Search Results for author: Andreas Maier

Found 207 papers, 40 papers with code

ICC++: Explainable Image Retrieval for Art Historical Corpora using Image Composition Canvas

no code implementations22 Jun 2022 Prathmesh Madhu, Tilman Marquart, Ronak Kosti, Dirk Suckow, Peter Bell, Andreas Maier, Vincent Christlein

In this work, we present a novel approach called Image Composition Canvas (ICC++) to compare and retrieve images having similar compositional elements.

Image Retrieval

ConFUDA: Contrastive Fewshot Unsupervised Domain Adaptation for Medical Image Segmentation

no code implementations8 Jun 2022 Mingxuan Gu, Sulaiman Vesal, Mareike Thies, Zhaoya Pan, Fabian Wagner, Mirabela Rusu, Andreas Maier, Ronak Kosti

Then, to align the source and target features and tackle the memory issue of the traditional contrastive loss, we propose the centroid-based contrastive learning (CCL) and a centroid norm regularizer (CNR) to optimize the contrastive pairs in both direction and magnitude.

Contrastive Learning Medical Image Segmentation +3

DeepTechnome: Mitigating Unknown Bias in Deep Learning Based Assessment of CT Images

no code implementations26 May 2022 Simon Langer, Oliver Taubmann, Felix Denzinger, Andreas Maier, Alexander Mühlberg

Reliably detecting diseases using relevant biological information is crucial for real-world applicability of deep learning techniques in medical imaging.

Building Brains: Subvolume Recombination for Data Augmentation in Large Vessel Occlusion Detection

no code implementations5 May 2022 Florian Thamm, Oliver Taubmann, Markus Jürgens, Aleksandra Thamm, Felix Denzinger, Leonhard Rist, Hendrik Ditt, Andreas Maier

The best configuration detects LVOs with an AUC of 0. 91, LVOs in the ICA with an AUC of 0. 96, and in the MCA with 0. 91 while accurately predicting the affected side.

Data Augmentation

Continual Learning for Peer-to-Peer Federated Learning: A Study on Automated Brain Metastasis Identification

no code implementations26 Apr 2022 Yixing Huang, Christoph Bert, Stefan Fischer, Manuel Schmidt, Arnd Dörfler, Andreas Maier, Rainer Fietkau, Florian Putz

With iterative continual learning (i. e., the shared model revisits each center multiple times during training), the sensitivity is further improved to 0. 914, which is identical to the sensitivity using mixed data for training.

Continual Learning Federated Learning

An Algorithm for the Labeling and Interactive Visualization of the Cerebrovascular System of Ischemic Strokes

no code implementations26 Apr 2022 Florian Thamm, Markus Jürgens, Oliver Taubmann, Aleksandra Thamm, Leonhard Rist, Hendrik Ditt, Andreas Maier

In the work at hand, we place the algorithm in a clinical context by evaluating the labeling and occlusion detection on stroke patients, where we have achieved labeling sensitivities comparable to other works between 92\,\% and 95\,\%.

Disentangled Latent Speech Representation for Automatic Pathological Intelligibility Assessment

1 code implementation8 Apr 2022 Tobias Weise, Philipp Klumpp, Kubilay Can Demir, Andreas Maier, Elmar Noeth, Bjoern Heismann, Maria Schuster, Seung Hee Yang

Our results are among the first to show that disentangled speech representations can be used for automatic pathological speech intelligibility assessment, resulting in a reference speaker pair invariant method, applicable in scenarios with only few utterances available.

Simulation-Driven Training of Vision Transformers Enabling Metal Segmentation in X-Ray Images

no code implementations17 Mar 2022 Fuxin Fan, Ludwig Ritschl, Marcel Beister, Ramyar Biniazan, Björn Kreher, Tristan M. Gottschalk, Steffen Kappler, Andreas Maier

Since the generation of high quality clinical training is a constant challenge, this study proposes to generate simulated X-ray images based on CT data sets combined with self-designed computer aided design (CAD) implants and make use of convolutional neural network (CNN) and vision transformer (ViT) for metal segmentation.

Neural Network based Successor Representations of Space and Language

no code implementations22 Feb 2022 Paul Stoewer, Christian Schlieker, Achim Schilling, Claus Metzner, Andreas Maier, Patrick Krauss

We conclude that cognitive maps and neural network-based successor representations of structured knowledge provide a promising way to overcome some of the short comings of deep learning towards artificial general intelligence.

Deep learning-guided weighted averaging for signal dropout compensation in diffusion-weighted imaging of the liver

no code implementations20 Feb 2022 Fasil Gadjimuradov, Thomas Benkert, Marcel Dominik Nickel, Tobit Führes, Marc Saake, Andreas Maier

Methods: Given a set of image repetitions for a slice, a locally adaptive weighted averaging is proposed which aims to suppress the contribution of image regions affected by signal dropouts.

Learning Perspective Deformation in X-Ray Transmission Imaging

no code implementations13 Feb 2022 Yixing Huang, Andreas Maier, Rainer Fietkau, Christoph Bert, Florian Putz

The experiments on patients' chest and head data demonstrate that learning perspective deformation using dual complementary views is also applicable in anatomical X-ray data, allowing accurate cardiothoracic ratio measurements in chest X-ray images and cephalometric analysis in synthetic cephalograms from cone-beam X-ray projections.

