News
07-16-2024 — New publication in Medical Physics
Tom Kumschier from the Chair of Biomedical Physics has published a new paper in Medical Physics. In this work, with involvement from Tobias Lasser, a method for streak artifact removal in X-ray darkf-field computed tomography using CNNs is presented.
Related publications:
- Streak artefact removal in x-ray dark-field computed tomography using a convolutional neural network. Medical Physics 51, 2024 DOI
05-30-2024 — New publication in PR
Our own Peng Tang has published a new paper in Pattern Recognition (IF 8). In this work we present a joint-individual fusion structure for multi-modal skin cancer classification.
Related publications:
- Joint-Individual Fusion Structure with Fusion Attention Module for Multi-Modal Skin Cancer Classification. Pattern Recognition 154, 2024 DOI
05-16-2024 — New publication in IEEE TMI
Jakob Häusele from the Chair of Biomedical Physics has published a new paper in IEEE TMI (IF 10.05). In this work, with involvement from Tobias Lasser, a robust sample processing pipeline for a clinical X-ray dark-field computed tomography scanner is presented.
Related publications:
- Robust Sample Information Retrieval in Dark-Field Computed Tomography with a Vibrating Talbot-Lau Interferometer. IEEE Transactions on Medical Imaging , 2024 DOI
05-11-2024 — New publication in Radiology: Artificial Intelligence
Our partners from the Chair of Biomedical Physics have published a new paper in the journal Radiology: Artificial Intelligence on improving hemorrhage detection in sparse-view X-ray computed tomography using U-net based artifact reduction.
Related publications:
- Improving Automated Hemorrhage Detection in Sparse-view Computed Tomography via Deep Convolutional Neural Network based Artifact Reduction. Radiology: Artificial Intelligence 6, 2024 DOI
05-03-2024 — New publication in European Radiology Experimental
Our partners from the Chair of Biomedical Physics have published a new paper in the journal European Radiology Experimental on improving lung metastasis detection in sparse-view X-ray computed tomography.
Related publications:
- Improving image quality of sparse-view lung tumor CT images with U-Net. European Radiology Experimental 8, 2024 DOI
04-29-2024 — Tobias Lasser is now an associated fellow of relAI
Our own Tobias Lasser is now an associated fellow of the Konrad Zuse School of Excellence in Reliable AI (relAI).
04-26-2024 — New publication in RöFo
Our own Alessandro Wollek has published a new paper in RöFo - Fortschritte auf dem Gebiet der Röntgensrahlen und der bildgebenden Verfahren. In this work we present language-model based automatic labeling for Chest X-ray radiology reports.
Related publications:
- Language model-based labeling of German thoracic radiology reports. Fortschr Röntgenstr , 2024 DOI
03-16-2024 — One presentation at CT Meeting 2024
Our own David Frank will present the curadon library at the CT Meeting 2024 in August 2024 in Bamberg, Germany.
Related publications:
- curadon: an Open-Source Foundational Library for Highly Efficient Differentiable Projectors in X-ray Computed Tomography. International Conference on Image Formation in X-ray Computed Tomography (CT Meeting), Bamberg, Germany, August, 2024
02-09-2024 — New publication in JOSS
Our own David Frank has published a new paper in The Journal of Open Source Software. In this work we present our open-source framework elsa for tomographic reconstruction.
Related publications:
- elsa: an elegant framework for tomographic reconstruction. Journal of Open Source Software 9, 2024 DOI
02-01-2024 — New publication in RöFo
Our own Alessandro Wollek has published a new paper in RöFo - Fortschritte auf dem Gebiet der Röntgensrahlen und der bildgebenden Verfahren. In this work we present a German CheXpert labeler for Chest X-ray radiology reports.
Related publications:
- German CheXpert Chest X-ray Radiology Report Labeler. Fortschr Röntgenstr 196, 2024 DOI
01-30-2024 — New publication in Biomedical Optics Express
Our own Josué Page Vizcaíno has published a new paper in Biomedical Optics Express. In this work we present the Conditional Wavelet Flow Architecture (CWFA), a technique for fast light-field 3D microscopy with out-of-distribution detection and adaption.
