Check out the updated conference hotel
See the transformation
>
This conference will address digital and computational pathology, from acquisition of pathology data to its management, analysis, and interpretation by observers. With the recent advances in whole slide scanners and novel instrumentation for multispectral, multiparametric tissue imaging the use of digital pathology data is growing in importance. Both the pre-clinical and clinical modeling of disease states are addressed by the developing field of computational pathology. The evolving concepts of human intelligence-artificial intelligence interactions in our understanding of image data are foundational in computational pathology. There is evidence that digital and computational pathology can improve diagnosis and grading of cancer and other pathology tasks, but there are still limitations and challenges that must be addressed before it can be fully incorporated into the clinical workflow.

Although there has been great progress in the development and application of computational pathology methods over recent years, there are several significant computational challenges specific to pathology imaging that distinguish it from its radiological counterpart. There are also unique challenges in terms of how digitized pathology specimens and correlated data are presented to, modified and interpreted by clinicians and computers.

We invite submissions that address specific problems related to image acquisition, display, interpretation, computer-aided diagnosis, and quantitative image analysis of pathology specimens. We particularly welcome contributions that identify and address challenges encountered in digital pathology imaging as well as in new approaches for image capture and analysis.

TOPIC AREAS: For this conference only
During the submission process, you will be asked to choose no more than three topics from the following list to assist in the review process.

Image Acquisition, Storage and Display Quantitative Image Analysis Information Fusion Digital/Computational Pathology and the Pathologist ;
In progress – view active session
Conference 12039

Digital and Computational Pathology

20 - 21 February 2022 | Palm 1
View Session ∨
  • 1: Segmentation of Cellular and Tissue Structures
  • 2: Grading and Classification of Pathology Images
  • Sunday/Monday Poster Viewing
  • 3: Automated Quantification of Tissue Biomarkers
  • 4: Multispectral, Multimodality, and Fused Imaging
  • 5: Multi-Stain and Multiplexed Image Analysis
  • 6: Computational Pathology: From Research to Application
  • 7: Computer-Aided Diagnosis, Prognosis, and Predictive Analysis
  • SPIE Medical Imaging Awards and Plenary Session + 50th Anniversary Panel
  • Monday Poster Session
Information

Check the conference schedule frequently for updates | Presentation times are subject to change

  • Presenters: Please inform SPIE of any changes by 4 February
  • Presentation times will be finalized on 16 February
Session 1: Segmentation of Cellular and Tissue Structures
20 February 2022 • 8:50 AM - 10:10 AM PST
12039-1
Author(s): Saarthak Kapse, Prateek Prasanna, Rajarsi Gupta, Stony Brook Univ. (United States)
20 February 2022 • 8:50 AM - 9:10 AM PST
Show Abstract + Hide Abstract
Shaped-based descriptors from Computed Tomography (CT) scans and whole slide digital pathology images were used to differentiate the two major histopathological subtypes of non-small-cell lung cancer (NSCLC). Our two hypotheses are 1) Encoding information on local heterogeneity will augment the model's classification capabilities 2) Shape-based biomarkers from radiology and pathology can complement each other. Shape features were extracted from the tumor map from pathology and radiology images. In pathology, tumor-microenvironment features were encoded by clustering the tumor map into phenotype maps. These features performed better than the features from whole tumor map. Integration of radio-pathomics performed best, achieving 0.802 AUC.
12039-2
Author(s): Alvaro Sandino Garzon, Univ. Nacional de Colombia (Colombia); Ruchika Verma, Yijang Chen, Case Western Reserve Univ. (United States); David Becerra Tovar, Eduardo Romero Castro, Univ. Nacional de Colombia (Colombia); Pallavi Tiwari, Case Western Reserve Univ. (United States)
20 February 2022 • 9:10 AM - 9:30 AM PST
Show Abstract + Hide Abstract
Glioblastoma (GBM) is a highly aggressive brain tumor and is notoriously known for its intra-tumoral heterogeneity. Diagnosis of GBM is based on histopathology confirmation via tissue samples obtained from intra-cranial biopsies. After surgical intervention, histopathology tissue slides are visually analyzed by neuro-pathologists to identify distinct GBM histological hallmarks. GBMs may be histologically undergraded based on the amount of viable tissue due to sampling errors associated with small tissue samples obtained. Consequently, there is a need for automatic identification of histopathological GBM hallmarks. In this work, we present a hierarchical deep learning strategy to automatically segment distinct GBM niches including necrosis, cellular tumor, and hyperplastic blood-vessels, on H&E digitized histopathology slides. Our approach includes first segmenting necrosis and cellular tumor regions, then identifying hyperplastic blood-vessels within cellular tumor regions.
12039-3
Author(s): Jonathan Folmsbee, Scott Doyle, Rakesh Choudhary, Univ. at Buffalo (United States); Margaret Brandwein-Weber, Icahn School of Medicine at Mount Sinai (United States); Jawaria Rahman, Case Western Reserve Univ. (United States)
20 February 2022 • 9:30 AM - 9:50 AM PST
Show Abstract + Hide Abstract
In our previous work, we have demonstrated that it is possible to use a small bootstrap set of fully annotated regions of interest (ROIs) to generate segmentation results on the WSI scale. In this work, pathologists were asked to edit the previously generated annotations on 150 WSIs, focusing on only the tumor class. Of these re-annotated WSIs, 21 were then sampled from, and used to train a new version of the classifier. Segmentation results were then generated for the remainder of the images. This work demonstrates an improvement in specificity of the segmentation of the tumor class.
12039-4
Author(s): Huu Giao Nguyen, Amjad Khan, Heather Dawson, Alessandro Lugli, Inti Zlobec, Univ. Bern (Switzerland)
20 February 2022 • 9:50 AM - 10:10 AM PST
Show Abstract + Hide Abstract
Precise tissue segmentation of histopathology images is often a crucial step in computational pathology pipelines. Here, a group affinity weakly supervised segmentation method is proposed to conquer this task. First, we create a cluster image by extracting the pixel visual feature using CNN and clustering it into different classes. Then, we create a target image by refining this cluster image with the constraints on prior selected tissue. Finally, a backpropagation process is considered to evaluate the error signals between cluster and target images and update the network parameters. We validate our method with extracellular mucin-to-tumor area quantification using a colorectal cancer dataset with 163 H&E WSIs from 97 patients. Inter-observer agreement between pathologists and the proposed algorithm is 0.917. This result is a high average performance and excellent reliability when applied to histopathology images and possibly is a promising method for inclusion into clinical practice.
