CHI-ANNOUNCEMENTS Archives

ACM SIGCHI General Interest Announcements (Mailing List)

CHI-ANNOUNCEMENTS@LISTSERV.ACM.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
George Bebis <[log in to unmask]>
Reply To:
George Bebis <[log in to unmask]>
Date:
Wed, 1 Jun 2022 15:45:40 -0700
Content-Type:
multipart/mixed
Parts/Attachments:
text/plain (12 kB)

17th International Symposium on Visual Computing (ISVC'22)
October 3-5, 2022
San Diego, CA
https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.isvc.net%2F&amp;data=05%7C01%7Cisvc_cfp%40lists.unr.edu%7Cc92828a5341f43a15cb208da44207589%7C523b4bfc0ebd4c03b2b96f6a17fd31d8%7C0%7C1%7C637897203448729229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=y%2Fe0yiT5HdZ4qlh7PdLYZuPCBHa%2FveuBTZkuJfytqdE%3D&amp;reserved=0

We would like to bring to your attention three Special Tracks that are 
being organized this year at ISVC'22 (details are provided on the ISVC 
website under the PROGRAM tab but also attached below for your help). If 
you are interested in submitting a paper to a special track, just select 
the track of your interest from the list of topics provided during your 
paper submission. The deadline for submitting a paper to a special track
or the main symposium is on July 18, 2022 (11:59 PM PST)

Warm regards,
George Bebis
ISVC'22 Steering Committee Chair

*** Special Tracks ***

ST1: Biomedical Image Analysis Techniques for Cancer Detection, Diagnosis 
and Management

Multiple biomedical imaging modalities are used in cancer detection, 
diagnosis and management including X-ray (plain film and Computed 
Tomography (CT)), Ultrasound (US), Magnetic Resonance Imaging (MRI), 
Single-Photon Emission Computed Tomography (SPECT), Positron Emission 
Tomography (PET), Optical Imaging and Digital Pathology. These imaging 
modalities form an essential part of cancer clinical decision making and 
are able to furnish morphological, structural, metabolic and functional 
information. In particular, biomedical imaging has become an important 
element for early cancer detection, for determining the stage and precise 
localization of cancer lesions to aid in directing surgery and other 
cancer treatments, or to check if cancer has recurred.

This special track invites research contributions on innovative biomedical 
image analysis techniques for cancer screening, diagnosis and staging, 
guiding cancer treatments, determining if a treatment works, and 
monitoring for cancer recurrence. Of particular interest are research 
contributions employing modern computer vision techniques, powered by 
statistical and machine/deep learning models, addressing the above 
challenges.

The authors of all accepted papers in the special track will be invited to 
submit an extended version of their work for review and possible 
publication in a Special Issue of the Mathematical Biosciences and 
Engineering journal (published by the American Institute of Mathematical 
Sciences) with an expected submission deadline in the second quarter of 
2023.

Topics of interest include but are not limited to:

Biomedical image analysis (e.g., detection, segmentation, classification, 
registration)
Computer-aided detection/diagnosis of various cancer types in biomedical 
images
Multi-modality fusion (e.g., MRI/PET, PET/CT, X-ray/ultrasound, etc.) for 
diagnosis, image analysis and image guided interventions
Image reconstruction for biomedical imaging
Cellular image analysis (e.g., genotype, phenotype, classification, 
identification, cell tracking)
Molecular/pathologic image analysis (e.g., PET, digital pathology)
Statistical and machine/deep learning models for biomedical image analysis
Evaluating and interpreting machine/deep learning models
Designing and building interfaces between algorithms and clinicians


Organizers:
George Bebis, University of Nevada, Reno
Sokratis Makrogiannis, Delaware State University

---------------------------------------------

ST2: Neuro-inspired Artificial Intelligence

This special track will focus on research that integrates themes in 
neuroscience that have yet to be thoroughly explored in machine learning 
and artificial intelligence. The current state-of-the-art is dominated by 
deep learning; however, recent research has uncovered critical issues that 
limits its advancement. While deep learning has had incredible success, 
especially when used in narrow, supervised settings, deep learning needs 
huge labeled databases to be successful. But, new breakthroughs in 
intelligence will not simply come from using more labeled data. As noted 
by LeCun, Bengio, and Hinton, “… we expect unsupervised learning to become 
far more important in the longer term. Human and animal learning is 
largely unsupervised: we discover the structure of the world by observing 
it, not by being told the name of every object.”

