch of the encoder-decoder structure restores the transformed anatomical patterns by computing a distance between each unique anatomical pattern and a reconstructed anatomical pattern via a loss function comparing the reconstructed anatomical pattern with a ground truth pattern corresponding to an anatomical pattern obtained from the plurality of unlabeled medical images.

10. The system of claim 1 , further comprising the AI model applying a multi-task loss function defined on each transformed anatomical pattern as L=λ cls L cls +λ rec L rec , where λ cls and λ rec regulate weights of classification and reconstruction losses, respectively.

11. The system of claim 1 :

wherein the system is further configured to perform operations including one or both of (i) fine-tuning the trained encoder-decoder structure for target classification and segmentation tasks.

12. The system of claim 4 wherein each of the assigned pseudo labels carries information about the unique anatomical features each reoccurring at the respective one of the plurality of fixed coordinates across the plurality of unlabeled medical images, the information selected from a group of unique anatomical features consisting of:

anterior ribs 2 through 4;

anterior ribs 1 through 3;

a right pulmonary artery; and

a left ventricle.

13. The system of claim 5 , wherein once the auto-encoder network is trained, utilizing latent representation vectors, which are extracted using the auto-encoder, as an indicator of each of the plurality of human patients.

14. A non-transitory computer-readable storage media having instructions stored thereupon that, when executed by a system having at least a processor and a memory therein and being specially configured to perform self-supervised learning for an artificial intelligence (AI) model having a trained encoder-decoder structure with skip connections in between and a classification head at an output of the encoder portion, and preceding a decoder portion, of the trained encoder-decoder structure, the instructions cause the system to perform operations including:

receiving via an interface a plurality of unlabeled medical images obtained from a plurality of human patients;

performing a self-discovery operation that crops two-dimensional (2D) patches or crops three-dimensional (3D) cubes representing a plurality of unique anatomical patterns each reoccurring at a respective one of a plurality of unique fixed coordinates across the plurality of unlabeled medical images, and assigns one of a plurality of pseudo labels to each of the cropped 2D patches or 3D cubes based on their respective one of the unique plurality of fixed coordinates across the plurality of unlabeled medical images;

transforming each of the cropped 2D patches or the cropped 3D cubes to generate transformed 2D anatomical patterns or transformed 3D anatomical patterns (hereinafter “the transformed anatomical patterns”);

performing a self-classification operation on the transformed anatomical patterns to learn semantically enriched visual representations of a human body derived from the plurality of unique anatomical patterns each reoccurring at the respective one of the plurality of unique fixed coordinates across the plurality of unlabeled medical images by formulating a classification task on the plurality of pseudo labels; and

performing a self-restoration operation by recovering anatomical patterns from the transformed anatomical patterns to learn different sets of the semantically enriched visual representations of the human body.

15. A method performed by a system having at least a processor and a memory therein to execute instructions and being specially configured to perform self-supervised learning for an artificial intelligence (AI) model having a trained encoder-decoder structure with skip connections in between and a classification head at an output of the encoder portion, and preceding a decoder portion, of the trained encoder-decoder structure, wherein the method comprises:

receiving via an interface a plurality of unlabeled medical images obtained from a plurality of human patients;

performing a self-discovery operation that crops two-dimensional (2D) patches or crops three-dimensional (3D) cubes representing a plurality of unique anatomical patterns each reoccurring at a respective one of a plurality of unique fixed coordinates across the plurality of unlabeled medical images, and assigns one of a plurality of pseudo labels to each of the cropped 2D patches or 3D cubes based on their respective one of the plurality of unique fixed coordinates across the plurality of unlabeled medical images;

transforming each of the cropped 2D patches or the cropped 3D cubes to generate transformed 2D anatomical patterns or transformed 3D anatomical patterns (hereinafter “the transformed anatomical patterns”);

performing a self-classification operation on the transformed anatomical patterns to learn semantically enriched visual representations of a human body derived from the plurality of unique anatomical patterns each reoccurring at the respective one of the plurality of unique fixed coordinates across the plurality of unlabeled medical images by formulating a classification task on the plurality of pseudo labels; and

performing a self-restoration operation by recovering anatomical patterns from the transformed anatomical patterns to learn different sets of the semantically enriched visual representations of the human body.