Session 2: Grading and Classification of Pathology Images
20 February 2022 • 10:50 AM - 11:50 AM PST
12039-6
Author(s): Chuheng Chen, Case Western Reserve Univ. (United States); Joseph Willis, Univ. Hospitals Cleveland Medical Ctr. (United States); Anant Madabhushi, Case Western Reserve Univ. (United States), Louis Stokes Cleveland Veterans Administration Medical Ctr. (United States)
20 February 2022 • 10:50 AM - 11:10 AM PST
Show Abstract + Hide Abstract
Metastatic carcinoma, either at time of presentation or during clinical care occurs in a large percentage of cancer patients - with the most predominant histological type being adenocarcinoma. This pilot study demonstrates the utility of analysis of hand-crafted computer extracted cellular features to identify the origination of liver metastasis. A dataset of 120 patients with primary adenocarcinomas and liver metastases from the breast, colon, esophagus, and pancreas was constructed. A combination of supervised classification and unsupervised clustering was applied to identify tumors’ origination using the computer extracted histomorphometric features from the dataset. Furthermore, a CBIR was developed and deployed to investigate the relationship between the metastatic tumors and the tumor heterogeneity within their corresponding primary tumors.
12039-9
Author(s): Rakesh Choudhary, Dhadma Balachandran, Jonathan Folmsbee, Univ. at Buffalo (United States); Jawaria Rahman, Case Western Reserve Univ. (United States); Margaret Brandwein-Weber, Icahn School of Medicine at Mount Sinai (United States); Scott Doyle, Univ. at Buffalo (United States)
20 February 2022 • 11:10 AM - 11:30 AM PST
12039-10
Author(s): Kristyna Manouskova, Sorbonne Univ. (France); Valentin Abadie, Institut du Cerveau et de la Moelle Épinière (France); Mehdi Ounissi, Gabriel Jimenez, Sorbonne Univ. (France); Lev Stimmer, Benoit Delatour, Stanley Durrleman, Institut du Cerveau et de la Moelle Épinière (France); Daniel Racoceanu, Sorbonne Univ. (France)
20 February 2022 • 11:30 AM - 11:50 AM PST
Show Abstract + Hide Abstract
The neuropathological signs of Alzheimer’s disease include tau protein aggregates, which can take the form of neurites, neurofibrillary tangles, and neuritic plaques. Using histological whole slide images annotated by specialists, we trained deep learning models and created a state-of-the-art pipeline to detect and segment these objects. The spatial distribution and form of the proteins is hypothesised to be correlated with the advancement of the disease. Thus this study represents a baseline for a deeper understanding of Alzheimer’s disease and the clinicopathological correlation in its different subtypes.
Sunday/Monday Poster Viewing
20 February 2022 • 12:00 PM - 7:00 PM PST | Golden State Hall
Posters will be on display Sunday and Monday with extended viewing until 7:00 pm on Sunday. The poster session with authors in attendance will be Monday evening from 5:30 to 7:00 pm. Award winners will be identified with ribbons during the reception. Award announcement times are listed in the conference schedule.
Session 3: Automated Quantification of Tissue Biomarkers
20 February 2022 • 1:20 PM - 3:00 PM PST
12039-11
Author(s): Thomas E. Tavolara, Arijit Dutta, Martin V. Burks, Wake Forest Univ. School of Medicine (United States); Wei Chen, Wendy Frankel, The Ohio State Univ. Wexner Medical Ctr. (United States); Metin N. Gurcan, M. Khalid Khan Niazi, Wake Forest Univ. School of Medicine (United States)
20 February 2022 • 1:20 PM - 1:40 PM PST
Show Abstract + Hide Abstract
Tumor budding (TB) is a cluster of one to four tumor cells at the tumor invasive front. High inter- and intra- observer disagreement on H&E hampers its prognostic utility. Pan-cytokeratin IHC staining increases agreement but is costly, nonroutine, and yields false positives, complicating algorithm development. Therefore, we are developing methods to generate TB ground truth from H&E while benefiting pan-cytokeratin. Our preliminary method resulted in precision/recall of 0.3856/0.3254 on H&E, which exceeded 0.3470/0.1932 when comparing pathologists on H&E to pan-cytokeratin. Results evidence the feasibility of our method to generate ground truth for TB using H&E.
12039-12
Author(s): David R. Chambers, Southwest Research Institute (United States); Bradley B. Brimhall, The Univ. of Texas Health Science Ctr. at San Antonio (United States); Donald R. Poole, Southwest Research Institute (United States); Edward A. Medina, The Univ. of Texas Health Science Ctr. at San Antonio (United States)
20 February 2022 • 1:40 PM - 2:00 PM PST
Show Abstract + Hide Abstract
In our work, we present an approach to regressing breast cancer cellularity in patches extracted from Whole Slide Imagery (WSI) on Hematoxylin and Eosin (H&E) stains using a fully-convolutional neural network which is trained with two heads: one which computes a global average pool for weakly-labeled data (data with a cellularity score of 0-1.0) and another which enforces pixel-wise activations for strongly-labeled (segmentation) data. Our method was the top-performing algorithm of all submissions to the BreastPathQ challenge, achieving a prediction probability of 0.941.
12039-13
Author(s): Yoni Schirris, Mendel Engelaer, Andreas Panteli, Hugo M. Horlings, The Netherlands Cancer Institute (Netherlands); Efstratios Gavves, Univ. of Amsterdam (Netherlands); Jonas Teuwen, The Netherlands Cancer Institute (Netherlands)
20 February 2022 • 2:00 PM - 2:20 PM PST
Show Abstract + Hide Abstract
We present WeakSTIL, an interpretable deep learning pipeline for scoring the percentage of stromal tumor infiltrating lymphocytes (sTIL%) in H&E-stained whole-slide images (WSI). WeakSTIL compresses the WSI with a feature extractor pre-trained with self-supervised learning on unlabeled histopathology data and predicts localized sTIL% scores by using a multiple instance learning regressor that only requires WSI-level labels. WeakSTIL is at least as good as TIL detection methods at predicting the WSI-level sTIL% score, and its tile-level predictions are highly interpretable. In the future, WeakSTIL can provide consistent and interpretable sTIL% predictions to stratify cancer patients into targeted therapy arms.