Traditional supervised models are also susceptible to adversarial attacks 
and are easily fooled. For example, small perturbations in the pixel 
intensities of an image that are imperceptible to humans, can
easily alter the output to a target class. This can partially be 
attributed to the fact that most supervised learning is based upon 
discriminative learning algorithms. In other words, it models the decision 
boundary between classes instead of modeling the distribution of classes 
themselves. Current AI models are also not robust to out-of-distribution 
data (e.g., ImageNet-C, ImageNet-R, ObjectNet). The out-of-distribution 
datasets includes cases that naturally happen during model deployment. For 
example, ImageNet-C provides a standard perturbation benchmark that 
simulates 75 real-world corruption examples that illustrate the weaknesses 
of current models.

In general, neural networks have gradually moved away from biological 
thematics. This has largely been due to engineering breakthroughs and 
brute force tactics in the past several years that have transformed the 
field of machine learning. Further engineering of these networks is 
reaching a saturation point where incremental novelty in the number of 
layers, activation function, parameter tuning, gradient function, etc., is 
only producing incremental accuracy improvements. Although there is 
evidence that AI has reached human levels on certain narrowly defined 
tasks, for general applications, biological AI remains far superior to 
that of any computer. Evidence from neuroscience suggest algorithmic and 
architectural methodology that could drive exciting and new research 
directions. For example, 95% of synapses in cortex are not related to 
feed-forward bottom-up drive but rather reflect local inhibitory, 
long-range lateral and top-down feedback projections, pathways that are 
mostly ignored by deep learning architectures. Spike timing may be a 
critical aspect of biological information encoding that have been 
abstracted away from current ML frameworks. Given the independent advances 
in neuromorphic software and hardware, machine learning, and neuroscience, 
the fields are again well positioned for cross pollination.

Topics of interest include but are not limited to:

Neuromorphic Computing
Spiking Neural Networks
Self-supervised/ Unsupervised Learning
Learning with Less Labels
Robust Classification
Generative Machine Learning
Neuro-inspired AI
Biologically Plausible AI
Sparse Coding
Sparse Distributed Representations
Energy Efficient Machine Learning
Top Down Feedback in Machine Learning
Inhibitory and Excitatory Lateral/Feedback connections
Cognitive Neural Architectures
Non-von Neumann computing architectures and models
Event based systems

Organizers:
Edward Kim, Drexel University
Yijing Watkins, Pacific Northwest National Laboratory
Garrett T. Kenyon, Los Alamos National Laboratory

------------------------------------------------------------------------------

ST3: Machine Learning in Ophthalmology

The advent of deep learning, a sub-field in Artificial Intelligence (AI), 
has made a significant impact on many biomedical imaging applications from 
detection of malignant tissue in mammographs to determining calcium signal 
propagations in smooth muscle cells. Recently, deep learning models have 
attracted the attention of researchers and clinicians in the ophthalmic 
domain. This has led to the approval of the first AI system for automatic 
diagnosis of diabetic retinopathy (DR) by the food and drug administration 
(FDA).  In this special track, we plan to bring researchers and clinicians 
from the two fields of machine learning and ophthalmology to discuss the 
most recent advances in deep learning that has made significant impacts on 
the way ophthalmic data is visualized, interpreted, and analyzed for 
diagnosis of vision threatening diseases.

