About

Cancer is one of the leading causes of mortality worldwide. According to the World Health Organization (WHO), 70% of deaths occur in low- and middle-income countries. The goal of this workshop is to identify the potential of AI to overcome global disparities in access, diagnosis and treatment in cancer care.

ICLR 2020

The International Conference on Learning Representations (ICLR) is one of the leading machine learning conferences. The 2020 conference will be held in Addis Ababa, Ethiopia from April 26-30, with a focus on “AI for Social Good.” Please check the main conference website for information about registration, schedule, venue, and travel arrangements. More information in the Registration section below.

Due to COVID-19 concerns, ICLR 2020 will be held as a virtual conference. Full details here: https://iclr.cc/Conferences/2020/virtual

Workshop Schedule & Attendee Instructions

The workshop will be streamed live on Sunday, April 26 over the ICLR virtual website and Zoom webinar. Attendees will have an opportunity to directly engage with the speakers and ask questions during Q&A sessions over Zoom. Slides-live presentations can be accessed by clicking the "video" links below.

Title Speakers Format Date Start time (EST) End time (EST)
Introductory remarks Thomas Fuchs VIDEO 4/26/20 10:00 AM 10:15 AM
Addressing South Africa's Cancer Reporting Delay with Machine Learning (Invited talk) Waheeda Banu Saib VIDEO 4/26/20 10:15 AM 10:45 AM
Deep Learning in Digital Pathology (Invited talk) Amit Sethi VIDEO 4/26/20 10:45 AM 11:15 AM
AI for cancer diagnosis in resource deprived areas (Invited talk) Yuchun Ding VIDEO 4/26/20 11:15 AM 11:45 AM
Q&A Session I
Introductory remarks Thomas Fuchs (Moderator) Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Invited talk Waheeda Banu Saib Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Panel Olusegun Isaac Alatise Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Invited talk Oluwatoyin P. Popoola Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Invited talk Amit Sethi Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Invited talk Yuchun Ding Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Closing remarks Johan Lundin Zoom Webinar 4/26/20 12:00 PM 1:00 PM
Afternoon session
Panel Olusegun Isaac Alatise Zoom Presentation 4/26/20 1:00 PM 1:15 PM
Cancer in low- and middle- income countries: MSK's Global Cancer Disiparities Initiatives (Panel) T. Peter Kingham (Moderator) VIDEO 4/26/20 1:15 PM 1:30 PM
Challenges in Pathology Diagnosis in Low to Middle Income Countries (LMICs) (Panel) Marcia Edelweiss VIDEO 4/26/20 1:30 PM 1:45 PM
Panel Peter Ntiamoah VIDEO 4/26/20 1:45 PM 2:00 PM
Making Decisions with (Limited) Data: Off-Policy RL, Model-Based Optimization, and Meta-Learning (Keynote) Sergey Levine VIDEO                    Q&A at 5 PM 4/26/20 2:00 PM 2:45 PM
A Multimodal Imaging System for Cervical Pre-Cancer and Cancer Detection (Invited talk) David Brenes VIDEO 4/26/20 3:00 PM 3:30 PM
Invited talk Oluwatoyin P. Popoola VIDEO 4/26/20 3:30 PM 4:00 PM
Poster Blitz
Computer Aided Diagnosis System for Classification of Abnormalities in Thyroid Nodules Ultrasound Images using Deep Learning Oluwadare A Adebisi VIDEO 4/26/20 4:00 PM 4:05 PM
Detection of western blot protein bands using transfer learning in the assessment of risk of gastric and oesophageal cancer Girmaw Abebe Tadesse, Daniel Chapman, Ling Yang, Pang Yao, Iona Millwood, Zhengming Chen, Tingting Zhu, Girmaw Abebe Tadesse VIDEO 4/26/20 4:05 PM 4:10 PM
A Study of Deep Learning Colon Cancer Detection in Limited Data Access Scenarios Apostolia Tsirikoglou, Karin Stacke, Gabriel Eilertsen, Martin Lindvall, Jonas Unger VIDEO 4/26/20 4:10 PM 4:15 PM
Multi-Task Learning in Histo-pathology for Widely Generalizable Model Jevgenij Gamper, Navid Alemi Koohbanani VIDEO 4/26/20 4:15 PM 4:20 PM
Deep Dictionary Learning for Colorectal Cancer Grading Nima Hatami, Mohsin Bilal, Nasir Rajpoot VIDEO 4/26/20 4:20 PM 4:25 PM
Artificial Intelligence and Telemedicine for Dermatological Care Marc Combalia, Sebastián Podlipnik, Susana Puig, Josep Malvehy VIDEO 4/26/20 4:25 PM 4:30 PM
CNN-based approach for cervical cancer in whole-slide histopathology images Ferdaous Idlahcen VIDEO 4/26/20 4:30 PM 4:35 PM
Addressing Ancestry Disparities in Genomic Medicine: A Geographic-aware Algorithm Daniel Mas Montserrat, Arvind Kumar, Carlos Bustamante, Alex Ioannidis VIDEO 4/26/20 4:35 PM 4:40 PM
Closing remarks Johan Lundin VIDEO 4/26/20 4:40 PM 5:00 PM
Q & A Session II
Panel T. Peter Kingham (Moderator) Zoom Webinar 4/26/20 5:00 PM 6:00 PM
Keynote Sergey Levine Zoom Webinar 4/26/20 5:00 PM 6:00 PM
Panel Marcia Edelweiss Zoom Webinar 4/26/20 5:00 PM 6:00 PM
Panel Peter Ntiamoah Zoom Webinar 4/26/20 5:00 PM 6:00 PM
Panel Olusegun Isaac Alatise Zoom Webinar 4/26/20 5:00 PM 6:00 PM
Introductory remarks Thomas Fuchs Zoom Webinar 4/26/20 5:00 PM 6:00 PM
Invited talk David Brenes Zoom Webinar 4/26/20 5:00 PM 6:00 PM