Related publications:
- Fast light-field 3D microscopy with out-of-distribution detection and adaptation through Conditional Normalizing Flows. Biomedical Optics Express 15, 2024 DOI
12-12-2023 — One presentation at AAAI 2024
Our own Peng Tang will present his work on single image defocus deblurring at the AAAI conference 2024.
Related publications:
- Prior and Prediction Inverse Kernel Transformer for Single Image Defocus Deblurring. Annual AAAI Conference on Artificial Intelligence, Vancouver, Canada, February, 2024
12-06-2023 — New publication in J. Imaging
Our own Alessandro Wollek has published a new paper in the Journal of Imaging. In this work we present WindowNet, a technique to improve chest X-ray classification by including learnable windows.
Related publications:
- WindowNet: Learnable Windows for Chest X-ray Classification. Journal of Imaging 9, 2023 DOI
12-05-2023 — One presentation at SPIE Medical Imaging 2024
Our partners from the Chair of Biomedical Physics will present a new work on dark-field imaging on a clinical X-ray CT system relating helix and surview scans, with a little involvement from our group, at the SPIE Medical Imaging 2024 conference from Feb. 18-22, 2024, in San Diego, USA.
Related publications:
- Advancing X-ray Dark-Field Imaging for Human-Scale Applications: Helical CT and Surview Imaging. SPIE Medical Imaging, San Diego, USA, February, 2024 DOI
11-15-2023 — PhD defense Erdal Pekel on November 15, 2023
Our own Erdal Pekel is defending his PhD thesis on Wednesday, November 15, 2023.
The oral exam is taking place on Wednesday, November 15, 2023, at 10:00, in room 1.211 (MIBE) in presence, and remotely via Zoom (request passcode via email).
The title of the dissertation is X-ray Computed Tomography with a Robotic Sample Holder.
11-08-2023 — Tobias Lasser wins the 2023 Outstanding Editorial Board Member Award by IEEE SPS
Our own Tobias Lasser won the 2023 Outstanding Editorial Board Member Award from the IEEE Signal Processing Society. It was awarded for outstanding editorial board service for the IEEE Transactions on Computational Imaging.
11-03-2023 — New publication in ERX
Erdal Pekel has published a new paper in the journal Engineering Research Express. In this work, we present an acquisition trajectories optimization method for an X-ray computed tomography system with a seven degree of freedom robotic sample holder that does not require any prior knowledge about a sample.
Related publications:
- Runtime optimization of acquisition trajectories for X-ray computed tomography with a robotic sample holder. Engineering Research Express 5, 2023 DOI
10-25-2023 — PhD defense Alessandro Wollek on October 25, 2023
Our own Alessandro Wollek is defending his PhD thesis on Wednesday, October 25, 2023.
The oral exam is taking place on Wednesday, October 25, 2023, at 15:00, in room 01.07.014 in presence, and remotely via BBB (request passcode via email).
The title of the dissertation is Deep Learning for Clinical Decision Support Systems in Chest Radiography.
10-13-2023 — New publication in Med. Phys.
Our own Alessandro Wollek has published a new paper in the journal Medical Physics. In this work, we present a novel method for detecting out-of-distribution samples using in-distribution voting, with the example of chest X-ray classification.
Related publications:
- Out-of-distribution detection with in-distribution voting using the medical example of chest X-ray classification. Medical Physics 51, 2024 DOI
09-30-2023 — New publication in WAOJ
Our cooperation partner R. Kaczmarczyk from the Department of Dermatology and Allergy (TUM), has published a new paper in the World Allergy Organization Journal. In this work, we study IgE sensitization patterns in Hymenoptra venom allergy.
Related publications:
- Revealing clinically relevant specific IgE sensitization patterns in Hymenoptera venom allergy with dimension reduction and clustering. World Allergy Organization Journal 16, 2023 DOI
08-18-2023 — New publication in ERX
Erdal Pekel has published a new paper in the journal Engineering Research Express. In this work, we present spherical acquisition trajectories for an X-ray computed tomography system with a seven degree of freedom robotic sample holder.
Related publications:
- Spherical acquisition trajectories for X-ray computed tomography with a robotic sample holder. Engineering Research Express 5, 2023 DOI
07-25-2023 — PhD defense Josue Page Vizcaino on July 25, 2023
Our own Josue Page Vizcaino is defending his PhD thesis on Tuesday, July 25, 2023.