This special track aims at presenting work inspired and implemented by 
advances made in the field of computer vision (e.g. deep learning) to help 
with diagnosis diseases that affect human visual perception. Thus the 
research presented in this special track covers both aspects of human 
perception and machine perception in the visual domain.

Topics of interest include but are not limited to:

Automated Diagnosis of Ophthalmic Conditions
Anomaly Detection in Ophthalmic Images
2D/3D Segmentation of Ophthalmic Data
Generative Models for Ophthalmic Data Fusion, Analysis, and Interpretation
Self/Semi- Supervised Learning from Limited Ophthalmic Data

Organizers:
Alireza Tavakkoli, University of Nevada, Reno
Julia Owen, University of Washington


*** ISVC'22 Important Information ***

*** Important Dates ***

(All deadlines are at 11:59PM Pacific Daylight Time)

Paper submissions: July 18, 2022
Notification of acceptance: August 22, 2022
Final paper: September 12, 2022
Early Registration: September 12, 2022
Hotel Reservation September 12, 2022
ISVC Symposium: October 3-5, 2022

*** Keynote Speakers ***
Shechtman Eli, Adobe Research
Solomon Justin, MIT
Bowman Doug, Virginia Tech
Jacobs David, University of Maryland
Liu Ce, Microsoft Azure AI
North Chris, Virginia Tech

*** Computer Vision Chairs ***
Li Bo, University of Illinois Urbana-Champaign
Yao Angela, National University of Singapore

*** Computer Graphics Chairs ***
Liu Yang, Microsoft Research Asia
Duan Ye, University of Missouri

*** Visualization Chairs ***
Crisan Ana, Tableau Research
Chang Remco, Tufts University

*** Virtual Reality Chairs ***
Lau Manfred, City University of Hong Kong
Khadka Rajiv, Idaho National Laboratory

*** Steering Committee ***
Bebis George, Univ of Nevada, Reno (chair)
Coquillart Sabine, INRIA
Klosowski James, AT&T Labs Research
Kuno Yoshinori, Saitama University
Lin Steve, Microsoft Research
Lindstrom Peter, Lawrence Livermore Nat Lab
Moreland Kenneth, Oak Ridge National Laboratory
Nefian Ara, NASA Ames Research Center
Tafti Ahmad P., University of Southern Maine

-------------------------------------------------------------------------------
Dr. George Bebis               	                         (775) 784-6463 (office)
Foundation Professor 		                         (775) 784-1877 (FAX)
Dept of Computer Science & Engineering/171               [log in to unmask]
Director of Computer Vision Laboratory 	                 https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.cse.unr.edu%2FCVL&amp;data=05%7C01%7Cisvc_cfp%40lists.unr.edu%7Cc92828a5341f43a15cb208da44207589%7C523b4bfc0ebd4c03b2b96f6a17fd31d8%7C0%7C1%7C637897203448729229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=vZ38Tc06eG0GbAICUyzy4YNP%2BrFI2mpuE2yDgoHUw2M%3D&amp;reserved=0
Senior Adjunct Scientist, Renown Institute for Cancer
University of Nevada                                     https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.cse.unr.edu%2F~bebis%2F&amp;data=05%7C01%7Cisvc_cfp%40lists.unr.edu%7Cc92828a5341f43a15cb208da44207589%7C523b4bfc0ebd4c03b2b96f6a17fd31d8%7C0%7C1%7C637897203448729229%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=N9nzw%2BdC93MbRfP9203zQbP%2FCmjUYxs%2B378D38KmdmM%3D&amp;reserved=0
1664 N. Virginia Str.
Reno, NV 89557
------------------------------------------------------------------------------

    ----------------------------------------------------------------------------------------
    To unsubscribe from CHI-ANNOUNCEMENTS send an email to:
     mailto:[log in to unmask]

    To manage your SIGCHI Mailing lists or read our polices see:
     https://sigchi.org/operations/listserv/
    ----------------------------------------------------------------------------------------

ATOM RSS1 RSS2