Call for extended abstracts

According to the World Health Organization (WHO), cancer is the second leading cause of death globally and was responsible for an estimated 9.6 million deaths in 2018. Significantly, 70% of cancer-related deaths occur in low- and middle-income countries (LMIC), highlighting enormous disparities in healthcare standards across the world. Overcoming these disparities requires confronting complex historical, cultural and political issues. However, the international scientific community can help in a meaningful way by developing and implementing technological and sociological solutions focused on improving patient outcomes in LMIC.

The purpose of this workshop is to bring together experts working in clinical cancer care and machine learning, and to facilitate discussions which include (1) an assessment of challenges in clinical global cancer care, (2) highlight current AI solutions to these problems, and (3) facilitate discourse between clinicians and machine learning experts to determine how access to these technologies can be expanded in a way that is both realistic and impactful.

We have identified three main areas of the cancer care work-flow that could greatly benefit from the application of machine learning technologies:

  1. AI for Accessibility: Due to the lack of public awareness regarding most common cancer symptoms and the absence of regular pre-cancer screening programs, patients frequently present late-stage diseases which are difficult to treat effectively. Technologies which guide potential cancer patients to a doctor faster are needed.
  2. AI for Diagnosis: Cancer diagnosis is based on microscopic examination of tissue by a pathologist. Yet pathologists are scarce, face high workloads, and due to lack of specialized training and poor infrastructure conditions, are less accurate and have slow turnaround times. Decision support systems for making diagnosis more accurate and efficient are necessary.
  3. AI for Treatment Planning: In LMIC, the number of medical professionals with cancer care expertise is limited. In addition, many treatment options are unavailable. For these reasons, the development of recommendation systems is imperative to facilitate timely and effective treatment.

We strongly encourage any submissions for machine learning methods relevant to cancer care. These include, but are not limited to:

  • AI-Agents for patient engagement
  • NLP for rare and indigenous languages
  • Optimization for automated scheduling
  • Computational pathology for diagnosis
  • Liquid based diagnosis
  • Radiographic diagnosis
  • Mobile Microscopy conferencing and diagnostic assessment
  • AI for portable medical devices
  • Longitudinal modeling
  • Patient trial selection prediction
  • Telemedicine/consultation solutions
  • Patient privacy
  • Active learning
  • Robust models on noisy labels
  • Transfer learning
  • Weakly supervised learning
  • Learning methods for multi-modal datasets
  • ML methods for genomic and proteomic data

Accepted submissions will be presented as posters during the workshop.