The oral exam is taking place on Tuesday, July 25, 2023, at 15:00, in room 00.13.010 in presence, and via Zoom (request passcode via email).
The title of the dissertation is Enabling real-time light-field 3D fluorescence microscopy through computational microscopy and deep learning.
07-17-2023 — Our presentations at Fully3D 2023
Today, David Frank from our group will present an oral presentation about our open-source software library elsa in session D1-2 at the Fully3D conference in Stony Brook, USA. From our partners at the Chair of Biomedical Physics, Jakob Häusele will also present an oral presentation in session D1-2 about current work on the clinical X-ray dark-field computed tomography scanner.
Tomorrow, on July 18, 2023, Theodor Cheslerean-Boghiu will present posters on his own work on X-ray CT reconstruction using deep learning (session D2-P1) and Erdal Pekel's work on robotic X-ray CT (session D2-P2).
Related publications:
- An update to elsa - an elegant framework for tomographic reconstruction. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
- Translating sparse sinogram measurements into reconstruction patches without geometry information via the attention mechanism. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
- Spherical acquisition trajectories for X-ray Computed Tomography with a robotic sample holder. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
- X-ray Dark-Field Imaging at the Human Scale: Helical Computed Tomography and Surview Imaging. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
07-12-2023 — Seminar and practical course in winter term 2023/24
In the upcoming winter term 2023/24, CIIP is offering various courses again.
This includes:- the lecture Concepts of C++ programming (IN2377)
- the seminar course Computational Methods for Image Reconstruction
- the practical course Large Scale Software Architectures: Analysis, Design, and Implementation
Registration for the seminar and practical courses takes place via the Matching Tool, but you also have to fill out a registration form with some information about you (this is mandatory, check the course websites for the link).
A preliminary meeting for the practical course Large Scale Software Architectures: Analysis, Design, and Implementation will take place on Friday, July 14, 2023, at 10:00 via BBB.
04-29-2023 — New publication in IEEE TMI
Jakob Häusele from the Chair of Biomedical Physics has published a new paper in IEEE TMI (IF 10.05). In this work, with involvement from Tobias Lasser, a novel processing pipeline for a clinical X-ray dark-field computed tomography scanner is presented.
Related publications:
- Advanced Phase-Retrieval for Stepping-Free X-Ray Dark-Field Computed Tomography. IEEE Transactions on Medical Imaging 42, 2023 DOI
04-15-2023 — Four presentations at Fully3D 2023
Our group will be presenting several contributions at this year's Fully3D on July 16-21, 2023, in Stony Brook, USA.
From CIIP directly, we will present works on our open-source software library elsa and deep learning methods for sparse X-ray computed tomography, as well as on spherical acquisitions for robotic X-ray CT systems.
In collaboration with our partners from the Chair of Biomedical Physics, we will present one work on dark-field imaging on a clinical X-ray CT system.
Related publications:
- An update to elsa - an elegant framework for tomographic reconstruction. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
- Translating sparse sinogram measurements into reconstruction patches without geometry information via the attention mechanism. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
- Spherical acquisition trajectories for X-ray Computed Tomography with a robotic sample holder. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
- X-ray Dark-Field Imaging at the Human Scale: Helical Computed Tomography and Surview Imaging. Fully3 Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully 3D), Stony Brook, USA, July, 2023
03-01-2023 — New publication in Radiology AI
Alessandro Wollek has published a new paper in the journal Radiology: Artificial Intelligence. In this work, we investigate chest radiograph classification performance of vision transformers and interpretability of attention-based saliency using the example of pneumothorax classification.
Related publications:
- Attention-based Saliency Maps Improve Interpretability of Pneumothorax Classification. Radiology: Artificial Intelligence 5, 2023 DOI
01-27-2023 — New publication in IEEE TCI
Theodor Cheslerean-Boghiu has published a new paper in the journal IEEE Transactions on Computational Imaging. In this work, we present a dual-domain hybrid technique for reconstructing sparse X-ray CT acquisitions, where we use both deep learning and known physical operators.