12039-14
Author(s): Thomas E. Tavolara, M. Khalid Khan Niazi, Wake Forest Univ. School of Medicine (United States); Gary Tozbikian, Robert Wesolowski, The Ohio State Univ. Wexner Medical Ctr. (United States); Metin N. Gurcan, Wake Forest Univ. School of Medicine (United States)
20 February 2022 • 2:20 PM - 2:40 PM PST
Show Abstract + Hide Abstract
The presence of HER2 informs prognosis and treatment in breast cancer. In clinical practice, a pathologist assigns a score from 0 to 3+ depending on the intensity and distribution of its IHC staining. Due to the variations in staining and scoring subjectivity, we are developing methods to predict HER2 scores from HER2 and H&E slide images. Our preliminary methods achieved 88% accuracy on 0/1+ and 85% accuracy on 2+ and 3+. We further demonstrate that identified positive regions from HER2 can be transferred to H&E via registration. Qualitative results suggest that HER2 may be scored using H&E images alone.
12039-15
Author(s): Bi Song, Albert Huang, Ming-Chang Liu, Sony Corp. of America (United States)
20 February 2022 • 2:40 PM - 3:00 PM PST
Show Abstract + Hide Abstract
Residual cancer burden measures postneoadjuvant breast cancer response and is shown to be prognostic for long term survival. The assessment of cellularity is an important component of tumor burden assessment. In the current clinical practice, tumor cellularity is manually estimated by pathologists on H&E-stained slides. We present a deep ordinal regression framework to automatically assess cellularity from pathological images. We formulate the cellularity assessment as an ordinal regression problem and address by an end-to-end learning approach using deep convolutional neural networks. We evaluated the proposed methods on the SPIE BreastPathQ dataset and achieved significant higher agreement with expert pathologist scoring.
Session 4: Multispectral, Multimodality, and Fused Imaging
20 February 2022 • 3:30 PM - 5:10 PM PST
12039-16
Author(s): Laura Quintana, Samuel Ortega, Raquel Leon, Himar Fabelo, Institute of Applied Microelectronics (Spain); Francisco J. Balea-Fernández, Univ. de Las Palmas de Gran Canaria (Spain); Esther Sauras, Hospital de Tortosa Verge de la Cinta (Spain); Marylène Lejeune, Hospital de Tortosa Verge de la Cinta (Spain); Ramon Bosch, Carlos López, Hospital de Tortosa Verge de la Cinta (Spain); Gustavo M. Callicó, Institute of Applied Microelectronics (Spain)
20 February 2022 • 3:30 PM - 3:50 PM PST
Show Abstract + Hide Abstract
Hyperspectral (HS) imaging (HSI) is a novel technique that allows an improved identification of materials compared to other imaging modalities. Specifically, HSI technology applied to breast cancer histology, may significantly reduce the time of tumor diagnosis, minimizing the histopathology current bottleneck in cancer diagnosis. In this work, breast tumor slides have been digitally scanned to whole-slides and further annotated at cell level. The annotated regions have also been captured with an HS microscopic acquisition system. Supervised spectral and spatial-spectral classifications were carried out to automatically detect tumor cells from the rest of the coexisting cells in breast histological samples.
12039-17
Author(s): Ruining Deng, Vanderbilt Univ. (United States); Haichun Yang, Vanderbilt Univ. Medical Ctr. (United States); Zuhayr Asad, Zheyu Zhu, Shiru Wang, Vanderbilt Univ. (United States); Lee E. Wheless, Agnes B. Fogo, Vanderbilt Univ. Medical Ctr. (United States); Yuankai Huo, Vanderbilt Univ. (United States)
20 February 2022 • 3:50 PM - 4:10 PM PST
Show Abstract + Hide Abstract
There has been a long pursuit for precise and reproducible glomerular quantification in the field of renal pathology in both research and clinical practice. Currently, 3D glomerular identification and reconstruction of large-scale glomeruli are labor-intensive tasks, and time-consuming by manual analysis on whole slide imaging in 2D serial sectioning representation. Moreover, there are no approaches to present 3D glomerular visualization for human examination. In this paper, we introduce an end-to-end holistic deep-learning-based method that achieves automatic detection, segmentation and multi-object tracking of individual glomeruli with large-scale glomerular-registered assessment in a 3D context on WSIs. The high-resolution WSIs are the inputs, while the outputs are the 3D glomerular reconstruction and volume estimation. This pipeline achieves 81.8 in IDF1 as MOT performance, while the proposed volume estimation achieves 0.84 Spearman correlation coefficient with manual annotation
12039-18
Author(s): Ximing Zhou, Ling Ma, The Univ. of Texas at Dallas (United States); James Little, Amy Chen, Emory Univ. (United States); Larry Myers, Baran Sumer, The Univ. of Texas Southwestern Medical Ctr. at Dallas (United States); Baowei Fei, The Univ. of Texas at Dallas (United States)
20 February 2022 • 4:10 PM - 4:30 PM PST
12039-19
Author(s): Minh Tran, The Univ. of Texas at Dallas (United States)
20 February 2022 • 4:30 PM - 4:50 PM PST
Show Abstract + Hide Abstract
This study compares performances of image classifiers on whole slide images using different imaging modalities as training data. From 33 fixed tissues from patients with thyroid cancer, we produced three different datasets: an RGB image dataset that was acquired from a whole slide image scanner, a hyperspectral (HS) dataset that was acquired with a compact hyperspectral camera, and a HS-synthesized RGB image dataset. We show that deep learning classifier trained on HS data has an AUC-ROC of 0.966, higher than that trained on RGB and synthesized RGB data. This study demonstrates hyperspectral images can improve cancer classification performance.
