Distilling Knowledge from Ensembles of Cluster-Constrained-Attention Multiple-Instance Learners for Whole Slide Image Classification

Alamudun, Folami; Dash, Sajal; Hinkle, Jacob; Hernandez Arreguin, Benjamin; Tsaris, Aristeidis (aris); Yoon, Hong-Jun

doi:10.1109/BigData55660.2022.10020938

Title: Distilling Knowledge from Ensembles of Cluster-Constrained-Attention Multiple-Instance Learners for Whole Slide Image Classification

Full Record
Other Related Research

Abstract

The peculiar nature of whole slide imaging (WSI), digitizing conventional glass slides to obtain multiple high resolution images which capture microscopic details of a patient’s histopathological features, has garnered increased interest from the computer vision research community over the last two decades. Given the unique computational space and time complexity inherent to gigapixel-size whole slide image data, researchers have proposed novel machine learning algorithms to aid in the performance of diagnostic tasks in clinical pathology. One effective algorithm represents a Whole slide image as a bag of smaller image patches, which can be represented as low-dimension image patch embeddings. Weakly supervised deep-learning methods, such as cluster-constrained-attention multiple instance learning (CLAM), have shown promising results when combined with image patch embeddings. While traditional ensemble classifiers yield improved task performance, such methods come with a steep cost in model complexity. Through knowledge distillation, it is possible to retain some performance improvements from an ensemble, while minimizing costs to model complexity. In this work, we implement a weakly supervised ensemble using clustering-constrained-attention multiple-instance learners (CLAM), which uses attention and instance-level clustering to identify task salient regions and feature extraction in whole slides. By applying logit-based and attention-based knowledge distillation, we show it ismore »« less

Authors:

^[1]; Dash, Sajal ^[1];

^[1];

^[1]; Tsaris, Aristeidis (aris) ^[1];

^[1]

ORNL

Publication Date:: 2022-12-01

Research Org.:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Org.:: USDOE Office of Science (SC)

OSTI Identifier:: 1965255

DOE Contract Number:: AC05-00OR22725

Resource Type:: Conference

Resource Relation:: Conference: The 4th International Workshop on Big Data Tools, Methods, and Use Cases for Innovative Scientific Discovery - Osaka, , Japan - 12/17/2022 5:00:00 AM-

Country of Publication:: United States

Language:: English

Citation Formats


                    Alamudun, Folami, Dash, Sajal, Hinkle, Jacob, Hernandez Arreguin, Benjamin, Tsaris, Aristeidis, and Yoon, Hong-Jun. Distilling Knowledge from Ensembles of Cluster-Constrained-Attention Multiple-Instance Learners for Whole Slide Image Classification.  United States: N. p., 2022. 
        Web.  doi:10.1109/BigData55660.2022.10020938.

Copy to clipboard


                    Alamudun, Folami, Dash, Sajal, Hinkle, Jacob, Hernandez Arreguin, Benjamin, Tsaris, Aristeidis, & Yoon, Hong-Jun. Distilling Knowledge from Ensembles of Cluster-Constrained-Attention Multiple-Instance Learners for Whole Slide Image Classification.  United States.  https://doi.org/10.1109/BigData55660.2022.10020938

Copy to clipboard


                    Alamudun, Folami, Dash, Sajal, Hinkle, Jacob, Hernandez Arreguin, Benjamin, Tsaris, Aristeidis, and Yoon, Hong-Jun. 2022.  
        "Distilling Knowledge from Ensembles of Cluster-Constrained-Attention Multiple-Instance Learners for Whole Slide Image Classification".  United States.  https://doi.org/10.1109/BigData55660.2022.10020938.  https://www.osti.gov/servlets/purl/1965255.

Copy to clipboard


                    
@article{osti_1965255,

  title        = {Distilling Knowledge from Ensembles of Cluster-Constrained-Attention Multiple-Instance Learners for Whole Slide Image Classification},

  author       = {Alamudun, Folami and Dash, Sajal and Hinkle, Jacob and Hernandez Arreguin, Benjamin and Tsaris, Aristeidis and Yoon, Hong-Jun},

  abstractNote = {The peculiar nature of whole slide imaging (WSI), digitizing conventional glass slides to obtain multiple high resolution images which capture microscopic details of a patient’s histopathological features, has garnered increased interest from the computer vision research community over the last two decades. Given the unique computational space and time complexity inherent to gigapixel-size whole slide image data, researchers have proposed novel machine learning algorithms to aid in the performance of diagnostic tasks in clinical pathology. One effective algorithm represents a Whole slide image as a bag of smaller image patches, which can be represented as low-dimension image patch embeddings. Weakly supervised deep-learning methods, such as cluster-constrained-attention multiple instance learning (CLAM), have shown promising results when combined with image patch embeddings. While traditional ensemble classifiers yield improved task performance, such methods come with a steep cost in model complexity. Through knowledge distillation, it is possible to retain some performance improvements from an ensemble, while minimizing costs to model complexity. In this work, we implement a weakly supervised ensemble using clustering-constrained-attention multiple-instance learners (CLAM), which uses attention and instance-level clustering to identify task salient regions and feature extraction in whole slides. By applying logit-based and attention-based knowledge distillation, we show it is possible to retain some performance improvements resulting from the ensemble at zero cost to model complexity.},

