Our journal paper entitled “A lung graph model for the radiological assessment of chronic thromboembolic pulmonary hypertension in CT” by O. Jimenez del Toro, Y. Dicente Cid, A. Platon, A-L. Hachulla Lemaire, F. Lador, P-A. Poletti and H. Müller has been accepted for publication in Computers in Biology and Medicine.
Our paper entitled entitled “Variability of Muscle Synergies in Hand Grasps: analysis of intra- and inter-session data”, by Una Pale, Manfredo Atzori, Henning Müller and Alessandro Scano has been accepted for publication in Sensors. The full paper is now available online: https://www.mdpi.com/1424-8220/20/15/4297.
Our paper entitled “Semi-Weakly Supervised Learning for Prostate Cancer Image Classification with Teacher-Student Deep Convolutional Networks“, by S. Otálora, N. Marini, M. Atzori and H. Müller has been accepted for publication at the MICCAI workshop LABELS 2020: Large-scale Annotation of Biomedical data and Expert Label Synthesis.
In this paper, we propose a teacher-student approach for training data-hungry models using images automatically annotated by a deep CNN model. Our results in the challenging task of automatic grading of prostate cancer images show that using this approach is significantly better than training the CNN with the small set of ground-truth annotations. It opens the possibility to use vast amounts of non-annotated data.
Congratulations to Matteo Cognolato who successfully defended his PhD thesis entitled “Multimodal data fusion to improve the control of myoelectric prosthetic hands“, supervised byProf. Dr. Roger Gassert, Prof. Dr. Henning Müller, and Dr. Manfredo Atzori.
Feature attribution techniques explain Convolution Neural Networks (CNNs) in terms of the input pixels. The abstraction to higher level impacting factors, however, can be difficult when only a pixel-based analysis is performed. In this paper, we generate explanations in terms of prognostic factors for breast cancer, such as nuclei pleomorphism.
The Image Biomarker Standardisation Initiative (IBSI) is proud to launch chapter 2, with the goal of standardizing computations of imaging filters in radiomics. You can find more details on our website.
We are happy to announce the official start of the HECKTOR challenge (at MICCAI 2020) with the release of the training data. It comprises 201 cases of Head and Neck tumors with manual annotations on PET-CT images for segmentation of primary Gross Tumor Volumes.
Congratulations to Dr. Nora Turoman who successfully defended her PhD thesis. Under the supervision of Paul Matusz (HES-SO/UNIL) and Micah M. Murray (UNIL), Nora studied the development of attentional control in real-world settings from childhood to adulthood.
Thanks to the EU project PROCESS, we’re now training convolutional neural networks on more than 7Tb of data. Our PhD student Mara Graziani answers questions and illustrates our software architecture in these videos.
Stay tuned for the code release in the coming months.
How do you interpret your machine learning models? Our third-year Ph.D. student Mara Graziani gave a virtual tutorial at IBM Research Zurich about applying interpretability techniques. The recording was kindly shared on the Youtube channel of our school.
This talk provides an overview of the main interpretability approaches and a glimpse of our future research directions.
Our paper on the “Effect of movement type on the classification of electromyography data for the control of dexterous prosthetic hands” by Manfredo Atzori, Elisa Rosanda, Giorgio Pajardi, Franco Bassetto and Henning Müller has been accepted for presentation at the IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob 2020). The conference is currently planned to take place 29 Nov. – 2 Dec. 2020, in New York City, NY, USA.
Our Bridging Grant 2019 application has been accepted, in collaboration with the Research Center for Molecular and Cellular Imaging in Tehran University of Medical Sciences. This work will aim at the analysis of low-grade diffuse gliomas in MRI images. A large cohort of patients with brain cancer will be imaged with modern MRI scanners and analyzed using state of the art medical image analysis methods.