About DEF-AI-MIA

In the past few years, Deep Learning techniques have made rapid advances in many medical image analysis tasks. In pathology and radiology applications, they managed to increase the accuracy and precision of medical image assessment, which is often considered subjective and not optimally reproducible. This is due to the fact that they can extract more clinically relevant information from medical images than what is possible in current routine clinical practice by human assessors. Nevertheless, considerable development and validation work lies ahead before AI-based methods can be fully integrated ad used in routine clinical tasks.

Of major importance is research on domain adaptation, fairness and explainability in AI-enabled medical image analysis. This research constitutes the main target of the DEF-AI-MIA Workshop. The workshop aims to foster discussion and presentation of ideas to tackle these challenges in the field, as well as identify research opportunities in this context.

The DEF-AI-MIA Workshop is the fourth in the AI-MIA series of Workshops held at IEEE ICASSP 2023, ECCV 2022 and ICCV 2021 Conferences.

For any requests or enquiries regarding the Workshop, please contact: stefanos@cs.ntua.gr.

Organisers

General Chairs

Dimitris N. Metaxas

Rutgers University USA dnm@cs.rutgers.edu

Stefanos Kollias

National Technical University of Athens, Greece stefanos@cs.ntua.gr

Program Chairs

Dimitrios Kollias

Queen Mary University of London, UK d.kollias@qmul.ac.uk

Xujiong Ye

University of Lincoln, UK xye@lincoln.ac.uk

Francesco Rundo

STMicroelectronics srl, Italy francesco.rundo@st.com

The Workshop

Call for Papers

Original high-quality contributions (in terms of databases, surveys, studies, foundation models, techniques and methodologies) are solicited on -but are not limited to- the following topics and technologies:

Explainable 2-D & 3D-CNN, RNN, CNN-RNN, Transformers, Foundation models, unimodal/multimodal Large Language Models (LLMs), unsupervised, self-supervised Machine Learning (ML) models for medical diagnosis

Sensing “salient features” of AI/ML models related to decision-making, in spatial (images), temporal (video), volumetric (3-D) data

Optimal visualization of salient features and areas in the input data

Low/Middle/High level feature extraction & analysis for model interpretatability and explainability

In temporal and 3-D data: explanation of which features and at what time, or slice, or respective intervals, are the most prominent for the provided decision

In multimodal data: explainable data correlations for predictions in data streams

Joint optimization of positive and negative saliencies

Global and local models for prediction or classification

Attention and self-attention mechanisms in DL/AI approaches

Interpretability at training time through adversarial regularization

Learning new data (from multiple sources) by leveraging knowledge already extracted and codified, through domain adaptation

Generalizable ML/DL methods when the training medical image datasets are small

Generalizable ML/DL methods in cases of images with potential domain shift

Unsupervised, weakly supervised and semi-supervised model adaptation

Uncertainty estimation and quantification, self-training

Adaptation and prompt engineering in Foundation Models (e.g., LLMs) for explainable decisions and prediction

Algorithmic fairness

Zero- and one- shot learning, avoidance of catastrophic forgetting

Workshop Important Dates

Paper Submission Deadline (UPDATED): 23:59:59 AoE (Anywhere on Earth) March 29, 2024

Review decisions sent to authors; Notification of acceptance: April 10, 2024

Camera ready version: April 14, 2024

Submission Information

The paper format should adhere to the paper submission guidelines for main CVPR 2024 proceedings style. Please have a look at the Submission Guidelines Section here.

We welcome full long paper submissions (between 4 and 8 pages, excluding references or supplementary materials). All submissions must be anonymous and conform to the CVPR 2024 standards for double-blind review.

All papers should be submitted using this CMT website.

All accepted manuscripts will be part of CVPR 2024 conference proceedings.

The Workshop's Agenda

Keynote Speaker: Dr. Ronald M. Summers

Biography

Dr. Summers is a pioneer in the use of artificial intelligence in radiology. His lab has made seminal contributions to the advancement of cancer diagnosis using radiology. Dr. Summers received the BA degree in physics and the MD and PhD degrees in Medicine/Anatomy and Cell Biology from the University of Pennsylvania. He completed a medical internship at the Presbyterian-University of Pennsylvania Hospital, Philadelphia, PA, a radiology residency at the University of Michigan, Ann Arbor, MI, and an MRI fellowship at Duke University, Durham, NC. In 1994, he joined the Radiology and Imaging Sciences Department at the NIH Clinical Center in Bethesda, MD., where he is now a tenured Senior Investigator and Staff Radiologist. He is a Fellow of the Society of Abdominal Radiologists and of the American Institute for Medical and Biological Engineering (AIMBE). He directs the Imaging Biomarkers and Computer-Aided Diagnosis (CAD) Laboratory and is former and founding Chief of the NIH Clinical Image Processing Service. In 2000, he received the Presidential Early Career Award for Scientists and Engineers, presented by Dr. Neal Lane, President Clinton's science advisor. In 2012, he received the NIH Director's Award, presented by NIH Director Dr. Francis Collins. In 2017, he received the NIH Clinical Center Director's Award. He has co-authored over 500 journal, review and conference proceedings articles and is a co-inventor on 14 patents. He is a member of the editorial boards of the Journal of Medical Imaging, Radiology: Artificial Intelligence and Academic Radiology and a past member of the editorial board of Radiology. He is a program committee member of the Computer-aided Diagnosis section of the annual SPIE Medical Imaging conference and was co-chair of the entire conference in 2018 and 2019. He was Program Co-Chair of the 2018 IEEE ISBI symposium.

Presentation Slides

The presentation slides of Dr. Summers can be found here.

Keynote Speaker: Prof. Greg Slabaugh

Biography

Greg is Professor of Computer Vision and AI and Director of the Digital Environment Research Institute (DERI) at Queen Mary. His primary research interests include computer vision and deep learning, with applications to computational photography and medical image computing. Prior to joining Queen Mary University of London, he was Chief Scientist in Computer Vision (EU) for Huawei Technologies R&D where he led a team of research scientists working in computational photography, studying the camera image signal processor (ISP) pipeline including denoising, demosaicing, automatic white balance, super-resolution, and colour enhancement for high quality photographs and video. Earlier industrial appointments include Medicsight, where he led a team of research scientists in detection of pre-cancerous lesions in the colon and lung, imaged with computed tomography; with the company's ColonCAD product receiving FDA clearance and CE marking. He also was an employee of Siemens, where he performed research in medical image computing and 3D shape modelling. He holds 36 granted patents and has roughly 200 publications. He earned a PhD in Electrical Engineering from Georgia Institute of Technology in Atlanta, USA where his thesis focused on reconstruction of 3D shapes from 2D photographs. For six years he was an academic at City, University of London where he taught modules in computer vision, graphics, computer games technology, and programming in addition to leading research grants funded by the European Commission, EPSRC and Innovate UK. He was awarded a university-wide Research Student Supervision Award in 2017, and a Teaching in the Schools award for the School of Mathematics, Computer Science, and Engineering in 2016.

The Competition

The 4th COV19D Competition is the 4th in the series of COV19D Competitions following the first 3 Competitions held in the framework of ICCV 2021, ECCV 2022 and ICASSP 2023 Conferences respectively. It includes two Challenges: i) Covid-19 Detection Challenge and ii) Covid-19 Domain Adaptation Challenge. More details can be found on this website.