CAD-RADS Scoring using Deep Learning and Task-Specific Centerline Labeling

no code implementations8 Feb 2022 Felix Denzinger, Michael Wels, Oliver Taubmann, Mehmet A. Gülsün, Max Schöbinger, Florian André, Sebastian J. Buss, Johannes Görich, Michael Sühling, Andreas Maier, Katharina Breininger

With coronary artery disease (CAD) persisting to be one of the leading causes of death worldwide, interest in supporting physicians with algorithms to speed up and improve diagnosis is high.

SliTraNet: Automatic Detection of Slide Transitions in Lecture Videos using Convolutional Neural Networks

no code implementations7 Feb 2022 Aline Sindel, Abner Hernandez, Seung Hee Yang, Vincent Christlein, Andreas Maier

With the increasing number of online learning material in the web, search for specific content in lecture videos can be time consuming.

online learning

Deep Learning for Ultrasound Speed-of-Sound Reconstruction: Impacts of Training Data Diversity on Stability and Robustness

no code implementations1 Feb 2022 Farnaz Khun Jush, Markus Biele, Peter M. Dueppenbecker, Andreas Maier

In this study, we investigated the impacts of training data diversity on the robustness of these networks by using multiple kinds of geometrical and natural simulated phantom structures.

Pan-Tumor CAnine cuTaneous Cancer Histology (CATCH) Dataset

no code implementations27 Jan 2022 Frauke Wilm, Marco Fragoso, Christian Marzahl, Jingna Qiu, Christof A. Bertram, Robert Klopfleisch, Andreas Maier, Katharina Breininger, Marc Aubreville

Due to morphological similarities, the differentiation of histologic sections of cutaneous tumors into individual subtypes can be challenging.

whole slide images

Ultra Low-Parameter Denoising: Trainable Bilateral Filter Layers in Computed Tomography

1 code implementation25 Jan 2022 Fabian Wagner, Mareike Thies, Mingxuan Gu, Yixing Huang, Sabrina Pechmann, Mayank Patwari, Stefan Ploner, Oliver Aust, Stefan Uderhardt, Georg Schett, Silke Christiansen, Andreas Maier

Due to the extremely low number of trainable parameters with well-defined effect, prediction reliance and data integrity is guaranteed at any time in the proposed pipelines, in contrast to most other deep learning-based denoising architectures.

Denoising SSIM

Initial Investigations Towards Non-invasive Monitoring of Chronic Wound Healing Using Deep Learning and Ultrasound Imaging

no code implementations25 Jan 2022 Maja Schlereth, Daniel Stromer, Yash Mantri, Jason Tsujimoto, Katharina Breininger, Andreas Maier, Caesar Anderson, Pranav S. Garimella, Jesse V. Jokerst

We conclude that deep learning-supported analysis of non-invasive ultrasound images is a promising area of research to automatically extract cross-sectional wound size and depth information with potential value in monitoring response to therapy.

Learned Cone-Beam CT Reconstruction Using Neural Ordinary Differential Equations

no code implementations19 Jan 2022 Mareike Thies, Fabian Wagner, Mingxuan Gu, Lukas Folle, Lina Felsner, Andreas Maier

Learned iterative reconstruction algorithms for inverse problems offer the flexibility to combine analytical knowledge about the problem with modules learned from data.

Numerical Integration

Superpixel Pre-Segmentation of HER2 Slides for Efficient Annotation

no code implementations19 Jan 2022 Mathias Öttl, Jana Mönius, Christian Marzahl, Matthias Rübner, Carol I. Geppert, Arndt Hartmann, Matthias W. Beckmann, Peter Fasching, Andreas Maier, Ramona Erber, Katharina Breininger

When evaluating the approaches on fully manually annotated images, we observe that the autoencoder-based superpixels achieve a 23% increase in boundary F1 score compared to the baseline SLIC superpixels.

Denoising Medical Image Segmentation +2

Segmentation of the Carotid Lumen and Vessel Wall using Deep Learning and Location Priors

no code implementations17 Jan 2022 Florian Thamm, Felix Denzinger, Leonhard Rist, Celia Martin Vicario, Florian Kordon, Andreas Maier

In this report we want to present our method and results for the Carotid Artery Vessel Wall Segmentation Challenge.

Deep learning for brain metastasis detection and segmentation in longitudinal MRI data

no code implementations22 Dec 2021 Yixing Huang, Christoph Bert, Philipp Sommer, Benjamin Frey, Udo Gaipl, Luitpold V. Distel, Thomas Weissmann, Michael Uder, Manuel A. Schmidt, Arnd Dörfler, Andreas Maier, Rainer Fietkau, Florian Putz

To improve brain metastasis detection performance with deep learning, a custom detection loss called volume-level sensitivity-specificity (VSS) is proposed, which rates individual metastasis detection sensitivity and specificity in (sub-)volume levels.

Ensemble Learning

Does Proprietary Software Still Offer Protection of Intellectual Property in the Age of Machine Learning? -- A Case Study using Dual Energy CT Data

no code implementations6 Dec 2021 Andreas Maier, Seung Hee Yang, Farhad Maleki, Nikesh Muthukrishnan, Reza Forghani

In the domain of medical image processing, medical device manufacturers protect their intellectual property in many cases by shipping only compiled software, i. e. binary code which can be executed but is difficult to be understood by a potential attacker.

Anomaly Detection in IR Images of PV Modules using Supervised Contrastive Learning

2 code implementations6 Dec 2021 Lukas Bommes, Mathis Hoffmann, Claudia Buerhop-Lutz, Tobias Pickel, Jens Hauch, Christoph Brabec, Andreas Maier, Ian Marius Peters

Instead, we frame fault detection as more realistic unsupervised domain adaptation problem where we train on labelled data of one source PV plant and make predictions on another target plant.