12039-20
Author(s): Jenny Romell, William Twengström, KTH Royal Institute of Technology (Sweden); Carlos Fernández Moro, Karolinska Univ. Hospital (Sweden), Karolinska Institute, Karolinska Univ. Hospital Huddinge (Sweden); Jakob C. Larsson, KTH Royal Institute of Technology (Sweden); Ernesto Sparrelid, Karolinska Institute (Sweden); Mikael Björnstedt, Karolinska Univ. Hospital (Sweden), Karolinska Institute, Karolinska Univ. Hospital Huddinge (Sweden); Hans M. Hertz, KTH Royal Institute of Technology (Sweden)
20 February 2022 • 4:50 PM - 5:10 PM PST
Show Abstract + Hide Abstract
Surgery is an essential part of the curative plan for most patients affected with solid tumors and is followed by a pathological assessment not only of the tumor itself, but also of the removed surrounding tissues and in particular the resection margins. For fast assessment of the resection margins, we propose virtual histology using x-ray phase-contrast micro-CT. We demonstrate phase-contrast CT of formalin-fixed paraffin-embedded tumors from liver and pancreas and compare with classical histology images. The agreement between CT and microscopy images is excellent and we conclude that phase-contrast micro-CT offers a complement to microscopy for histopathological assessment of tumors.
Session 5: Multi-Stain and Multiplexed Image Analysis
21 February 2022 • 8:20 AM - 9:40 AM PST
12039-21
Author(s): Wenchao Han, Sunnybrook Research Institute (Canada), Univ. of Toronto (Canada); Alison Cheung, Dan Wang, Kela Liu, Sunnybrook Research Institute (Canada); Martin J. Yaffe, Anne L. Martel, Sunnybrook Research Institute (Canada), Univ. of Toronto (Canada)
21 February 2022 • 8:20 AM - 8:40 AM PST
Show Abstract + Hide Abstract
Cell phenotyping is an essential step for analyzing high-dimensional cellular information. Literature illustrated the usability for unsupervised algorithms for cell phenotyping by validating the results against manual gated cell populations. To extend the knowledge for identifying unknown inclusive cell populations in our database of multiplexed immunofluorescence images of breast cancer tissue microarrays, we explored two commonly used methods (PhenoGraph and FlowSOM) using reference standard of clinical relevant cancer subtypes that were manually assigned based on immunohistochemistry scoring of serial sections. Our results showed PhenoGraph yielded better results but much larger variations using different parameter settings than FlowSOM.
12039-22
Author(s): Shunxing Bao, Yucheng Tang, Ho Hin Lee, Riqiang Gao, Qi Yang, Xin Yu, Vanderbilt Univ. (United States); Sophie Chiron, Lori A. Coburn, Keith T. Wilson, Joseph T. Roland, Vanderbilt Univ. Medical Ctr. (United States); Bennett A. Landman, Yuankai Huo, Vanderbilt Univ. (United States)
21 February 2022 • 8:40 AM - 9:00 AM PST
Show Abstract + Hide Abstract
MxIF is an emerging technique that allows for staining multiple cellular and histological markers to stain simultaneously on a single tissue section. However, with multiple rounds of staining and bleaching, it is inevitable that the tissue may be physically depleted. A digital way of synthesizing such missing tissue is appealing. We investigate the feasibility of GAN approaches to synthesize missing tissues using 11 MxIF structural molecular markers. We integrate a multi-channel high-resolution image synthesis approach to synthesize the missing tissue marker from the remaining markers. The performance of different methods is quantitatively evaluated via the downstream cell membrane segmentation task.
12039-23
Author(s): Alison Cheung, Dan Wang, Kela Liu, Sarah Hynes, Sunnybrook Research Institute (Canada); Ben Wang, Simone Stone, Pamela Ohashi, Princess Margaret Cancer Ctr. (Canada); Martin J. Yaffe, Sunnybrook Research Institute (Canada)
21 February 2022 • 9:00 AM - 9:20 AM PST
Show Abstract + Hide Abstract
Single cell phenotyping on histology tissue section is now possible with in situ protein multiplex staining and quantitative image analysis and computation. We have developed methods that can not only assess the densities of various immune cell types in the tumor microenvironment, but also quantify their cellular arrangement and spatial relationships. We investigated and compared using binary cell counts and clustering methods to identify cells with various marker-co-expression signatures. Their localizations in the lesion were assessed with neighborhood analysis and cell-to-cell distance mapping. The correlation of immune phenotype and spatial patterns to clinical outcomes will be evaluated.
12039-25
Author(s): Laurin Herbsthofer, CBmed GmbH (Austria), BioTechMed (Austria); Barbara Ehall, Medizinischen Univ. Graz (Austria), CBmed GmbH (Austria), BioTechMed (Austria); Martina Tomberger, CBmed GmbH (Austria); Barbara Prietl, CBmed GmbH (Austria), Medizinischen Univ. Graz (Austria), BioTechMed (Austria); Thomas R. Pieber, CBmed GmbH (Austria), Medizinischen Univ. Graz (Austria), JOANNEUM RESEARCH Forschungsgesellschaft mbH (Austria); Pablo López-García, CBmed GmbH (Austria)
21 February 2022 • 9:20 AM - 9:40 AM PST
Show Abstract + Hide Abstract
We present a novel method for the procedural generation of synthetic fluorescent multiplex immunohistochemistry (fm-IHC) images using human-defined rules and a conditional Generative Adversarial Network (cGAN). It consists of three steps: (1) training of a cGAN to reverse a cell-based image compression method (Cell2Grid), (2) definition of a procedural algorithm for the generation of synthetic Cell2Grid images, and (3) using the trained cGAN to convert synthetic Cell2Grid images into high-resolution fm-IHC images. We demonstrate our method by generating synthetic fm-IHC images of murine pancreatic islets in different stages of insulitis consisting of six antibody marker channels.
Session 6: Computational Pathology: From Research to Application
21 February 2022 • 10:10 AM - 12:10 PM PST
12039-500
Author(s): Faisal Mahmood, Harvard Medical School (United States)
21 February 2022 • 10:10 AM - 11:05 AM PST
Show Abstract + Hide Abstract
Advances in digital pathology and artificial intelligence have presented the potential to build assistive tools for objective diagnosis, prognosis and therapeutic-response and resistance prediction. In this talk we will discuss: 1) Data-efficient methods for weakly-supervised whole slide classification with examples in cancer diagnosis and subtyping, allograft rejection etc. (Nature Biomedical Engineering, 2021). 2) Harnessing weakly-supervised, fast and data-efficient WSI classification for identifying origins for cancers of unknown primary (Nature, 2021). 3) Discovering integrative histology-genomic prognostic markers via interpretable multimodal deep learning (IEEE TMI, 2020). 4) Deploying weakly supervised models in low resource settings without slide scanners, network connections, computational resources and expensive microscopes. 5) Bias and fairness in computational pathology algorithms.
