Content-Based Image Retrieval
Our abstract entitled ‘Revealing the most suitable CT radiomics features for retrospective studies with heterogeneous datasets’ by Jimenez-del-Toro et al. has been accepted to be presented at the ONLINE European Congress of Radiology 2021, for the session ‘AI in abdominal imaging’, held from March 3-7, 2021.
Our paper on “The development of attentional control mechanisms in multisensory environments” has been accepted for publication in Developmental Cognitive Neuroscience, Elsevier.
This work led by Nora Turoman, Pawel J. Matusz and their colleagues focused on using advanced multivariate EEG analyses to understand how children develop their attentional control in real-world like settings and when different processes that underlie this attention control reach the adult-like state.
Deep learning models may hide inherent risks: codification of biases, weak accountability and little transparency of the decision-making.
Our workshop at AMLD2021 (Building Interpretable AI for Digital Pathology) offers an all-round discussion on building interpretable AI for digital pathology.
Join us to hear about visualization methods, concept attribution and interpretable graph-networks.
Our team, led by Prof. Henning Müller, started a collaboration with the American College of Radiology to develop AI tools that help radiologists in their medical decision making. Le Nouvelliste wrote an article about this collaboration (in French).
We hosted Iam Palatnik (Pontifical Catholic University of Rio de Janeiro, Brazil) at MedGIFT for an invited talk on interpretability in medical imaging. Iam shared challenges and solutions for the Local Interpretable Model Agnostic Explanations (LIME) method. The video of his presentation is available online.
Abstract: An application of Local Interpretable Model Agnostic Explanations (LIME) is described for two case studies: Metastases and Malaria classification. Some of the key challenges of using LIME for this purpose – most notably the instability of explanations – are discussed, as well as some potential solutions. Namely, a genetic algorithm based solution called EvEx, where explanations are evolved as the average of a Pareto Front, and Squaregrid, a parameterless rough approximation. The results seem to show that EvEx finds more consistent explanations for regions of high explanatory weight, and that Squaregrid could be a viable way to diminish the need for segmentation parameter fine tuning.
The Springer LNCS proceedings of our HEad and neCK TumOR segmentation (HECKTOR) challenge are now available online: https://www.springer.com/gp/book/9783030671938
These proceedings feature the participants’ papers and our overview paper (summarizing the algorithms, results and outcomes of the challenge).
Our paper on “Breast Histopathology with High-Performance Computing and Deep Learning” (M. Graziani et. al) has been accepted for publication in Computing and Informatics, in the special issue on Providing Computing Solutions for Exascale Challenges.
In this work, we present our modular pipeline for detecting tumorous regions in digital specimens of breast lymph nodes with deep learning models. We evaluate challenges and benefits of training models on high-performance and cloud computing with millions of images.
Our PhD student Mara Graziani discussed the topic of defining machine learning interpretability at CIBM (Center for Biomedical Imaging, Switzerland). She also presented our latest applications to the histopathology domain. In particular, she covered our recent work on the “Evaluation and Comparison of CNN Visual Explanations for Histopathology”. She then explained how interpretability can be used in a proactive way to improve model performance. You can watch her talk online at this link: https://www.youtube.com/watch?v=7hs21U-3hgk&feature=youtu.be
Below, the information about the presentation:
Title: A myth-busting attempt for DL interpretability: discussing taxonomies, methodologies and applications to medical imaging.
Abstract:
Deep Learning (DL) models report almost perfect accuracy on some medical tasks, though this seems to plunge in real-world practices [1]. Started in 2009 as the generation of deep visualizations [2, 3], the field of interpretability has grown and developed over the years, with the intent of understanding why such failures happen and discovering hidden and erroneous behaviors. Several interpretability techniques were then developed to address the fundamentally incomplete problem of evaluating DL models on the sole task performance [4].
While defining the key terms used in the field, I will try to bust some myths on DL interpretability: are explainable and interpretable the same thing? Is a “transparent” model an “interpretable” model? Besides, within the application in the field of medical imaging, I will describe the risk of confirmation bias and present our work on evaluating the reliability of interpretability methods. Finally, I will bring examples from our previous works on how interpretable AI can be used to improve model performance and reduce the distance between the clinical and the DL ontologies.
[1] Yune, S., et al. “Real-world performance of deep-learning-based automated detection system for intracranial hemorrhage.” 2018 SIIM Conference on Machine Intelligence in Medical Imaging: San Francisco (2018).
[2] Erhan, D., et al. “Visualizing Higher-Layer Features of a Deep Network.” (2009).
