arXiv subjects

Andreas Maier

Publications and source records attributed to Andreas Maier.

At least 469 records · Page 26Linked to original sources

arXiv2018

Augmented Mitotic Cell Count using Field Of Interest Proposal

Histopathological prognostication of neoplasia including most tumor grading systems are based upon a number of criteria. Probably the most important is the number of mitotic figures which are most commonly determined as the mitotic count (MC), i.e. number of mitotic figures within 10 consecutive high power fields. Often the area with the highest mitotic activity is to be selected for the MC. However, since mitotic activity is not known in advance, an arbitrary choice of this region is considered one important cause for high variability in the prognostication and grading. In this work, we present an algorithmic approach that first calculates a mitotic cell map based upon a deep convolutional network. This map is in a second step used to construct a mitotic activity estimate. Lastly, we select the image segment representing the size of ten high power fields with the overall highest mitotic activity as a region proposal for an expert MC determination. We evaluate the approach using a dataset of 32 completely annotated whole slide images, where 22 were used for training of the network and 10 for test. We find a correlation of r=0.936 in mitotic count estimate.

cs.CV↗

arXiv2018

A Multi-task Framework for Skin Lesion Detection and Segmentation

Early detection and segmentation of skin lesions is crucial for timely diagnosis and treatment, necessary to improve the survival rate of patients. However, manual delineation is time consuming and subject to intra- and inter-observer variations among dermatologists. This underlines the need for an accurate and automatic approach to skin lesion segmentation. To tackle this issue, we propose a multi-task convolutional neural network (CNN) based, joint detection and segmentation framework, designed to initially localize the lesion and subsequently, segment it. A `Faster region-based convolutional neural network' (Faster-RCNN) which comprises a region proposal network (RPN), is used to generate bounding boxes/region proposals, for lesion localization in each image. The proposed regions are subsequently refined using a softmax classifier and a bounding-box regressor. The refined bounding boxes are finally cropped and segmented using `SkinNet', a modified version of U-Net. We trained and evaluated the performance of our network, using the ISBI 2017 challenge and the PH2 datasets, and compared it with the state-of-the-art, using the official test data released as part of the challenge for the former. Our approach outperformed others in terms of Dice coefficients ($>0.93$), Jaccard index ($>0.88$), accuracy ($>0.96$) and sensitivity ($>0.95$), across five-fold cross validation experiments.

cs.CV↗

arXiv2018

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

Segmentation of the left atrial chamber and assessing its morphology, are essential for improving our understanding of atrial fibrillation, the most common type of cardiac arrhythmia. Automation of this process in 3D gadolinium enhanced-MRI (GE-MRI) data is desirable, as manual delineation is time-consuming, challenging and observer-dependent. Recently, deep convolutional neural networks (CNNs) have gained tremendous traction and achieved state-of-the-art results in medical image segmentation. However, it is difficult to incorporate local and global information without using contracting (pooling) layers, which in turn reduces segmentation accuracy for smaller structures. In this paper, we propose a 3D CNN for volumetric segmentation of the left atrial chamber in LGE-MRI. Our network is based on the well known U-Net architecture. 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. The results show that including global context through the use of dilated convolutions, helps in domain adaptation, and the overall segmentation accuracy is improved in comparison to a 3D U-Net.

cs.CV↗

arXiv2018

Viewpoint Planning for Quantitative Coronary Angiography

In coronary angiography the condition of myocardial blood supply is assessed by analyzing 2-D X-ray projections of contrasted coronary arteries. This is done using a flexible C-arm system. Due to the X-ray immanent dimensionality reduction projecting the 3-D scene onto a 2-D image, the viewpoint is critical to guarantee an appropriate view onto the affected artery and, thus, enable reliable diagnosis. In this work we introduce an algorithm computing optimal viewpoints for the assessment of coronary arteries without the need for 3-D models. We introduce the concept of optimal viewpoint planning solely based on a single angiographic X-ray image. The subsequent viewpoint is computed such that it is rotated precisely around a vessel, while minimizing foreshortening. Our algorithm reduces foreshortening substantially compared to the input view and completely eliminates it for 90 degree rotations. Rotations around iso-centered foreshortening-free vessels passing the isocenter are exact. The precision, however, decreases when the vessel is off-centered or foreshortened. We evaluate worst case boundaries, providing insight in the maximal inaccuracies to be expected. This can be utilized to design viewpoints guaranteeing desired requirements, e.g. a true rotation around the vessel of at minimum 30 degree.

physics.med-ph↗

arXiv2018

Fast Epipolar Consistency without the Need for Pseudo Matrix Inverses

Interventional C-arm systems allow flexible 2-D imaging of a 3-D scene while being capable of cone beam computed tomography. Due to the flexible structure of the C-arm, the rotation speed is limited, increasing the acquisition time compared to conventional computed tomography. Therefore, patient motion frequently occurs during data acquisition inducing inconsistencies in the projection raw data. A framework using Grangeat's theorem and epipolar consistency was successfully applied for compensating rigid motion. This algorithm was efficiently parallelized, however, before each iteration, the pseudoinverse of each projection matrix must be calculated. We present a geometric modification of the presented algorithm which can be used without a pseudo-inverse. As such, the complete algorithm can be implemented for low-level hardware without the need of a linear algebra package that supports the calculation of matrix inverse. Both algorithms are applied for head motion compensation and the runtime of both is compared.

physics.med-ph↗

arXiv2018

Towards Arbitrary Noise Augmentation - Deep Learning for Sampling from Arbitrary Probability Distributions