Anomaly Detection Contrastive Learning +2

Prediction of Household-level Heat-Consumption using PSO enhanced SVR Model

no code implementations3 Dec 2021 Satyaki Chatterjee, Siming Bayer, Andreas Maier

In combating climate change, an effective demand-based energy supply operation of the district energy system (DES) for heating or cooling is indispensable.

Towards Super-Resolution CEST MRI for Visualization of Small Structures

1 code implementation3 Dec 2021 Lukas Folle, Katharian Tkotz, Fasil Gadjimuradov, Lorenz Kapsner, Moritz Fabian, Sebastian Bickelhaupt, David Simon, Arnd Kleyer, Gerhard Krönke, Moritz Zaiß, Armin Nagel, Andreas Maier

This work paves the way for the prospective investigation of neural networks for super-resolution CEST MRI and, followingly, might lead to a earlier detection of the onset of rheumatic diseases.

SSIM Super-Resolution

Detection of Large Vessel Occlusions using Deep Learning by Deforming Vessel Tree Segmentations

no code implementations3 Dec 2021 Florian Thamm, Oliver Taubmann, Markus Jürgens, Hendrik Ditt, Andreas Maier

Training the EfficientNetB1 architecture on 100 data sets, the proposed augmentation scheme was able to raise the ROC AUC to 0. 85 from a baseline value of 0. 56 using no augmentation.

Automatic Plane Adjustment of Orthopedic Intra-operative Flat Panel Detector CT-Volumes

no code implementations15 Sep 2021 Celia Martin Vicario, Florian Kordon, Felix Denzinger, Jan Siad El Barbari, Maxim Privalov, Jochen Franke, Sarina Thomas, Lisa Kausch, Andreas Maier, Holger Kunze

The most important benefit of the MTL approach is that it is a single network for standard plane regression for all body regions with a reduced number of stored parameters.

Multi-Task Learning

Automated Cardiac Resting Phase Detection Targeted on the Right Coronary Artery

no code implementations6 Sep 2021 Seung Su Yoon, Elisabeth Preuhs, Michaela Schmidt, Christoph Forman, Teodora Chitiboi, Puneet Sharma, Juliano Lara Fernandes, Christoph Tillmanns, Jens Wetzl, Andreas Maier

The purpose of this work is to propose a fully automated framework that allows the detection of the right coronary artery (RCA) RP within CINE series.

Module-Power Prediction from PL Measurements using Deep Learning

no code implementations31 Aug 2021 Mathis Hoffmann, Johannes Hepp, Bernd Doll, Claudia Buerhop-Lutz, Ian Marius Peters, Christoph Brabec, Andreas Maier, Vincent Christlein

While these areas can be easily identified from electroluminescense (EL) images, this is much harder for photoluminescence (PL) images.

InSE-NET: A Perceptually Coded Audio Quality Model based on CNN

no code implementations30 Aug 2021 Guanxin Jiang, Arijit Biswas, Christian Bergler, Andreas Maier

Automatic coded audio quality assessment is an important task whose progress is hampered by the scarcity of human annotations, poor generalization to unseen codecs, bitrates, content-types, and a lack of flexibility of existing approaches.

Data Augmentation

Fiducial marker recovery and detection from severely truncated data in navigation assisted spine surgery

no code implementations25 Aug 2021 Fuxin Fan, Björn Kreher, Holger Keil, Andreas Maier, Yixing Huang

For direct detection from distorted markers in reconstructed volumes, an efficient automatic marker detection method using two neural networks and a conventional circle detection algorithm is proposed.

Deep Iterative 2D/3D Registration

no code implementations21 Jul 2021 Srikrishna Jaganathan, Jian Wang, Anja Borsdorf, Karthik Shetty, Andreas Maier

A refinement step using the classical optimization-based 2D/3D registration method applied in combination with Deep Learning-based techniques can provide the required accuracy.

Optical Flow Estimation

SmartPatch: Improving Handwritten Word Imitation with Patch Discriminators

1 code implementation21 May 2021 Alexander Mattick, Martin Mayr, Mathias Seuret, Andreas Maier, Vincent Christlein

As of recent generative adversarial networks have allowed for big leaps in the realism of generated images in diverse domains, not the least of which being handwritten text generation.

Data Augmentation Handwritten Text Recognition +1

Robust partial Fourier reconstruction for diffusion-weighted imaging using a recurrent convolutional neural network

no code implementations19 May 2021 Fasil Gadjimuradov, Thomas Benkert, Marcel Dominik Nickel, Andreas Maier

The algorithm is trained on DW liver data of 60 volunteers and evaluated on retrospectively and prospectively sub-sampled data of different anatomies and resolutions.

MR-Contrast-Aware Image-to-Image Translations with Generative Adversarial Networks

no code implementations3 Apr 2021 Jonas Denck, Jens Guehring, Andreas Maier, Eva Rothgang

Our approach is motivated by style transfer networks, whereas the "style" for an image is explicitly given in our case, as it is determined by the MR acquisition parameters our network is conditioned on.

Data Augmentation Style Transfer

Quantifying the Scanner-Induced Domain Gap in Mitosis Detection

1 code implementation30 Mar 2021 Marc Aubreville, Christof Bertram, Mitko Veta, Robert Klopfleisch, Nikolas Stathonikos, Katharina Breininger, Natalie ter Hoeve, Francesco Ciompi, Andreas Maier

Hypothesizing that the scanner device plays a decisive role in this effect, we evaluated the susceptibility of a standard mitosis detection approach to the domain shift introduced by using a different whole slide scanner.