Submission instructions

Submissions should be anonymized extended abstracts up to 2 pages in PDF format, typeset using the ICLR style. References do not count towards the page limit. You can make submissions through AI4CC submission site.

Travel grant support

Limited funding will be available to participants based on need: Travel grant application .

Code of Conduct

We expect all workshop participants to adhere to ICLR code of conduct.

Important dates

  • Call for Abstracts Open: December 11, 2019
  • Submission deadline: February 23, 2020 4:59 AM GMT
  • Notification of Acceptance: February 24, 2020
  • Travel grant application deadline: March 2, 2020
  • Workshop: April 26, 2020

Registration

ICLR registration here

List of Accepted Papers

Paper Title Authors
Addressing Ancestry Disparities in Genomic Medicine: A Geographic-aware Algorithm Daniel Mas Montserrat (Purdue University); Arvind Kumar (Stanford University); Carlos Bustamante (Stanford University); Alex Ioannidis (Stanford University)
A government call to cancer diagnosis with A.I. Emile R Engelbrecht (Stellenbosch University); Johan du Preez (Stellenbosch University)
Detection of western blot protein bands using transfer learning in the assessment of risk of gastric and oesophageal cancer Girmaw Abebe Tadesse (IBM); Daniel Chapman (University of Oxford); Ling Yang (University of Oxford); Pang Yao (University of Oxford); Iona Millwood (University of Oxford); Zhengming Chen (University of Oxford); Tingting Zhu (University of Oxford); Girmaw Abebe Tadesse (IBM Research)
Multi-View CNN for Automated Grading of Glioma Using Multi-Modal MRI Scans Abdela A Mossa (Cukurova Universiy); Ulus Cevik (Cukurova Universiy)
A Study of Deep Learning Colon Cancer Detection in Limited Data Access Scenarios Apostolia Tsirikoglou (Linköping University ); Karin Stacke (Linköping University); Gabriel Eilertsen (Linköping University); Martin Lindvall (Linköping University); Jonas Unger (Linköpings universitet)
Multi-Task Learning in Histo-pathology for Widely Generalizable Model Jevgenij Gamper (Cervest Ltd.); Navid Alemi Koohbanani (University of Warwick)
Deep Dictionary Learning for Colorectal Cancer Grading Nima Hatami (University of Warwick); Mohsin Bilal (University of Warwick); Nasir Rajpoot (University of Warwick)
Artificial Intelligence and Telemedicine for Dermatological Care Marc Combalia (Hospital Clínic de Barcelona); Sebastián Podlipnik (Hospital Clínic Barcelona); Susana Puig (Hospital Clínic de Barcelona); Josep Malvehy (Hospital Clínic de Barcelona)
CNN-based approach for cervical cancer in whole-slide histopathology images Ferdaous Idlahcen (LIMIARF)
AI healthcare bridge between disposable income & treatment cost in cancer treatment Isacko Sharamo Tuye (Iteconsultancy Ltd)
Computer Aided Diagnosis System for Classification of Abnormalities in Thyroid Nodules Ultrasound Images using Deep Learning Oluwadare A Adebisi (Polytechnic Ibadan)

Speakers

Keynote speaker

  • Sergey Levine
    UC Berkeley

    Talk title: Meta Learning for limited datasets

Disparities in Cancer Care Panel

Invited talks

  • David Brenes
    Rice University

  • Michael G. Kawooya
    ECUREI

    Talk title: Challenges and Opportunities for AI in Radiology
  • Waheeda Banu Saib
    IBM Research

    Talk title: Addressing South Africa's cancer reporting delay using machine learning
  • Amit Sethi
    Indian Institute of Technology Bombay

    Talk title: Deep Learning in Digital Pathology

Organizers