[3] Zeiler, Matthew D., and Rob Fergus. “Visualizing and understanding convolutional networks.” European conference on computer vision. Springer, Cham, 2014.
[4] Doshi-Velez, Finale, and Been Kim. “Towards A Rigorous Science of Interpretable Machine Learning.” stat 1050 (2017): 2.
Two papers have been accepted for presentation at the 43rd European Conference on Information Retrieval (ECIR 2021), that will take place online online March 28 – April 1, 2021.
The first paper: “The 2021 ImageCLEF Benchmark: Multimedia Retrieval in Medical, Nature, Internet and Social Media Applications“, Bogdan Ionesco et. al, presents an overview of the upcoming ImageCLEF 2021 campaign and is supported by the AI4media project.
The second paper: “LifeCLEF 2021 teaser: Biodiversity Identification and Prediction Challenges“, Hervé Goëau et. al, presents the different tasks of the LifeCLEF 2021 conference including PlantCLEF, BirdCLEF and GeoLifeCLEF.
The “Germaine de Staël” program promotes collaboration between French and Swiss researchers and research teams. Several exchanges between our group (Prof. Henning Müller) and Sorbonne university of histopathology image analysis (Nicolas Lomenie and Camille Kurtz) will be funded by this project to work on deep learning in digital pathology with gigapixel images of hepatic tissue: “BioGigaDeep -Apprentissage profond en pathologie digitale pour l’analyse d’images gigapixels de tissus hépatiques”.
Our paper entitled “InvNet: A Deep Learning Approach to Invert Complex Deformation Fields”, by Marek Wodzinski and Henning Müller, has been accepted for presentation at ISBI 2021, the IEEE International Symposium on Biomedical Imaging to be held virtually on April 13-16, 2021.
Prof. Henning Müller will give a keynote presentation on “Multimodal Medical Data Analysis: Machine Learning in Histopathology” at the MMDLCA workshop at the International Conference on Pattern Recognition (ICPR) 2021, on January 11, 12:00 CET.
The full program is available online: https://medical-and-multimedia-lab.github.io/MMDLCA2020/
Congratulations to Marek Wodzinski and Henning Müller for their 2nd Best Paper Award at the International Workshop on Biomedical Image Registration WBIR 2020 with their paper on “Learning-based affine registration of histological images“.
Our paper on the “Evaluation and Comparison of CNN Visual Explanations for Histopathology” by Mara Graziani, Thomas Lompech, Henning Müller and Vincent Andrearczyk has been accepted for presentation at the AAAI-21 Workshop on Explainable Agency in Artificial Intelligence.
In this work, we evaluate the alignment of XAI visualisations to cancer diagnostic procedures for breast tissue. Do the visualizations highlight specific nuclei types? Visual explanations may induce confirmation bias about CNN decisions.
Prof. Henning Müller will give an invited talk on “Machine learning on multimodal histopathology data” at the European Conference on Natural Language Processing and Information Retrieval (ECNLPIR 2021), Stockholm, Sweden, Aug. 13-15, 2021.
More info here: http://www.ecnlpir.org/kn2021
Our work on “The importance of feature aggregation in radiomics: a head and neck cancer study“, conducted by Pierre Fontaine, Oscar Acosta, Joël Castelli, Renaud De Crevoisier, Henning Müller and Adrien Depeursinge has been published in Scientific Reports.
We show that visual feature aggregation, often overlooked in radiomics studies, plays an important role in the prediction of disease characteristics.
We are looking forward to presenting our work at the AIDP2021 workshop at ICPR 2021. The following two papers were accepted for presentation.
Our paper “DeepHistReg: Unsupervised Deep Learning Registration Framework for Differently Stained Histology Samples“, by Marek Wodzinski and Henning Müller has been published in Computer Methods & Programs in Biomedicine.
The project IMAGINE (Radiomics for comprehensive patient and disease phenotyping in personalized health), a Swiss-wide initiative to promote image-based personalized medicine, had a dedicated Swiss Personalized Health Network (SPHN) webinar on Oct. 7 2020 to detail the recent progress of the consortium. You can view it under this link:
https://sphn.ch/seminar-training/imagine/
Congratulations to Sebastian Otálora, Niccolò Marini, Henning Müller and Manfredo Atzori who received the “People’s choice award” for their paper “Semi-weakly Supervised Learning for Prostate Cancer Image Classification with Teacher-Student Deep Convolutional Networks” at LABELS MICCAI 2020.