Mitosis Detection

Is Medical Chest X-ray Data Anonymous?

no code implementations15 Mar 2021 Kai Packhäuser, Sebastian Gündel, Nicolas Münster, Christopher Syben, Vincent Christlein, Andreas Maier

Our verification system is able to identify whether two frontal chest X-ray images are from the same person with an AUC of 0. 9940 and a classification accuracy of 95. 55%.

Pixel-wise Distance Regression for Glacier Calving Front Detection and Segmentation

no code implementations9 Mar 2021 Amirabbas Davari, Christoph Baller, Thorsten Seehaus, Matthias Braun, Andreas Maier, Vincent Christlein

In this work, we propose to mitigate the class-imbalance between the calving front class and the non-calving front class by reformulating the segmentation problem into a pixel-wise regression task.

Enhanced Magnetic Resonance Image Synthesis with Contrast-Aware Generative Adversarial Networks

no code implementations17 Feb 2021 Jonas Denck, Jens Guehring, Andreas Maier, Eva Rothgang

A Magnetic Resonance Imaging (MRI) exam typically consists of the acquisition of multiple MR pulse sequences, which are required for a reliable diagnosis.

Image Generation

Learning the Update Operator for 2D/3D Image Registration

no code implementations4 Feb 2021 Srikrishna Jaganathan, Jian Wang, Anja Borsdorf, Andreas Maier

We aim to address this gap by incorporating traditional methods in deep neural networks using known operator learning.

Image Registration Operator learning

Learning-Based Patch-Wise Metal Segmentation with Consistency Check

no code implementations26 Jan 2021 Tristan M. Gottschalk, Andreas Maier, Florian Kordon, Björn W. Kreher

Metal implants that are inserted into the patient's body during trauma interventions cause heavy artifacts in 3D X-ray acquisitions.

Metal Artifact Reduction

X-ray Scatter Estimation Using Deep Splines

no code implementations22 Jan 2021 Philipp Roser, Annette Birkhold, Alexander Preuhs, Christopher Syben, Lina Felsner, Elisabeth Hoppe, Norbert Strobel, Markus Korwarschik, Rebecca Fahrig, Andreas Maier

Algorithmic X-ray scatter compensation is a desirable technique in flat-panel X-ray imaging and cone-beam computed tomography.

Medical Physics Image and Video Processing

Learning to be EXACT, Cell Detection for Asthma on Partially Annotated Whole Slide Images

no code implementations13 Jan 2021 Christian Marzahl, Christof A. Bertram, Frauke Wilm, Jörn Voigt, Ann K. Barton, Robert Klopfleisch, Katharina Breininger, Andreas Maier, Marc Aubreville

We evaluated our pipeline in a cross-validation setup with a fixed training set using a dataset of six equine WSIs of which four are partially annotated and used for training, and two fully annotated WSI are used for validation and testing.

object-detection Object Detection +1

Coronary Plaque Analysis for CT Angiography Clinical Research

no code implementations11 Jan 2021 Felix Denzinger, Michael Wels, Christian Hopfgartner, Jing Lu, Max Schöbinger, Andreas Maier, Michael Sühling

However, to enable clinical research with the help of these algorithms, a software solution, which enables manual correction, comprehensive visual feedback and tissue analysis capabilities, is needed.

Glacier Calving Front Segmentation Using Attention U-Net

no code implementations8 Jan 2021 Michael Holzmann, Amirabbas Davari, Thorsten Seehaus, Matthias Braun, Andreas Maier, Vincent Christlein

An essential climate variable to determine the tidewater glacier status is the location of the calving front position and the separation of seasonal variability from long-term trends.

2-D Respiration Navigation Framework for 3-D Continuous Cardiac Magnetic Resonance Imaging

no code implementations26 Dec 2020 Elisabeth Hoppe, Jens Wetzl, Philipp Roser, Lina Felsner, Alexander Preuhs, Andreas Maier

Continuous protocols for cardiac magnetic resonance imaging enable sampling of the cardiac anatomy simultaneously resolved into cardiac phases.

Spatio-temporal Multi-task Learning for Cardiac MRI Left Ventricle Quantification

1 code implementation24 Dec 2020 Sulaiman Vesal, Mingxuan Gu, Andreas Maier, Nishant Ravikumar

In this paper, we propose a spatio-temporal multi-task learning approach to obtain a complete set of measurements quantifying cardiac LV morphology, regional-wall thickness (RWT), and additionally detecting the cardiac phase cycle (systole and diastole) for a given 3D Cine-magnetic resonance (MR) image sequence.

Multi-Task Learning

Combined subleading high-energy logarithms and NLO accuracy for W production in association with multiple jets

no code implementations18 Dec 2020 Jeppe R. Andersen, James A. Black, Helen M. Brooks, Emmet P. Byrne, Andreas Maier, Jennifer M. Smillie

Large logarithmic corrections in $\hat s/p_t^2$ lead to substantial variations in the perturbative predictions for inclusive $W$-plus-dijet processes at the Large Hadron Collider.

High Energy Physics - Phenomenology

Enhancing Human Pose Estimation in Ancient Vase Paintings via Perceptually-grounded Style Transfer Learning

1 code implementation10 Dec 2020 Prathmesh Madhu, Angel Villar-Corrales, Ronak Kosti, Torsten Bendschus, Corinna Reinhardt, Peter Bell, Andreas Maier, Vincent Christlein

(2) To improve the already strong results further, we created a small dataset (ClassArch) consisting of ancient Greek vase paintings from the 6-5th century BCE with person and pose annotations.