12039-26
Author(s): Ling Ma, The Univ. of Texas at Dallas (United States), Tianjin Univ. (China); Armand Rathgeb, Minh Tran, Baowei Fei, The Univ. of Texas at Dallas (United States)
21 February 2022 • 11:10 AM - 11:30 AM PST
Show Abstract + Hide Abstract
This study developed an unsupervised super resolution network for hyperspectral histologic imaging. High-resolution hyperspectral images of hemoxylin & eosin (H&E) stained slides were obtained at 10× magnification and downsampled 2x, 4×, and 5x to generate low-resolution hyperspectral imaging (HSI) data. High-resolution digital histologic RGB images of the same field of view (FOV) were cropped and registered to the corresponding high resolution HSI images. A modified U-Net uses the low-resolution HSI and high-resolution RGB as input and was trained with unsupervised methods to output high-resolution HSI data. The generated high-resolution HSI has similar spectral signatures and better image contrast than the original high-resolution HSI, which indicates that the super resolution network with RGB guidance can improve the image quality. The proposed method can reduce the acquisition time and save storage space taken up by hyperspectral images without compromising image quality.
12039-27
Author(s): Darshana Govind, Saber Meamardoost, Rabi Yacoub, Rudiyant Gunawan, John E. Tomaszewski, Pinaki Sarder, Univ. at Buffalo (United States)
21 February 2022 • 11:30 AM - 11:50 AM PST
Show Abstract + Hide Abstract
In this study, we investigated the importance of podocyte nuclear morphometrics in DKD. Our results revealed that podocyte nuclear texture was a key feature distinguishing diabetic from control podocytes. In tandem, the scRNA-seq highlighted EMT and apoptotic pathway, which are accompanied by variations in chromatin distribution. Additionally, a phenotypical change was observed in diabetic podocytes. These results suggest that podocyte nuclear textural features may potentially aid in the identification of injured podocytes during DKD.
12039-28
Author(s): Tianyuan Yao, Yuzhe Lu, Ruining Deng, Zheyu Zhu, Zuhayr Asad, Vanderbilt Univ. (United States); Haichun Yang, Lee E. Wheless, Agnes B. Fogo, Vanderbilt Univ. Medical Ctr. (United States); Yuankai Huo, Vanderbilt Univ. (United States)
21 February 2022 • 11:50 AM - 12:10 PM PST
Show Abstract + Hide Abstract
The detection and characterization of glomeruli are key elements in diagnostic and experimental nephropathology. Although the field of machine vision has already advanced the detection, classification, and prognostication of diseases in the specialties of radiology and oncology, renal pathology is just entering the digital imaging era. However, developing quantitative machine learning approaches (e.g., self-supervised deep learning) that characterize glomerular lesions (e.g., global glomerulosclerosis (GGS)) from whole slide images (WSIs) typically requires large-scale heterogeneous images, which is resource extensive for individual labs. In this study, we assess the feasibility of leveraging fine-grained GGS characterization via large-scale web image mining (e.g., from journals, search engines, websites) and self-supervised
Session 7: Computer-Aided Diagnosis, Prognosis, and Predictive Analysis
21 February 2022 • 1:30 PM - 3:50 PM PST
12039-29
Author(s): Samuel P. Border, Brandon Ginley, John E. Tomaszewski, Pinaki Sarder, Univ. at Buffalo (United States)
21 February 2022 • 1:30 PM - 1:50 PM PST
Show Abstract + Hide Abstract
We present HistoLens, a GUI developed to allow for the visualization of a large number of quantitative features within the renal glomerulus relating to specific biological compartments. By isolating the regions that directly contribute to the values of handcrafted features, we can present these features in a visual context that is more accessible to non-computational experts. Packaged in a user-friendly GUI, HistoLens enables users to establish an objective understanding of their data and develop quantitative biological hypotheses.
12039-30
Author(s): Salma Dammak, Matthew J. Cecchini, Aaron D. Ward, Western Univ. (Canada)
21 February 2022 • 1:50 PM - 2:10 PM PST
Show Abstract + Hide Abstract
A high tumor mutational burden is a promising biomarker for identifying lung cancer patients who would benefit from risky but potentially highly beneficial immunotherapy treatment. However, the cost and time it takes to obtain it makes it difficult to implement in the clinic. In this study, we used a deep learning model to estimate this biomarker based on hematoxylin and eosin histology slides of squamous cell carcinoma tumours from 30 patients from 20 centers included in The Cancer Genome Atlas. On the validation set (n=7), the system had perfect classification.
12039-31
Author(s): Nicholas Lucarelli, Univ. at Buffalo (United States); Donghwan Yun, Dohyun Han, Seoul National Univ. (Korea, Republic of); Brandon Ginley, Univ. at Buffalo (United States); Kyung C. Moon, Seoul National Univ. (Korea, Republic of); Avi Rosenberg, Johns Hopkins Univ. (United States); John E. Tomaszewski, Univ. at Buffalo (United States); Seung S. Han, Seoul National Univ. (Korea, Republic of); Pinaki Sarder, Univ. at Buffalo (United States)
21 February 2022 • 2:10 PM - 2:30 PM PST
Show Abstract + Hide Abstract
Histological image data and molecular profiles provide context into renal condition. Using AI, we linked urinary proteomics with renal biopsies, to investigate new and existing molecular links to image phenotypes. Our dataset contains renal biopsies from n = 56 DN patients, with 2038 proteins measured from patients’ urine. Glomeruli and tubules were segmented from other renal compartments using our H-AI-L pipeline. Handcrafted image features were measured for both glomeruli and tubules, and were input to a fully connected neural network, to predict their patient level protein signature. Using the extracted network weights, we investigated AI-learned relationships between image phenotype and molecular profile.