Related publications:
- WNet: A Data-driven Dual-domain Denoising Model for Sparse-view Computed Tomography with a Trainable Reconstruction Layer. IEEE Transactions on Computational Imaging 9, 2023 DOI
01-06-2023 — New publication in JEADV
Dr. Maximilian Schielein and Joshua Christl from the Department of Dermatology and Allergy of the Klinikum rechts der Isar have published a new paper in JEADV (IF 6.17). In this work, with involvement from Tobias Lasser, a deep learning technique for outlier detection in binary classification of non-melanoma skin diseases in dermatology is presented.
Related publications:
- Outlier Detection in Dermatology: Performance of different Convolutional Neural Networks for Binary Classification of Inflammatory Skin Diseases. Journal of the European Academy of Dermatology and Venereology 9, 2023 DOI
11-11-2022 — New publication in IEEE TMI
Clemens Schmid from the Chair of Biomedical Physics has published a new paper in IEEE TMI (IF 10.05). In this work, with involvement from Tobias Lasser, a novel processing pipeline for a clinical dark-field computed tomography scanner is presented.
Related publications:
- Modeling vibrations of a tiled Talbot-Lau interferometer on a clinical CT. IEEE Transactions on Medical Imaging 42, 2022 DOI
08-05-2022 — New publication in ERX
Erdal Pekel has published a new paper in the journal Engineering Research Express. In this work, we present an X-ray computed tomography system with a seven degree of freedom robotic sample holder.
Related publications:
- X-ray computed tomography with seven degree of freedom robotic sample holder. Engineering Research Express 4, 2022 DOI
07-25-2022 — New publication in JEADV
Dr. Sebastian Krammer from the Department of Dermatology and Allergy of the LMU Klinikum has published a new paper in JEADV (IF 6.17). In this work, with involvement from Peng Tang and Tobias Lasser, a deep learning technique for the classification of dermatological lesions is presented when only a limited amount of labeled training data is available.
Related publications:
- Deep learning-based classification of dermatological lesions given a limited amount of labeled data. Journal of the European Academy of Dermatology and Venereology 36, 2022 DOI
07-12-2022 — Three presentations at IEEE MIC 2022
Our partners from the Chair of Biomedical Physics will present several joint works on dark-field imaging using a clinical X-ray CT system at this year's IEEE Medical Imaging Conference from November 5-12, 2022, in Milano, Italy.
Related publications:
- Performance and Potential of Human Sized Dark-Field CT Based on Recent Results. IEEE Medical Imaging Conference, Milano, Italy, November, 2022
- Modelling vibrations of a Talbot-Lau interferometer with tiled analyzer grating on a clinical CT system. IEEE Medical Imaging Conference, Milano, Italy, November, 2022
- Sample Data Processing of the First Clinical Dark-Field CT Prototype. IEEE Medical Imaging Conference, Milano, Italy, November, 2022
05-04-2022 — New publication in JEADV
Theresa Willem from the Institut für Geschichte und Ethik der Medizin has published a new paper in JEADV (IF 6.17). In this work, with involvement from Tobias Lasser, the risks and benefits of machine learning healthcare applications in dermatology are analyzed from an ethical view point.
Related publications:
- Risks and benefits of dermatological machine learning healthcare applications - an overview an ethical analysis. Journal of the European Academy of Dermatology and Venereology 36, 2022 DOI
04-13-2022 — Five presentations at CT Meeting 2022
Our group will be presenting several contributions at this year's CT Meeting on June 13-16, 2022, in Baltimore, USA.
From CIIP directly, we will present works on a hybrid reconstruction method using shearlets and deep learning for sparse X-ray computed tomography, as well as on calibration methods for X-ray CT systems with robotic sample holders.
In collaboration with our partners from the Chair of Biomedical Physics, we will present several works on dark-field imaging on a clinical X-ray CT system.