Image Retrieval Pose Estimation +2

Robustness Investigation on Deep Learning CT Reconstruction for Real-Time Dose Optimization

no code implementations7 Dec 2020 Chang Liu, Yixing Huang, Joscha Maier, Laura Klein, Marc Kachelrieß, Andreas Maier

For organ-specific AEC, a preliminary CT reconstruction is necessary to estimate organ shapes for dose optimization, where only a few projections are allowed for real-time reconstruction.

Computed Tomography (CT) Image Reconstruction

Reconstruction of Voxels with Position- and Angle-Dependent Weightings

no code implementations27 Oct 2020 Lina Felsner, Tobias Würfl, Christopher Syben, Philipp Roser, Alexander Preuhs, Andreas Maier, Christian Riess

In this work we first formulate this reconstruction problem in terms of a system matrix and weighting part.

ICFHR 2020 Competition on Image Retrieval for Historical Handwritten Fragments

1 code implementation20 Oct 2020 Mathias Seuret, Anguelos Nicolaou, Dominique Stutzmann, Andreas Maier, Vincent Christlein

In particular, we investigate the performance of large-scale retrieval of historical document fragments in terms of style and writer identification.

Image Retrieval

Deep Learning-based Pipeline for Module Power Prediction from EL Measurements

1 code implementation30 Sep 2020 Mathis Hoffmann, Claudia Buerhop-Lutz, Luca Reeb, Tobias Pickel, Thilo Winkler, Bernd Doll, Tobias Würfl, Ian Marius Peters, Christoph Brabec, Andreas Maier, Vincent Christlein

However, knowledge of the power at maximum power point is important as well, since drops in the power of a single module can affect the performance of an entire string.

Graph convolutional regression of cardiac depolarization from sparse endocardial maps

no code implementations28 Sep 2020 Felix Meister, Tiziano Passerini, Chloé Audigier, Èric Lluch, Viorel Mihalef, Hiroshi Ashikaga, Andreas Maier, Henry Halperin, Tommaso Mansi

The training set consists of data produced by a computational model of cardiac electrophysiology on a large cohort of synthetically generated geometries of ischemic hearts.

Cephalogram Synthesis and Landmark Detection in Dental Cone-Beam CT Systems

no code implementations9 Sep 2020 Yixing Huang, Fuxin Fan, Christopher Syben, Philipp Roser, Leonid Mill, Andreas Maier

The method trained on conventional cephalograms can be directly applied to landmark detection in the synthetic cephalograms, achieving 93. 0% and 80. 7% successful detection rate in 4 mm precision range for synthetic cephalograms from 3D volumes and 2D projections respectively.

3D Reconstruction Super-Resolution

The Effect of Various Strengths of Noises and Data Augmentations on Classification of Short Single-Lead ECG Signals Using Deep Neural Networks

no code implementations14 Aug 2020 Faezeh Nejati Hatamian, Amirabbas Davari, Andreas Maier

Due to the multiple imperfections during the signal acquisition, Electrocardiogram (ECG) datasets are typically contaminated with numerous types of noise, like salt and pepper and baseline drift.

Denoising

A Learning-based Method for Online Adjustment of C-arm Cone-Beam CT Source Trajectories for Artifact Avoidance

no code implementations14 Aug 2020 Mareike Thies, Jan-Nico Zäch, Cong Gao, Russell Taylor, Nassir Navab, Andreas Maier, Mathias Unberath

We propose to adjust the C-arm CBCT source trajectory during the scan to optimize reconstruction quality with respect to a certain task, i. e. verification of screw placement.

Tomographic Reconstructions

Extracting and Leveraging Nodule Features with Lung Inpainting for Local Feature Augmentation

no code implementations5 Aug 2020 Sebastian Guendel, Arnaud Arindra Adiyoso Setio, Sasa Grbic, Andreas Maier, Dorin Comaniciu

However, because of the limited availability of scans containing nodules and the subtle properties of nodules in CXRs, state-of-the-art methods do not perform well on nodule classification.

The Notary in the Haystack -- Countering Class Imbalance in Document Processing with CNNs

no code implementations15 Jul 2020 Martin Leipert, Georg Vogeler, Mathias Seuret, Andreas Maier, Vincent Christlein

In classification, notarial instruments are distinguished from other documents, while the notary sign is separated from the certificate in the segmentation task.

Classification Data Augmentation +1

Re-ranking for Writer Identification and Writer Retrieval

no code implementations14 Jul 2020 Simon Jordan, Mathias Seuret, Pavel Král, Ladislav Lenc, Jiří Martínek, Barbara Wiermann, Tobias Schwinger, Andreas Maier, Vincent Christlein

We show that a re-ranking step based on k-reciprocal nearest neighbor relationships is advantageous for writer identification, even if only a few samples per writer are available.

Re-Ranking

JBFnet -- Low Dose CT Denoising by Trainable Joint Bilateral Filtering

no code implementations9 Jul 2020 Mayank Patwari, Ralf Gutjahr, Rainer Raupach, Andreas Maier

JBFnet is split into four filtering blocks, each of which performs Joint Bilateral Filtering.

Denoising

Inertial Measurements for Motion Compensation in Weight-bearing Cone-beam CT of the Knee

no code implementations9 Jul 2020 Jennifer Maier, Marlies Nitschke, Jang-Hwan Choi, Garry Gold, Rebecca Fahrig, Bjoern M. Eskofier, Andreas Maier

In our proposed multi-stage algorithm, these signals are transformed to the global coordinate system of the CT scan and applied for motion compensation during reconstruction.