12039-32
Author(s): Aya Aqeel, Case Western Reserve Univ. (United States), Cleveland State Univ. (United States); Germán Corredor, Vidya Sankar Viswanathan, Chuheng Chen, Pingfu Fu, Case Western Reserve Univ. (United States); Joseph Willis, Univ. Hospitals Cleveland Medical Ctr. (United States); Anant Madabhushi, Case Western Reserve Univ. (United States)
21 February 2022 • 2:30 PM - 2:50 PM PST
12039-33
Author(s): Jon J. Camp, Gregory Otteson, Jansen Seheult, Min Shi, Dragan Jevremovic, Horatiu Olteanu, Ahmad Nanaa, Aref Al-Kali, Mohamed Salama, David Holmes, Mayo Clinic (United States)
21 February 2022 • 2:50 PM - 3:10 PM PST
Show Abstract + Hide Abstract
Flow cytometry immunophenotyping of bone marrow cells is helpful in making the diagnosis of Myelodysplastic syndromes (MDS). Due to natural properties of the fluorescent dyes used in flow cytometry, raw digital data from the instrument must be compensated to account for the spillover of signal between fluorochromes to be interpreted by skilled technicians. A neural network was trained twice on a randomly selected 80% of 353,655,369 unique events (282,924,297 events), once on uncompensated data and again (from initialization) with the per-tube compensated data. Tested on the reserved data, the networks perform essentially identically.
12039-34
Author(s): Chang Hee Han, Sejong Univ. (Korea, Republic of); Jin Tae Kwak, Korea Univ. (Korea, Republic of)
21 February 2022 • 3:10 PM - 3:30 PM PST
Show Abstract + Hide Abstract
Perineural invasion refers to a process where tumor cells invade, surround, or pass through nerve cells, serving as an indicator of aggressive tumor and related to poor prognosis. Herein, we propose an efficient and effective hybrid computational method for an automated detection of perineural invasion junctions in multi-tissue digitized histology images. The proposed approach conducts the detection of perineural invasion junctions in three stages. The first state identifies candidate regions for perineural invasion. The second stage delineates perineural invasion junctions. The last stage eliminates any false positive regions for perineural invasion. In the first two stages, we exploit an advanced deep neural network. In the last stage, we utilize hand-crafted features and a conventional machine learning algorithm. To evaluate the proposed approach, we employ 150 whole slide images obtained from PAIP2021 Challenge: Perineural Invasion in Multiple Organ Cancer and conduct a five-fold c
12039-59
Author(s): Luoting Zhuang, Jana Lipkova, Richard Chen, Faisal Mahmood, Brigham and Women's Hospital (United States)
21 February 2022 • 3:30 PM - 3:50 PM PST
Show Abstract + Hide Abstract
Management of aggressive malignancies, such as glioma, is complicated by a lack of predictive biomarkers that could reliably stratify patients based on treatment outcome. The complex mechanisms driving glioma recurrence and treatment resistance cannot be fully understood without the integration of multiscale factors such as cellular morphology, tissue microenvironment, and macroscopic features of the tumor and the host tissue. We present a weakly-supervised, interpretable, multimodal deep learning-based model fusing histology, radiology, and genomics features for glioma survival predictions. The proposed framework demonstrates the feasibility of multimodal integration for improved survival prediction in glioma patients.
SPIE Medical Imaging Awards and Plenary Session + 50th Anniversary Panel
21 February 2022 • 4:00 PM - 6:00 PM PST | Town & Country A
Session Chairs: Metin N. Gurcan, Wake Forest Baptist Medical Ctr. (United States), Robert M. Nishikawa, Univ. of Pittsburgh (United States)
4:00 pm:
Symposium Chair Welcome and Best Student Paper Award Announcement
First-place winner and runner-up of the Robert F. Wagner All-Conference Best Student Paper Award

4:10 pm:
SPIE 2022 Presidents Welcome and New SPIE Fellows Acknowledgments

4:15 pm:
SPIE Harrison H. Barrett Award in Medical Imaging
Presented in recognition of outstanding accomplishments in medical imaging
12032-300
Author(s): Jennifer N. Avari Silva, Washington Univ. in St. Louis (United States)
21 February 2022 • 4:20 PM - 5:05 PM PST | Town & Country A
Show Abstract + Hide Abstract
With the increased availability of extended reality (XR) devices in the marketplace, there has been a rapid development of medical XR applications spanning from education, training, rehabilitation, pre-procedural planning, and intra-procedural use. We will explore various use case to understand the importance of technology-use case matches and focus on intra-procedural use cases which generally have the highest risk to patient and medical provider but may have the most sizable impact on benefit to patient and procedure.
12031-700
21 February 2022 • 5:10 PM - 6:00 PM PST | Town & Country A
Monday Poster Session
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
All symposium attendees are invited to attend the evening Monday Poster Session to view the high-quality posters and engage the authors in discussion. Attendees are required to wear their conference registration badges to access the Poster Session. Authors may set up their posters starting Sunday 20 February.*

*In order to be fully considered for a Poster Award, it is recommended to have your poster set up by 12:00pm on Sunday 20 February 2022. Posters should remain on display until the end of the Poster Session on Monday.
12039-37
Author(s): Tom R. L. Kimpe, Varun Vasudev, Albert Xthona, Barco N.V. (Belgium)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
The past few years digital pathology has been widely adopted. The display system is a crucial component in the overall digital pathology system, since pathologists decide based upon images visualized on the display. Quality of the display can influence clinical performance, but also workflow efficiency and ergonomics. Performance of radiology display systems has been extensively studied and this resulted into standardization and clear requirements and recommendations. Digital pathology images and viewing conditions are very different compared to radiology. Fewer effort has gone in understanding what makes a digital pathology display fit for use and there is no consensus yet in the digital pathology community about ideal specifications for digital pathology displays. This paper studies specific characteristics of digital pathology display systems, such as luminance, contrast and resolution. Effects of these characteristics on visibility of relevant pathological features is described,
12039-40
Author(s): Yuxuan Shi, Quan Liu, Jiachen Xu, Zuhayr Asad, Can Cui, Vanderbilt Univ. (United States); Hernan Correa, Yash Choksi, Girish Hiremath, Vanderbilt Univ. Medical Ctr. (United States); Yuankai Huo, Vanderbilt Univ. (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Eosinophilic esophagitis (EoE) is an immune-mediated, clinicopathologic disease of the esophagus. In this study, we propose to use transfer deep-learning via both the ImageNetpre-trained ResNet50 as well as the more recent Big Transfer (BiT) model to achieve automated EoE feature identification on whole slide images. Our study investigates five EoE-relevant histologic features including basal zone hyperplasia, dilated intercellular spaces, eosinophils, lamina propria fibrosis, and normal lamina propria simultaneously. From the results, the model achieved a promising testing balanced accuracy of 61.9%, which is better than that of its trained-from-scratch counterparts.