Assignments (2)
CONFIRMATORY LICENSE Recorded Nov 14, 2023
From: ARIZONA STATE UNIVERSITY-TEMPE CAMPUS
To: NATIONAL INSTITUTES OF HEALTH (NIH), U.S. DEPT. OF HEALTH AND HUMAN SERVICES (DHHS), U.S. GOVERNMENT
Reel/Frame 065565/0995 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 11, 2023
From: HAGHIGHI, FATEMEH; HOSSEINZADEH TAHER, MOHAMMAD REZA; ZHOU, ZONGWEI; LIANG, JIANMING
To: ARIZONA BOARD OF REGENTS ON BEHALF OF ARIZONA STATE UNIVERSITY
Reel/Frame 064864/0644 →
Continuity (3)
Provisional Application 63110265 · Nov 5, 2020
Provisional Application 63018335 · Apr 30, 2020
Related Publication 20210343014A1 · Nov 4, 2021
References Cited (80)
US 8204842B1 · Zhang et al. · 2012 [cited by applicant]
US 9811765B2 · Wang et al. · 2017 [cited by applicant]
US 20180196873A1 · Yerebakan et al. · 2018 [cited by applicant]
US 20190057774A1 · Velez et al. · 2019 [cited by applicant]
US 20200069973A1 · Lou · 2020 [cited by examiner]
US 20210319556A1 · Chauhan · 2021 [cited by examiner]
US 20220309811A1 · Haghighi et al. · 2022 [cited by applicant]
Chen et al., Self-supervised learning for medical image analysis using image context restoration, Jul. 2018, IEEE. (Year: 2018). [cited by examiner]
Tajbakhsh, N. et al., “Embracing imperfect datasets: A review of deep learning solutions for medical image segmentation,” Medical Image Analysis, 60, 2020, p. 101693. [cited by applicant]
Vincent, P. et al., “Extracting and composing robust features with denoising autoencoders,” Proceedings of the 25th International Conference on Machine Learning, 2008, pp. 1096-1103. [cited by applicant]
Wang, H. et al., “Comparison of machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer from 18 F-FDG PET/CT images,” EJNMMI research 7(1), 2017, pp. 1-11. [cited by applicant]
Wang, X. et al., “Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases,” Proceedings of the IEEE Conference on Computer Vision an… [cited by applicant]
Wang, Y. et al., “E2-train: Training state-of-the-art CNNs with over 80% energy savings.” Advances in Neural Information Processing Systems, 32, 2019. [cited by applicant]
Wu, B. et al., “Joint learning for pulmonary nodule segmentation, attributes and malignancy prediction,” 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 2018, pp. 1109-1113, IEEE. [cited by applicant]
Yan, X. et al., “Clusterfit: Improving generalization of visual representations,” Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 6509-6518. [cited by applicant]
Yosinski, J. et al., “How transferable are features in deep neural networks?” Advances in neural information processing systems, 27, 2014, pp. 3320-3328. [cited by applicant]
Yue-Hei Ng, J. et al., “Exploiting local features from deep networks for image retrieval,” Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2015, pp. 53-61. [cited by applicant]
Zhan, X. et al., “Online deep clustering for unsupervised representation learning,” Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 6688-6697. [cited by applicant]
Zhang, L. et al., “Aet vs. aed: Unsupervised representation learning by auto-encoding transformations rather than data,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 2547-… [cited by applicant]
Zhang, R. et al., “Colorful image colorization,” Computer Vision—ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, Oct. 11-14, 2016, Proceedings, Part III 14, 2016, Springer International Publishing. [cited by applicant]
Zhou, Z. et al., “Finetuning convolutional neural networks for biomedical image analysis: actively and incrementally,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 7340-7349. [cited by applicant]
Zhou, Z. et al., “Models genesis,” Medical Image Analysis, vol. 67, 2021, p. 101840. [cited by applicant]
Zhou, Z. et al., “Models genesis: Generic autodidactic models for 3d medical image analysis,” Medical Image Computing and Computer Assisted Intervention—MICCAI 2019: 22nd International Conference, Shenzhen, China, Oct. … [cited by applicant]
Zhuang, X. et al., “Self-supervised feature learning for 3d medical images by playing a rubik's cube,” Medical Image Computing and Computer Assisted Intervention—MICCAI 2019: 22nd International Conference, Shenzhen, Chi… [cited by applicant]
Alex, V. et al., “Semisupervised learning using denoising autoencoders for brain lesion detection and segmentation,” Journal of Medical Imaging 4(4), 2017, 041311. [cited by applicant]
Arandjelovic, R. et al., “Netvlad: Cnn architecture for weakly supervised place recognition,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 5297-5307. [cited by applicant]
Ardila, D. et al., “End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography,” Nature medicine 25(6), 2019, pp. 954-961. [cited by applicant]