Computed Tomography (CT) Motion Compensation

Simultaneous Estimation of X-ray Back-Scatter and Forward-Scatter using Multi-Task Learning

no code implementations8 Jul 2020 Philipp Roser, Xia Zhong, Annette Birkhold, Alexander Preuhs, Christopher Syben, Elisabeth Hoppe, Norbert Strobel, Markus Kowarschik, Rebecca Fahrig, Andreas Maier

Here, we propose a novel approach combining conventional techniques with learning-based methods to simultaneously estimate the forward-scatter reaching the detector as well as the back-scatter affecting the patient skin dose.

Multi-Task Learning

Appearance Learning for Image-based Motion Estimation in Tomography

no code implementations18 Jun 2020 Alexander Preuhs, Michael Manhart, Philipp Roser, Elisabeth Hoppe, Yixing Huang, Marios Psychogios, Markus Kowarschik, Andreas Maier

To this end, we train a siamese triplet network to predict the reprojection error (RPE) for the complete acquisition as well as an approximate distribution of the RPE along the single views from the reconstructed volume in a multi-task learning approach.

Motion Estimation Multi-Task Learning

Data Consistent CT Reconstruction from Insufficient Data with Learned Prior Images

no code implementations20 May 2020 Yixing Huang, Alexander Preuhs, Michael Manhart, Guenter Lauritsch, Andreas Maier

For example, for truncated data, DCR achieves a mean root-mean-square error of 24 HU and a mean structure similarity index of 0. 999 inside the field-of-view for different patients in the noisy case, while the state-of-the-art U-Net method achieves 55 HU and 0. 995 respectively for these two metrics.

Computed Tomography (CT) Image Reconstruction

Projection Inpainting Using Partial Convolution for Metal Artifact Reduction

no code implementations2 May 2020 Lin Yuan, Yixing Huang, Andreas Maier

In this work, partial convolution is applied for projection inpainting, which only relies on valid pixels values.

Metal Artifact Reduction

Will we ever have Conscious Machines?

no code implementations31 Mar 2020 Patrick Krauss, Andreas Maier

The question of whether artificial beings or machines could become self-aware or consciousness has been a philosophical question for centuries.

Recognizing Characters in Art History Using Deep Learning

1 code implementation31 Mar 2020 Prathmesh Madhu, Ronak Kosti, Lara Mührenberg, Peter Bell, Andreas Maier, Vincent Christlein

We present experiments and analysis on three different models and show that the model trained on domain related data gives the best performance for recognizing character.

Computer Vision

Spatio-Temporal Handwriting Imitation

2 code implementations24 Mar 2020 Martin Mayr, Martin Stumpf, Anguelos Nicolaou, Mathias Seuret, Andreas Maier, Vincent Christlein

Then, a method for online handwriting synthesis is used to produce a new realistic-looking text primed with the online input sequence.

The Effect of Data Augmentation on Classification of Atrial Fibrillation in Short Single-Lead ECG Signals Using Deep Neural Networks

no code implementations7 Feb 2020 Faezeh Nejati Hatamian, Nishant Ravikumar, Sulaiman Vesal, Felix P. Kemeth, Matthias Struck, Andreas Maier

In this study, we investigate the impact of various data augmentation algorithms, e. g., oversampling, Gaussian Mixture Models (GMMs) and Generative Adversarial Networks (GANs), on solving the class imbalance problem.

Classification Data Augmentation +2

CLCNet: Deep learning-based Noise Reduction for Hearing Aids using Complex Linear Coding

no code implementations28 Jan 2020 Hendrik Schröter, Tobias Rosenkranz, Alberto N. Escalante B., Marc Aubreville, Andreas Maier

To improve monaural speech enhancement in noisy environments, we propose CLCNet, a framework based on complex valued linear coding.

Speech Enhancement

An Investigation of Feature-based Nonrigid Image Registration using Gaussian Process

no code implementations12 Jan 2020 Siming Bayer, Ute Spiske, Jie Luo, Tobias Geimer, William M. Wells III, Martin Ostermeier, Rebecca Fahrig, Arya Nabavi, Christoph Bert, Ilker Eyupoglo, Andreas Maier

For a wide range of clinical applications, such as adaptive treatment planning or intraoperative image update, feature-based deformable registration (FDR) approaches are widely employed because of their simplicity and low computational complexity.

Image Registration

Limited Angle Tomography for Transmission X-Ray Microscopy Using Deep Learning

no code implementations8 Jan 2020 Yixing Huang, Shengxiang Wang, Yong Guan, Andreas Maier

Particularly, the U-Net, the state-of-the-art neural network in biomedical imaging, is trained from synthetic ellipsoid data and multi-category data to reduce artifacts in filtered back-projection (FBP) reconstruction images.

Denoising Image Reconstruction +1

Prediction of MRI Hardware Failures based on Image Features using Ensemble Learning

no code implementations5 Jan 2020 Nadine Kuhnert, Lea Pflüger, Andreas Maier

The other data level uses matrices which represent the overall coil condition and feeds a different neural network.

Ensemble Learning

COPD Classification in CT Images Using a 3D Convolutional Neural Network

no code implementations4 Jan 2020 Jalil Ahmed, Sulaiman Vesal, Felix Durlak, Rainer Kaergel, Nishant Ravikumar, Martine Remy-Jardin, Andreas Maier

Chronic obstructive pulmonary disease (COPD) is a lung disease that is not fully reversible and one of the leading causes of morbidity and mortality in the world.