12039-41
Author(s): Yusuke Murayama, Tohru Sugiyama, Yoshihiko Ogino, Hiroki Furuta, Yoichi Kajimura, Dai Nippon Printing Co., Ltd. (Japan); Michiie Sakamoto, Department of Pathology, Keio University School of Medicine (Japan)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
In pathological image diagnosis, there is a problem that the color reproduction differs depending on the device. To solve this problem, we have developed a color chart and color correction tool. In this report, we will describe the correction accuracy of color difference when our method is applied to actual stained pathological specimens taken with multiple WSI. It was confirmed that our color correction reduces the color difference between the images taken by the two models of WSI and the color difference between the specimen itself and the captured image.
12039-42
Author(s): Shayan Monabbati, Case Western Reserve Univ. (United States); Paula Toro, Ctr. for Computational Imaging & Personalized Diagnostics (United States); Pingfu Fu, Case Western Reserve Univ. (United States); Sylvia A. Lou, Univ. Hospitals Cleveland Medical Ctr. (United States); Anant Madabhushi, Ctr. for Computational Imaging & Personalized Diagnostics (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
12039-44
Author(s): Antong Chen, Tosha Shah, Andrew Brown, Haleh Akrami, Albert Swiston, Amir Vajdi, Merck & Co., Inc. (United States); Radha Krishnan, Merck, Sharp & Dohme Ltd. (United Kingdom); Razvan Cristescu, Merck & Co., Inc. (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Tumor mutation burden (TMB) is an important biomarker for the prediction of response to anti-PD-1 immunotherapies. Studies have shown that higher level of TMB (TMB-H) is associated with higher response rate to immunotherapies in patients with various types of advanced solid tumors. In this work, we assess the feasibility of predicting TMB-H based upon hematoxylin and eosin (H&E)-stained histopathology images. Using an Inception-V3 CNN as a baseline feature extractor, we compare adding a multi-layer perceptron (MLP) and a squeeze-and-excitation (SE) network on top of the baseline CNN. Training from random initialization and tuning with pretrained weights are also compared. A 4-fold cross-validation on the H&E whole-slide images (WSI) of The Cancer Genome Atlas (TCGA).show that the highest average area under the receiver operating characteristic curve (AUC) is 0.589, which implies that the prediction of TMB based on H&E WSI for melanoma remains a challenging problem.
12039-45
Author(s): Matthew McNeil, Cem Anil, Anne Martel, Univ. of Toronto (Canada)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Digital pathology involves the digitization of high quality tissue biopsies on microscope slides to be used by physicians for patient diagnosis and prognosis. These slides have become exciting avenues for deep learning applications to improve care. Despite this, labels are difficult to produce and thus remain rare. In this work, we create a sparse capsule network with a spatial broadcast decoder to perform representation learning on segmented nuclei patches extracted from the BreastPathQ dataset. This was able to produce disentangled latent space for categories such as rotations, and logistic regression classifiers trained on the latent space performed well.
12039-46
Author(s): Farzad Fereidouni, Richard Levenson, Univ. of California, Davis (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Hematoxylin- and eosin-stained tissue sections are central to patient diagnosis and management guidance. Direct viewing of these slides using a microscope is slowly but seemingly inevitably giving way to the new world of digital pathology. Having the histology images in digital format facilitates deployment of digital analysis tools, ranging from basic image enhancement to straightforward assessment of quantifiable metrics (area, intensity, molecular marker intensity and distribution) all the way to applications of novel machine learning and artificial intelligence tools. However, to date, virtually all such H&E whole-slide scans have been accomplished only in brightfield mode. We have observed that simple fluorescence imaging of H&E-stained slides can provide a great deal of useful histology content. We have developed a scanner that can acquire pixel-matched brightfield and fluorescence images in area-scanning mode, a process we termed DUET, for DUal-mode Emission and Transmission, i
12039-47
Author(s): Pierpaolo Vendittelli, Esther M. M. Smeets, Geert Litjens, Radboud Univ. Medical Ctr. (Netherlands)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Imaging techniques such as CT, MRI and ultrasound typically aim to provide the initial diagnosis for pancreatic cancer, but histopathological assessment is still the gold standard for final confirmation of disease presence and prognosis. We propose a multi-task convolutional neural network for detection and segmentation of Pancreatic cancer. We validated our approach on a dataset of 29 patients (for a total of 58 slides) at different resolutions. The best single task segmentation network achieved a median Dice of 0.885 ± 0.122 IQR. Our multi-task network improved on that with a median Dice score of 0.934 ± 0.077 IQR.
12039-48
Author(s): Neil Kavthekar, Brandon Ginley, Samuel P. Border, Nicholas Lucarelli, Univ. at Buffalo (United States); Kuang-Yu Jen, Univ. of California, Davis (United States); Pinaki Sarder, Univ. at Buffalo (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Tubular atrophy is a powerful indicator of chronic kidney disease progression. Despite its unquestioned utility, there is significant inter-pathologist variability, limiting its impact. We developed an automated tool to segment tubules in whole kidney biopsy, and further highlight specific tubule segmentations based on their morphometric properties. Our results suggest this tool may be helpful to increase the level of quantitation applied in kidney biopsy reporting. Future work will focus on defining reference standards of ‘healthy/normal’ tubule morphometrics in a controlled reference population.
12039-50
Author(s): Mauro Gwerder, Amjad khan, Christina Neppl, Inti Zlobec, Institute of Pathology, Univ. Bern (Switzerland)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Computational tools in pathology become more and more widespread, however, development of such tools usually needs large amounts of data. Previous approaches omitted the need for pixel-wise annotations and instead rely on global slide labels. Furthermore, a smart selection of relevant tiles within whole slide images reduces the amount of data needed for training. Such technique is feasible for end-to-end learning. A weakly supervised learning algorithm was trained on 668 WSI from 203 squamous cell lung carcinoma patients. Systematic experiments were designed to explore shallower deep learning models. We evaluated our study on different numbers of representative tiles for each slide.