Related publications:
- Hybrid Reconstruction Using Shearlets and Deep Learning for Sparse X-ray Computed Tomography. International Conference on Image Formation in X-ray Computed Tomography (CT Meeting), Baltimore, USA, June, 2022 DOI
- Geometric calibration of seven degree of freedom Robotic Sample Holder for X-ray CT. International Conference on Image Formation in X-ray Computed Tomography (CT Meeting), Baltimore, USA, June, 2022 DOI
- Dark-Field Imaging on a Clinical CT System: Performance and Potential based on first Results. International Conference on Image Formation in X-ray Computed Tomography (CT Meeting), Baltimore, USA, June, 2022 DOI
- Dark-Field Imaging on a Clinical CT System: Modelling of Interferometer Vibrations. International Conference on Image Formation in X-ray Computed Tomography (CT Meeting), Baltimore, USA, June, 2022 DOI
- Dark-Field Imaging on a Clinical CT System: Sample Data Processing and Reconstruction. International Conference on Image Formation in X-ray Computed Tomography (CT Meeting), Baltimore, USA, June, 2022 DOI
04-08-2022 — PhD defense of Anca Stefanoiu
On April 8, 2022, at 9:00, our own Anca Stefanoiu defends her PhD. Her thesis has the topic Fluorescence light field microscopy: wave optics modeling and 3D image reconstruction.
03-21-2022 — Course in Ferienakademie 2022
Together with Christian Riess from FAU Erlangen, we are again offering a course on Computational Medical Imaging in the Ferienakademie 2022 from September 18 to 30, 2022, in Sarntal Valley, South Tyrol, Italy.
If you want to take part, please apply until May 10, 2022.
02-08-2022 — New publication in PNAS
Manuel Viermetz from the Chair of Biomedical Physics has published a new paper in PNAS (IF 11.2). In this work, with involvement from Tobias Lasser, a novel implementation of a clinical dark-field computed tomography scanner is presented.
The corresponding TUM press release on Feb. 8, 2022.
Related publications:
- Dark-field computed tomography reaches the human scale. Proceedings of the National Academy of Sciences of the United States of America 119, 2022 DOI
11-23-2021 — New publication in Medical Image Analysis
Peng Tang has published a new paper in the journal Medical Image Analysis (IF 8.55). It presents a new multi-stage multi-modal learning algorithm for multi-label skin lesion classification.
Update: There was a corresponding TUM press release on Jan. 14, 2022.
Related publications:
- FusionM4Net: A multi-stage multi-modal learning algorithm for multi-label skin lesion classification. Medical Image Analysis 76, 2022 DOI
10-06-2021 — Student tutors wanted for BMT
We are UPDATE: no longer (since all positions have been filled) looking for student tutors for our course Basic Mathematical Tools for Imaging and Visualization in the winter term 2021/22. The job is 8h or 16h/week during the lecture period and involves the correction of homework.
If you are interested, please get in touch with Tobias Lasser.
For general information on student tutor jobs, please check out the department website.
07-23-2021 — Best Poster Award: Fully3D 2021
Ana Radutoiu won a best poster award at the Fully3D 2021 conference.
At the 16th Virtual International Meeting on Fully 3D Image Reconstruction in Radiology and Nuclear Medicine (Fully3D), Ana Radutoiu won the panel poster award for her poster on Laplace-Beltrami Regularization for Anisotropic X-ray Dark-field Tomography. Her work arose from her Bachelor thesis at Computational Imaging and Inverse Problems (CIIP).
06-17-2021 — CeDoSIA Supervisory Award 2021
Tobias Lasser won the CeDoSIA Supervisory Award 2021.
The Center of Doctoral Studies in Informatics and its Applications (CeDoSIA) honors the best doctoral supervisors at the Department of Informatics, TUM. In 2021, Tobias Lasser won the first place of the award.
06-17-2021 — TeachInf Award 2021
The course Grundlagen: Algorithmen und Datenstrukturen won the TeachInf Award for the best compulsory course in the summer term 2020.
The Student Council of the Departments of Mathematics, Physics, and Informatics honors the best courses each term according to the course evaluations conducted by the Student Council. The course Grundlagen: Algorithmen und Datenstrukturen by Tobias Lasser, Bernhard Radke, Christoph Anneser, and Sebastian Wolf won the TeachInf award for the best compulsory course in the summer term 2020 at the Department of Informatics.
07-24-2020 — TeachInf Corona Award 2020
Tobias Lasser was awarded the TeachInf "Corona" Award for enabling excellent teaching during the pandemic by the Student Council of the Departments of Mathematics, Physics, and Informatics.