Armato III, S.G. et al., “The lung image database consortium (LIDC) and image database resource initiative (IDRI): a completed reference database of lung nodules on CT scans,” Medical physics 38(2), 2011, pp. 915-931. [cited by applicant]
Bai, W. et al., “Self-supervised learning for cardiac mr image segmentation by anatomical position prediction,” Medical Image Computing and Computer Assisted Intervention—MICCAI 2019: 22nd International Conference, Shen… [cited by applicant]
Bakas, S. et al., “Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge,” arXiv preprint arXiv:1811.02629, 2018. [cited by applicant]
Bengio, Y. “Learning Deep Architectures for AI”, Foundations and Trends in Machine Learning, vol. 2, No. 1, 2009. pp 1-127. [cited by applicant]
Bilic, P. et al., “The liver tumor segmentation benchmark (lits),” arXiv preprint arXiv:1901.04056, 2019. [cited by applicant]
Caron, M. et al., “Deep clustering for unsupervised learning of visual features,” Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 132-149. [cited by applicant]
Caron, M. et al., “Unsupervised pre-training of image features on non-curated data,” Proceedings of the IEEE International Conference on Computer Vision, 2019, pp. 2959-2968. [cited by applicant]
Carreira, J. et al., “Quo vadis, action recognition? a new model and the kinetics dataset,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 6299-6308. [cited by applicant]
Chaitanya, K. et al., “Contrastive learning of global and local features for medical image segmentation with limited annotations,” Advances in neural information processing systems 33, 2020, pp. 12546-12558. [cited by applicant]
Chen, S. et al., Med3d: Transfer learning for 3d medical image analysis. arXiv preprint arXiv:1904.00625, 2019. [cited by applicant]
Deshpande, A. et al., “Learning large-scale automatic image colorization,” Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 567-575. [cited by applicant]
Doersch, C. et al., “Unsupervised visual representation learning by context prediction,” Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 1422-1430. [cited by applicant]
Feng, Z. et al., “Self-supervised representation learning by rotation feature decoupling,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019, pp. 10364-10374. [cited by applicant]
Gansbeke, W.V. et al., “Scan: Learning to classify images without labels,” European conference on computer vision, 2020, pp. 268-285, Springer International Publishing. [cited by applicant]
Gibson, E. et al., “Automatic multi-organ segmentation on abdominal CT with dense V-networks,” IEEE transactions on medical imaging, vol. 37, No. 8, 2018, pp. 1822-1834. [cited by applicant]
Gibson, E. et al., “Niftynet: a deep-learning platform for medical imaging,” Computer methods and programs in biomedicine 158, 2018, pp. 113-122. [cited by applicant]
Gidaris, S. et al., “Learning representations by predicting bags of visual words,” Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 6928-6938. [cited by applicant]
Gidaris, S. et al., “Unsupervised representation learning by predicting image rotations,” arXiv preprint arXiv:1803.07728, 2018. [cited by applicant]
Gong, Y. et al., “Multi-scale orderless pooling of deep convolutional activation features,” Computer Vision—ECCV 2014: 13th European Conference, Zurich, Switzerland, Sep. 6-12, 2014, Proceedings, Part VII 13, 2014, pp. … [cited by applicant]
Goyal, P. et al., “Scaling and benchmarking self-supervised visual representation learning,” Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 6391-6400. [cited by applicant]
Haghighi, F. et al., “Learning semantics-enriched representation via self-discovery, self-classification, and self-restoration,” Medical Image Computing and Computer Assisted Intervention—MICCAI 2020: 23rd International… [cited by applicant]
Hendrycks, D. et al., “Using self-supervised learning can improve model robustness and uncertainty,” Advances in Neural Information Processing Systems, 32, 2019, pp. 15637-15648. [cited by applicant]
Huh, M. et al., “What makes imagenet good for transfer learning?,” arXiv preprint arXiv:1608.08614 (2016). [cited by applicant]
Isensee, F. et al., “Automated design of deep learning methods for biomedical image segmentation,” arXiv preprint arXiv:1904.08128, 2019. [cited by applicant]
Jenni, S. et al., “Steering self-supervised feature learning beyond local pixel statistics,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 6408-6417. [cited by applicant]
Johnson, T.B. et al., “Training deep models faster with robust, approximate importance sampling,” Advances in Neural Information Processing Systems 31, 2018, pp. 7265-7275. [cited by applicant]