Classification Computed Tomography (CT) +2

Analyzing an Imitation Learning Network for Fundus Image Registration Using a Divide-and-Conquer Approach

no code implementations19 Dec 2019 Siming Bayer, Xia Zhong, Weilin Fu, Nishant Ravikumar, Andreas Maier

In this work, we propose an imitation learning framework for the registration of 2D color funduscopic images for a wide range of applications such as disease monitoring, image stitching and super-resolution.

Image Registration Image Stitching +2

Deep Learning Algorithms for Coronary Artery Plaque Characterisation from CCTA Scans

no code implementations13 Dec 2019 Felix Denzinger, Michael Wels, Katharina Breininger, Anika Reidelshöfer, Joachim Eckert, Michael Sühling, Axel Schmermund, Andreas Maier

Analysing coronary artery plaque segments with respect to their functional significance and therefore their influence to patient management in a non-invasive setup is an important subject of current research.

Texture Classification

Deep Learning-based Denoising of Mammographic Images using Physics-driven Data Augmentation

no code implementations11 Dec 2019 Dominik Eckert, Sulaiman Vesal, Ludwig Ritschl, Steffen Kappler, Andreas Maier

In this study, we propose a deep learning method based on Convolutional Neural Networks (CNNs) for mammogram denoising to improve the image quality.

Data Augmentation Denoising

ICDAR 2019 Competition on Image Retrieval for Historical Handwritten Documents

1 code implementation8 Dec 2019 Vincent Christlein, Anguelos Nicolaou, Mathias Seuret, Dominique Stutzmann, Andreas Maier

This competition investigates the performance of large-scale retrieval of historical document images based on writing style.

Image Retrieval

Epoch-wise label attacks for robustness against label noise

no code implementations4 Dec 2019 Sebastian Guendel, Andreas Maier

With a simple one-class problem, the classification of tuberculosis, we measure the performance on a clean evaluation set when training with label-corrupt data.

Learning New Tricks from Old Dogs -- Inter-Species, Inter-Tissue Domain Adaptation for Mitotic Figure Assessment

no code implementations25 Nov 2019 Marc Aubreville, Christof A. Bertram, Samir Jabari, Christian Marzahl, Robert Klopfleisch, Andreas Maier

We were able to show that domain adversarial training considerably improves accuracy when applying mitotic figure classification learned from the canine on the human data sets (up to +12. 8% in accuracy) and is thus a helpful method to transfer knowledge from existing data sets to new tissue types and species.

Domain Adaptation

Projection-to-Projection Translation for Hybrid X-ray and Magnetic Resonance Imaging

no code implementations19 Nov 2019 Bernhard Stimpel, Christopher Syben, Tobias Würfl, Katharina Breininger, Philipp Hoelter, Arnd Dörfler, Andreas Maier

Additionally, a weighting scheme in the loss computation that favors high-frequency structures is proposed to focus on the important details and contours in projection imaging.

Image Enhancement Translation

Merging-ISP: Multi-Exposure High Dynamic Range Image Signal Processing

no code implementations12 Nov 2019 Prashant Chaudhari, Franziska Schirrmacher, Andreas Maier, Christian Riess, Thomas Köhler

As such, it is end-to-end trainable, circumvents the use of hand-crafted and potentially complex algorithms, and mitigates error propagation.

Demosaicking

What Do We Really Need? Degenerating U-Net on Retinal Vessel Segmentation

no code implementations6 Nov 2019 Weilin Fu, Katharina Breininger, Zhaoya Pan, Andreas Maier

Results show that for retinal vessel segmentation on DRIVE database, U-Net does not degenerate until surprisingly acute conditions: one level, one filter in convolutional layers, and one training sample.

Retinal Vessel Segmentation

Superpixel-Based Background Recovery from Multiple Images

no code implementations4 Nov 2019 Lei Gao, Yixing Huang, Andreas Maier

Background candidate images are obtained from input raw images with the masks.

Superpixels

Learning to Avoid Poor Images: Towards Task-aware C-arm Cone-beam CT Trajectories

no code implementations19 Sep 2019 Jan-Nico Zaech, Cong Gao, Bastian Bier, Russell Taylor, Andreas Maier, Nassir Navab, Mathias Unberath

Metal artifacts in computed tomography (CT) arise from a mismatch between physics of image formation and idealized assumptions during tomographic reconstruction.

Computed Tomography (CT)

Automated Multi-sequence Cardiac MRI Segmentation Using Supervised Domain Adaptation

no code implementations21 Aug 2019 Sulaiman Vesal, Nishant Ravikumar, Andreas Maier

We first train an encoder-decoder CNN on T2-weighted and balanced-Steady State Free Precession (bSSFP) MR images with pixel-level annotation and fine-tune the same network with a limited number of Late Gadolinium Enhanced-MR (LGE-MR) subjects, to adapt the domain features.

Domain Adaptation Left Ventricle Segmentation +3

Data Consistent Artifact Reduction for Limited Angle Tomography with Deep Learning Prior

no code implementations19 Aug 2019 Yixing Huang, Alexander Preuhs, Guenter Lauritsch, Michael Manhart, Xiaolin Huang, Andreas Maier

Robustness of deep learning methods for limited angle tomography is challenged by two major factors: a) due to insufficient training data the network may not generalize well to unseen data; b) deep learning methods are sensitive to noise.

Deep Generalized Max Pooling

1 code implementation14 Aug 2019 Vincent Christlein, Lukas Spranger, Mathias Seuret, Anguelos Nicolaou, Pavel Král, Andreas Maier

Global pooling layers are an essential part of Convolutional Neural Networks (CNN).