12039-51
Author(s): Brendon Lutnick, Univ. at Buffalo (United States); David Manthey, Kitware, Inc. (United States); Jan U. Becker, Univ. zu Köln (Germany); Jonathan E. Zuckerman, Univ. of California, Los Angeles (United States); Luis Rodrigues, Coimbra Univ. Hospital (Portugal); Kuang-Yu Jen, Univ. of California, Davis (United States); Pinaki Sarder, Univ. at Buffalo (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
It is commonly known that diverse datasets of WSIs are beneficial when training convolutional neural networks, however sharing medical data between institutions is often hindered by regulatory concerns. We have developed a cloud-based tool for federated WSI segmentation, allowing collaboration between institutions without the need to directly share data. We demonstrate this tool by segmenting IFTA from three institutions, and show that keeping the three datasets separate does not hinder segmentation performance.
12039-52
Author(s): Farzad Fereidouni, Taryn Morningstar, Alexander Borowsky, Richard Levenson, Univ. of California, Davis (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Anatomic pathology is still the gold standard for tissue-based disease diagnosis, is centered on interpretation of H&E-stained sections on glass slides, a process requires labor-intensive and time-consuming specimen processing. We propose a simple, clinically relevant solution, termed FIBI (fluorescence imitating brightfield imaging), for directly creating diagnostic-quality images from unsectioned, fresh or fixed tissue specimens. FIBI can generate full-color histology-grade images within minutes. We have demonstrated the validity of this method by collecting 50 tissue samples from various organs and pathologies and comparing the diagnosis obtained using FIBI images with those determined from adjacent, conventionally prepared H&E-stained slides.
12039-54
Author(s): Michael Brehler, Allison Lowman, Samuel Bobholz, Savannah Duenweg, Fitzgerald Kyereme, Medical College of Wisconsin (United States); Cassandra Naze, Marquette Univ. (United States); John Sherman, Kenneth Iczkowski, Peter S. LaViolette, Medical College of Wisconsin (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
Early detection of prostate cancer drastically improves survival. With digital pathology playing an increasingly important role, automatic prescreening can help facilitate decision making and allow for a faster, more consistent diagnosis. We demonstrate the use of a fine-tuned neural network to streamline the analysis of prostate histology samples for automated Gleason grading. In addition, we present a novel way to interpret network response in the form of prostate cancer probability maps. Overall accuracy tested on a left-out test dataset was 79%. Based on the area of automatically annotated Gleason patterns, two exemplary whole mounts were correctly diagnosed as Gleason 4+5.
12039-56
Author(s): Parmida Ghahremani, Arie Kaufman, Stony Brook Univ (United States)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
The development of deep convolutional networks requires access to large quantities of labeled images for training and evaluation. Since image annotation is a tedious task for biomedical experts, recruiting non-expert crowd workers is economical and efficient to provide a rich dataset of annotated images. We first present a CrowdSourcing framework that enables fast and efficient acquisition of nuclei-segmented masks from the crowd by providing manual and semi-automatic annotation methods. We then present CrowdDeep, a novel technique to improve the segmentation accuracy of deep learning models trained with expert annotation on H&E slides by efficiently hiring crowd-annotated data.
12039-58
Author(s): Kouther Noureddine, Paul Gallagher, Martial Guillaud, Calum MacAulay, BC Cancer Research Centre (Canada)
21 February 2022 • 6:00 PM - 7:30 PM PST | Golden State Hall
Show Abstract + Hide Abstract
The tumour microenvironment(TME) is a complex mixture containing epithelium, stroma and a diverse network of immune cells & the spatial organization of these immune cells within the TME reflects a crucial process in anti-tumor immunity. By combining a multiplexed immunohistochemistry technique enabling analysis of complex immune cell populations on a single slide with deep learning segmentation methods to segment individual cell nuclei in tissue sections with an accuracy comparable to human annotation. We can analyze the cell-cell interactions between immune and tumour cells, enhancing our ability to perform molecularly based single cell analysis of multiple cell types simultaneously within the tissue.
Conference Chair
Univ. at Buffalo (United States)
Conference Chair
The Univ. of Western Ontario (Canada)
Program Committee
Bilkent Univ. (Turkey)
Program Committee
Univ. of Michigan Health System (United States)
Program Committee
Univ. of Illinois at Urbana-Champaign (United States)
Program Committee
Ulf-Dietrich Braumann
Fraunhofer-Institut für Zelltherapie und Immunologie IZI (Germany)
Program Committee
The Univ. of Texas Health Science Ctr. at San Antonio (United States)
Program Committee
U.S. Food and Drug Administration (United States)
Program Committee
Penn State College of Medicine (United States)
Program Committee
U.S. Food and Drug Administration (United States)
Program Committee
NEC Labs. America, Inc. (United States)
Program Committee
Rutgers, The State Univ. of New Jersey (United States)
Program Committee
Alton B. Farris
Emory Univ. (United States)
Program Committee
The Univ. of Pennsylvania Health System (United States)
Program Committee
AstraZeneca Pharmaceuticals LP (United States)
Program Committee
Ryerson Univ. (Canada)
Program Committee
Barco N.V. (Belgium)
Program Committee
Emory Univ. School of Medicine (United States)
Program Committee
Univ. of California, Davis (United States)
Program Committee
Univ. de Caen Basse-Normandie (France)
Program Committee
Geert Litjens
Radboud Univ. Medical Ctr. (Netherlands)
Program Committee
Case Western Reserve Univ. (United States)
Program Committee
Derek R. Magee
Univ. of Leeds (United Kingdom)
Program Committee
Sunnybrook Research Institute (Canada)
Program Committee
The Univ. of New South Wales (Australia)
Program Committee
Inspirata, Inc. (United States)
Program Committee
IBM Research (United States)
Program Committee
Lawrence Berkeley National Lab. (United States)
Program Committee
The Univ. of Warwick (United Kingdom)
Program Committee
U.S. Food and Drug Administration (United States)
Program Committee
Univ. at Buffalo (United States)
Program Committee
Amazon Lab126 (United States)
Program Committee
Univ. of Leeds (United Kingdom)
Program Committee
Radboud Univ. Nijmegen Medical Ctr. (Netherlands)
Program Committee
Mitko Veta
Technische Univ. Eindhoven (Netherlands)
Program Committee
Sunnybrook Research Institute (Canada)
Program Committee
Rensselaer Polytechnic Institute (United States)
Additional Information

POST-DEADLINE ABSTRACT SUBMISSIONS CLOSED

  • Submissions accepted through 27-December
  • Notification of acceptance by 12-January

View Call for Papers PDF