Kim, D. et al., “Learning image representations by completing damaged jigsaw puzzles,” 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), 2108, pp. 793-802, IEEE. [cited by applicant]
Kornblith, S. et al., “Do better imagenet models transfer better?” Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 2661-2671. [cited by applicant]
Larsson, G. et al., “Colorization as a proxy task for visual understanding,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 6874-6883. [cited by applicant]
Larsson, G. et al., “Learning representations for automatic colorization,” Computer Vision—ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, Oct. 11-14, 2016, Proceedings, Part IV, 2016, pp. 577-593, Spri… [cited by applicant]
Lowe, D. G., “Distinctive image features from scale-invariant keypoints,” International Journal of Computer Vision, 60, 2004, pp. 91-110. [cited by applicant]
Minderer, M. et al., “Automatic shortcut removal for self-supervised representation learning,” International Conference on Machine Learning. PMLR, 2020, pp. 6927-6937. [cited by applicant]
Mundhenk, T. N. et al., “Improvements to context based self-supervised learning.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 9339-9348. [cited by applicant]
Newell, A. et al., “How useful is self-supervised pretraining for visual tasks?” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 7345-7354. [cited by applicant]
Neyshabur, B. et al., “What is being transferred in transfer learning?” Advances in neural information processing systems 33, 2020, pp. 512-523. [cited by applicant]
Noroozi, M. et al., “Unsupervised learning of visual representations by solving jigsaw puzzles,” European Conference on Computer Vision, 2016, pp. 69-84, Springer International Publishing. [cited by applicant]
Pathak, D. et al., “Context encoders: Feature learning by inpainting,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 2536-2544. [cited by applicant]
Raghu, M. Zhang, J. Kleinberg, and S. Bengio, “Transfusion: Understanding transfer learning with applications to medical imaging,” arXiv preprint arXiv:1902.07208, 2019. [cited by applicant]
Ross, T. et al., “Exploiting the potential of unlabeled endoscopic video data with self-supervised learning,” International journal of computer assisted radiology and surgery 13(6), 2018, pp. 925-933. [cited by applicant]
Sabokrou, M. et al., “Self-supervised representation learning via neighborhood-relational encoding,” Proceedings of IEEE/CVF International Conference on Computer Vision, 2019, pp. 8010-8019. [cited by applicant]
Setio, A.A.A. et al., “Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge,” Medical image analysis 42, 2017, pp. 1-13. [cited by applicant]
Siim-acr pneumothorax segmentation (2019), https://www.kaggle.com/c/siim-acr-pneumothorax-segmentation/. [cited by applicant]
Simonyan, K. et al., “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014. [cited by applicant]
Simpson, A.L. et al., “A large annotated medical image dataset for the development and evaluation of segmentation algorithms,” arXiv preprint arXiv:1902.09063, 2019. [cited by applicant]
Sivic, J. et al., “Video google: a text retrieval approach to object matching in videos,” Proceedings ninth IEEE international conference on computer vision, 2003, pp. 1470-1477, IEEE. [cited by applicant]
Tajbakhsh, N. et al., “Computer-aided pulmonary embolism detection using a novel vessel-aligned multi-planar image representation and convolutional neural networks,” Medical Image Computing and Computer-Assisted Interve… [cited by applicant]
Tajbakhsh, N. et al., “Convolutional neural networks for medical image analysis: Full training or fine tuning?,” IEEE transactions on medical imaging, vol. 35, No. 5, 2016, pp. 1299-1312. [cited by applicant]
Non-Final Office Action for U.S. Appl. No. 17/676,134, mailed Aug. 13, 2024, 14 pages. [cited by applicant]
Fan, Z., et al., “A Generic Unified Deep Model for Learning from Multiple Tasks,” 34 pages. [cited by applicant]
Guo, Z., et al., “Discriminative, Restorative, and Adversarial Learning: Stepwise Incremental Pretraining,” 15 pages. [cited by applicant]
Haghighi, F., et al., “DiRA: Discriminative, Restorative, and Adversarial Learning for Self-supervised Medical Image Analysis,” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2022)… [cited by applicant]
Haghighi, F., et al., Transferable Visual Worlds, IEEE Transactions on Medical Imaging, 2020, 21 pages. [cited by applicant]
Taher, M.R.H., et al., “CAiD: A Self-supervised Learning Framework for Empowering Instance Discrimination in Medical Imaging,” Proceedings of Machine Learning Research, 2022, 20 pages. [cited by applicant]