  doi          = {10.1109/BigData55660.2022.10020938},

  url          = {https://www.osti.gov/biblio/1965255},
  journal      = {},
number       = ,

  volume       = ,

  place        = {United States},

  year         = {2022},

  month        = {12}

}

Copy to clipboard

Conference:

View Conference (2.54 MB)

https://doi.org/10.1109/BigData55660.2022.10020938

Other availability

Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:

Export Metadata

Save to My Library

Similar records in OSTI.GOV collections:

A comparison of histopathology imaging comprehension algorithms based on multiple instance learning

Conference Saunders, Adam; Dash, Sajal; Tsaris, Aristeidis (aris); ...

Whole slide imaging (WSI), also called digital virtual microscopy, is a new imaging modality. It allows for the application of AI and machine learning methods to cancer pathology to help establish a means for the automatic diagnosis of cancer cases. However, designing machine-learning models for WSI is computationally challenging due to its required ultra-high resolution. The current state-of-the-art models use multiple instance learning (MIL). MIL is a weakly-supervised learning method in which the model uses an array of inferences from many smaller instances to make a final classification about the entire set. In the context of WSI, researchers divide themore »« less
https://doi.org/10.1117/12.2654265

Full Text Available
A Scalable Pipeline for Gigapixel Whole Slide Imaging Analysis on Leadership Class HPC Systems

Conference Dash, Sajal; Hernandez Arreguin, Benjamin; Tsaris, Aristeidis (aris); ...

Whole Slide Imaging (WSI) captures microscopic details of a patient's histopathological features at multiple res-olutions organized across different levels. Images produced by WSI are gigapixel-sized, and saving a single image in memory requires a few gigabytes which is scarce since a complicated model occupies tens of gigabytes. Performing a simple met-ric operation on these large images is also expensive. High-performance computing (HPC) can help us quickly analyze such large images using distributed training of complex deep learning models. One popular approach in analyzing these images is to divide a WSI image into smaller tiles (patches) and then train a simplermore »« less
https://doi.org/10.1109/IPDPSW55747.2022.00223

Full Text Available
Scaling Resolution of Gigapixel Whole Slide Images Using Spatial Decomposition on Convolutional Neural Networks

Conference Tsaris, Aristeidis (aris); Yoon, Hong-Jun; Dash, Sajal; ...

Gigapixel images are prevalent in scientific domains ranging from remote sensing, and satellite imagery to microscopy, etc. However, training a deep learning model at the natural resolution of those images has been a challenge in terms of both, overcoming the resource limit (e.g. HBM memory constraints), as well as scaling up to a large number of GPUs. In this paper, we trained Residual neural Networks (ResNet) on 22,528 x 22,528-pixel size images using a distributed spatial decomposition method on 2,304 GPUs on the Summit Supercomputer. We applied our method on a Whole Slide Imaging (WSI) dataset from The Cancer Genomemore »« less
https://doi.org/10.1145/3592979.3593401

Full Text Available
Integrate Light-Weight Deep Learning Tools with Internet of Things

Technical Report Sun, Yu; Yu, Dantong

Today’s smart devices, such as a collision avoidance system, rely on detection of past and current eventsand then react to mitigate and reduce the severity of the ultimate consequence. The lack of proactivenessoften renders such a detection system less useful for an emergency. The ability to predict the future basedon what has been heard, sensed, observed at the current moment is of utmost importance for real-timesystems, such as autonomous driving, safety, and security monitoring and warning. It is quite commonthat a prediction system completely reconstructs (i.e., predicts) the raw image pixels of the future frame.However it does not necessarily reflectmore »« less
Final Phase I Technical Report - Deep Learning Enabled FAIR Data Management for Center for Functional Nanomaterials

Technical Report Sun, Yu

Modern materials science experiments are pushing towards several frontiers: 1) faster data collection rates, for example the Center for Functional Nanomaterials (CFN) and National Synchrotron Light Source II (NSLS-II) at BNL alone generates data in the range of 500TB/week, 2) larger arrays of detectors with higher pixel counts, 3) multi-modal measurement of elemental, structural, chemical, and physical properties at different lengths and scales of material system, and 4) sophisticated reconstruction of the shape, density, and strain field inside a crystal. These trends, once combined, highlight an enormous challenge concerning the scale of data analytics. It is clear that modern scientificmore »« less

Similar Records