Multi-task Localization and Segmentation for X-ray Guided Planning in Knee Surgery

no code implementations24 Jul 2019 Florian Kordon, Peter Fischer, Maxim Privalov, Benedict Swartman, Marc Schnetzke, Jochen Franke, Ruxandra Lasowski, Andreas Maier, Holger Kunze

A deep multi-task stacked hourglass network is trained on 149 conventional lateral X-ray images to jointly localize two femoral landmarks, to predict a region of interest for the posterior femoral cortex tangent line, and to perform semantic segmentation of the femur, patella, tibia, and fibula with adaptive task complexity weighting.

Semantic Segmentation

A Divide-and-Conquer Approach towards Understanding Deep Networks

no code implementations14 Jul 2019 Weilin Fu, Katharina Breininger, Roman Schaffert, Nishant Ravikumar, Andreas Maier

We start with a high-performance U-Net and show by step-by-step conversion that we are able to divide the network into modules of known operators.

Retinal Vessel Segmentation Semantic Segmentation

RinQ Fingerprinting: Recurrence-informed Quantile Networks for Magnetic Resonance Fingerprinting

no code implementations9 Jul 2019 Elisabeth Hoppe, Florian Thamm, Gregor Körzdörfer, Christopher Syben, Franziska Schirrmacher, Mathias Nittka, Josef Pfeuffer, Heiko Meyer, Andreas Maier

Although the acquisition is highly accelerated, the state-of-the-art reconstruction suffers from long computation times: Template matching methods are used to find the most similar signal to the measured one by comparing it to pre-simulated signals of possible parameter combinations in a discretized dictionary.

Magnetic Resonance Fingerprinting Template Matching

Analysis by Adversarial Synthesis -- A Novel Approach for Speech Vocoding

no code implementations1 Jul 2019 Ahmed Mustafa, Arijit Biswas, Christian Bergler, Julia Schottenhamml, Andreas Maier

Recently, autoregressive deep generative models such as WaveNet and SampleRNN have been used as speech vocoders to scale up the perceptual quality of the reconstructed signals without increasing the coding rate.

A 2D dilated residual U-Net for multi-organ segmentation in thoracic CT

no code implementations19 May 2019 Sulaiman Vesal, Nishant Ravikumar, Andreas Maier

Automatic segmentation of organs-at-risk (OAR) in computed tomography (CT) is an essential part of planning effective treatment strategies to combat lung and esophageal cancer.

Computed Tomography (CT) Medical Image Segmentation +1

Multi-task Learning for Chest X-ray Abnormality Classification on Noisy Labels

no code implementations15 May 2019 Sebastian Guendel, Florin C. Ghesu, Sasa Grbic, Eli Gibson, Bogdan Georgescu, Andreas Maier, Dorin Comaniciu

Chest X-ray (CXR) is the most common X-ray examination performed in daily clinical practice for the diagnosis of various heart and lung abnormalities.

Classification General Classification +1

Dilated deeply supervised networks for hippocampus segmentation in MRI

1 code implementation20 Mar 2019 Lukas Folle, Sulaiman Vesal, Nishant Ravikumar, Andreas Maier

Tissue loss in the hippocampi has been heavily correlated with the progression of Alzheimer's Disease (AD).

Hippocampus Semantic Segmentation

A 3-D Projection Model for X-ray Dark-field Imaging

no code implementations11 Nov 2018 Shiyang Hu, Lina Felsner, Andreas Maier, Veronika Ludwig, Gisela Anton, Christian Riess

A key step of the reconstruction algorithm is the inversion of a forward projection model.

A Gentle Introduction to Deep Learning in Medical Image Processing

no code implementations12 Oct 2018 Andreas Maier, Christopher Syben, Tobias Lasser, Christian Riess

This paper tries to give a gentle introduction to deep learning in medical image processing, proceeding from theoretical foundations to applications.

Image Registration Semantic Segmentation

Dilated Convolutions in Neural Networks for Left Atrial Segmentation in 3D Gadolinium Enhanced-MRI

no code implementations5 Aug 2018 Sulaiman Vesal, Nishant Ravikumar, Andreas Maier

We employ a 3D fully convolutional network, with dilated convolutions in the lowest level of the network, and residual connections between encoder blocks to incorporate local and global knowledge.

Domain Adaptation Medical Image Segmentation +1

User Loss -- A Forced-Choice-Inspired Approach to Train Neural Networks directly by User Interaction

no code implementations24 Jul 2018 Shahab Zarei, Bernhard Stimpel, Christopher Syben, Andreas Maier

This approach opens the way towards implementation of direct user feedback in deep learning and is applicable for a wide range of application.

Image Denoising Medical Image Denoising

Deriving Neural Network Architectures using Precision Learning: Parallel-to-fan beam Conversion

no code implementations9 Jul 2018 Christopher Syben, Bernhard Stimpel, Jonathan Lommen, Tobias Würfl, Arnd Dörfler, Andreas Maier

The results demonstrate that the proposed method is superior to ray-by-ray interpolation and is able to deliver sharper images using the same amount of parallel-beam input projections which is crucial for interventional applications.

Adversarial and Perceptual Refinement for Compressed Sensing MRI Reconstruction

1 code implementation28 Jun 2018 Maximilian Seitzer, Guang Yang, Jo Schlemper, Ozan Oktay, Tobias Würfl, Vincent Christlein, Tom Wong, Raad Mohiaddin, David Firmin, Jennifer Keegan, Daniel Rueckert, Andreas Maier

In addition, we introduce a semantic interpretability score, measuring the visibility of the region of interest in both ground truth and reconstructed images, which allows us to objectively quantify the usefulness of the image quality