Veranstaltungen
zum Themenbereich Medizinische Bild- und Signalverarbeitung
Virtuelles AG Meeting im September
Liebe Mitglieder und Interessierte der GMDS AG Medizinische Bild- und Signalverarbeitung,
nachdem leider unser geplantes Treffen in Berlin aufgrund der corona-bedingten kurzfristigen Absage der Präsenzveranstaltung ausfallen musste, möchten wir Sie sehr herzlich zu unserem September-Meeting einladen. Da die Corona-Pandemie weiterhin Präsenztreffen schwierig machen und auch die GMDS Jahrestagung virtuell abgehalten wird, möchten wir auch das AG Meeting virtuell als Zoom-Meeting abhalten. Im Rahmen dieses Meetings wird auch die AG Leitung turnusgemäß neu gewählt.
Die Daten:
Mittwoch, 30.9.2020
ab 17 Uhr
Einwahllink:
https://oth-regensburg.zoom.us/j/94573810172?pwd=NWEyREx0RkQxUmFTODBhN3R6QUd6UT09
Meeting-ID: 945 7381 0172
Kenncode: 087176
Die Agenda:
- Begrüßung und evtl. Vorstellung der Teilnehmer
- Bericht der AG Leitung zu aktuellen Aktivitäten der Arbeitsgruppe, insbes. Unterstützung der BVM 2021 in Regensburg
-> Vorstellung der Planungen der BVM 2021 unter Corona-Bedingungen - Wahlen des AG Leiters bzw. der Leiterin und des stellvertr. AG Leiters bzw. der stellv. AG Leiterin
- Diskussion über künftige Aktivitäten der Arbeitsgruppe
-> Künftige Organisation des BVM Newsletters
-> Vorstellung der AG als Video auf dem GMDS-Youtube-Kanal - Sonstiges
Für die Wahl der AG Leiter*in gibt es bisher folgende Kandidat*innen:
- Prof. Dr. Dagmar Krefting, Universität Göttingen (Leiterin)
- Prof. Dr. Dennis Säring, FH Wedel (Stellv. Leiter)
Weitere Kandidaturen sind sehr willkommen und können vorab an uns weitergeleitet oder auch spontan in der Sitzung selbst geäußert werden.
Wahlberechtigt sind nur Mitglieder der GMDS.
Ich würde mich sehr freuen, wenn Sie zahlreich und aktiv am ersten virtuellen AG Meeting teilnehmen.
Beste Grüße,
Christoph Palm (AG Leiter)
Dennis Säring (Stellv. AG Leiter)
Learn2Reg MICCAI 2020 Registration Challenge 8th October
https://learn2reg.grand-challenge.org
Medical image registration plays a very important role in improving clinical workflows, computer-assisted interventions and diagnosis
Deep learning for medical registration is currently starting to show promising advances that could improve the robustness, computation speed
and accuracy of conventional algorithms.
The Learn2reg challenge provides a simplified challenge design that removes many of the common pitfalls for learning and applying transformations.
We will provide pre-preprocessed data (resample, crop, pre-align, etc.) that can be directly employed by most conventional and learning frameworks.
Only displacement fields in voxel dimensions in a standard orientation will have to be provided by participants.
We invite interested researcher from academia and industry to apply either learning based or conventional medical registration algorithms to one or up to 4 clinically relevant sub-tasks.
The tasks cover both intra- and inter-patient alignment, CT, ultrasound and MRI modalities, neuro-, thorax and abdominal anatomies that requires participants to address ubiquitous challenges of medical image registration: learning from small datasets; estimating large deformations; dealing with multi-modal scans and learning from noisy annotations
Important Dates:
- Early June 2020: all training scans for all 4 tasks available for download
- Early July 2020: open submission and release evaluation script for validation (subset of training data)
September 25th: submission of docker that computes displacement fields for test scans (inference on NVIDIA GPUs, relevant for computation time bonus) October 2nd: rolling submissions of new participants without the ability to gain ranking scores for computation time
Publications and Prizes:
- All participants of the L2R challenge may submit a short LNCS Springer paper (up to 4 pages) with a deadline of October 16th.
- We encourage authors of novel contributions to submit extended papers to MELBA (The Journal of Machine Learning for Biomedical Imaging).
- Highly ranked teams (for single tasks or overall performance) will be invited to contribute to a challenge overview paper, which will be submitted to a high impact journal (MedIA/TMI). Invitations will follow in the weeks after the workshop in October.
- We are very happy to announce that Nvidia is sponsoring a Titan RTX (value $2500) for the overall winner and Scaleway is offering a 500€ prize (in GPU computing) for the best GPU-based submission.
Please contact us at learn2reg@gmail.com for any queries (Mattias, Lasse, Alessa and Adrian)
Mattias Heinrich (Associate Professor, Universität zu Lübeck)
AutoImplant:
The MICCAI 2020 Cranial Implant Design Challenge
Summary:
Cranioplasty is the surgical process where a skull defect is repaired using a cranial implant, which must fit precisely against the borders of the defect, as replacement to the removed cranial bone. The design of the cranial implant is a challenging task. It usually involves intensive human interaction, is time-consuming and requires expertise in the specific medical domain. Therefore, a fast and automatic design of cranial implants is highly desirable and would furthermore enable in-operation room manufacturing of the implants for the patient. For the AutoImplant challenge, which will be held on October 8, 2020 as a half-day event at MICCAI 2020 virtual conference, we seek data-driven algorithms (such as deep learning) for automatic skull defect restoration. We provide 200 high-resolution healthy and defected data pairs, on which participants can base their solutions. We expect the participants' submissions between September 1 and September 10, 2020. The challenge results will be published as LNCS challenge proceedings and, depending on the outcome, as a joint publication in a top tier medical imaging journal.
Submission period: September 1, 2020 - September 10, 2020
Challenge date: October 8, 2020
Website: https://autoimplant.grand-challenge.org/
Contact: autoimplant.challenge@gmail.com
Call for Papers:
The Second International Workshop on Thoracic Image Analysis
A MICCAI 2020 Workshop
The Second Workshop on Thoracic Image Analysis (TIA) brings together medical image analysis researchers in the area of thoracic imaging to discuss recent advances in this rapidly developing field. COVID-19 infection has brought a lot of attention to lung imaging, and the role of CT imaging in the diagnostic workflow of suspects is an important research topic. In addition to that, cardiovascular disease, lung cancer and COPD, three diseases all visible on thoracic imaging, are amongst the top causes of death worldwide. Many imaging modalities are currently available to study the pulmonary and cardiac system, including radiography, CT, PET, MRI, and ultrasound. We invite papers that deal with all aspects of image analysis of thoracic data and particularly welcome novel work focused around the need for new methodologies for predisposition, diagnosis, staging, and resolution assessment of COVID-19 infections.
Paper submission deadline: June 20, 2020
Workshop date: October 8, 2020
Website: www.lungworkshop.org/2020/
Contact: TIA2020@iibi.uiowa.edu
Von der AG mit organisiert oder unterstützt
Workshop Bildverarbeitung für die Medizin (BVM)
turnus: jährlich
nächster Workshop: 15.3.-17.3.2020, Berlin, Link
Workshop Deep Learning in der Medizinischen Informatik und der Bioinformatik
am 9.9.2019 in Dortmund Link
Biomedical Image and Signal Computing (BISC)
gemeinsamer Workshop zusammen mit der DGBMT
- BISC im Rahmen der GMDS Jahrestagung in Lübeck, 2013
- BISC als Focussession im Rahmen der BMT in Lübeck, 2015