Accurate noise modelling is important for training of deep learning reconstruction algorithms. While noise models are well known for traditional imaging techniques, the noise distribution of a novel sensor may be difficult to determine a priori. Therefore, we propose learning arbitrary noise distributions. To do so, this paper proposes a fully connected neural network model to map samples from a uniform distribution to samples of any explicitly known probability density function. During the training, the Jensen-Shannon divergence between the distribution of the model's output and the target distribution is minimized. We experimentally demonstrate that our model converges towards the desired state. It provides an alternative to existing sampling methods such as inversion sampling, rejection sampling, Gaussian mixture models and Markov-Chain-Monte-Carlo. Our model has high sampling efficiency and is easily applied to any probability distribution, without the need of further analytical or numerical calculations.

cs.LG↗

arXiv2018

Precision Learning: Reconstruction Filter Kernel Discretization

In this paper, we present substantial evidence that a deep neural network will intrinsically learn the appropriate way to discretize the ideal continuous reconstruction filter. Currently, the Ram-Lak filter or heuristic filters which impose different noise assumptions are used for filtered back-projection. All of these, however, inhibit a fully data-driven reconstruction deep learning approach. In addition, the heuristic filters are not chosen in an optimal sense. To tackle this issue, we propose a formulation to directly learn the reconstruction filter. The filter is initialized with the ideal Ramp filter as a strong pre-training and learned in frequency domain. We compare the learned filter with the Ram-Lak and the Ramp filter on a numerical phantom as well as on a real CT dataset. The results show that the network properly discretizes the continuous Ramp filter and converges towards the Ram-Lak solution. In our view these observations are interesting to gain a better understanding of deep learning techniques and traditional analytic techniques such as Wiener filtering and discretization theory. Furthermore, this will allow fully trainable data-driven reconstruction deep learning approaches.

cs.CV↗

arXiv2018

Adversarial and Perceptual Refinement for Compressed Sensing MRI Reconstruction

Deep learning approaches have shown promising performance for compressed sensing-based Magnetic Resonance Imaging. While deep neural networks trained with mean squared error (MSE) loss functions can achieve high peak signal to noise ratio, the reconstructed images are often blurry and lack sharp details, especially for higher undersampling rates. Recently, adversarial and perceptual loss functions have been shown to achieve more visually appealing results. However, it remains an open question how to (1) optimally combine these loss functions with the MSE loss function and (2) evaluate such a perceptual enhancement. In this work, we propose a hybrid method, in which a visual refinement component is learnt on top of an MSE loss-based reconstruction network. 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 for image post-processing and analysis. Applied on a large cardiac MRI dataset simulated with 8-fold undersampling, we demonstrate significant improvements ($p<0.01$) over the state-of-the-art in both a human observer study and the semantic interpretability score.

cs.CV↗

arXiv2018

SkinNet: A Deep Learning Framework for Skin Lesion Segmentation

There has been a steady increase in the incidence of skin cancer worldwide, with a high rate of mortality. Early detection and segmentation of skin lesions are crucial for timely diagnosis and treatment, necessary to improve the survival rate of patients. However, skin lesion segmentation is a challenging task due to the low contrast of lesions and their high similarity in terms of appearance, to healthy tissue. This underlines the need for an accurate and automatic approach for skin lesion segmentation. To tackle this issue, we propose a convolutional neural network (CNN) called SkinNet. The proposed CNN is a modified version of U-Net. We compared the performance of our approach with other state-of-the-art techniques, using the ISBI 2017 challenge dataset. Our approach outperformed the others in terms of the Dice coefficient, Jaccard index and sensitivity, evaluated on the held-out challenge test data set, across 5-fold cross validation experiments. SkinNet achieved an average value of 85.10, 76.67 and 93.0%, for the DC, JI, and SE, respectively.

cs.CV↗

arXiv2018

Augmented Reality-based Feedback for Technician-in-the-loop C-arm Repositioning

Interventional C-arm imaging is crucial to percutaneous orthopedic procedures as it enables the surgeon to monitor the progress of surgery on the anatomy level. Minimally invasive interventions require repeated acquisition of X-ray images from different anatomical views to verify tool placement. Achieving and reproducing these views often comes at the cost of increased surgical time and radiation dose to both patient and staff. This work proposes a marker-free "technician-in-the-loop" Augmented Reality (AR) solution for C-arm repositioning. The X-ray technician operating the C-arm interventionally is equipped with a head-mounted display capable of recording desired C-arm poses in 3D via an integrated infrared sensor. For C-arm repositioning to a particular target view, the recorded C-arm pose is restored as a virtual object and visualized in an AR environment, serving as a perceptual reference for the technician. We conduct experiments in a setting simulating orthopedic trauma surgery. Our proof-of-principle findings indicate that the proposed system can decrease the 2.76 X-ray images required per desired view down to zero, suggesting substantial reductions of radiation dose during C-arm repositioning. The proposed AR solution is a first step towards facilitating communication between the surgeon and the surgical staff, improving the quality of surgical image acquisition, and enabling context-aware guidance for surgery rooms of the future. The concept of technician-in-the-loop design will become relevant to various interventions considering the expected advancements of sensing and wearable computing in the near future.

cs.CV↗

arXiv2018

Action Learning for 3D Point Cloud Based Organ Segmentation

We propose a novel point cloud based 3D organ segmentation pipeline utilizing deep Q-learning. In order to preserve shape properties, the learning process is guided using a statistical shape model. The trained agent directly predicts piece-wise linear transformations for all vertices in each iteration. This mapping between the ideal transformation for an object outline estimation is learned based on image features. To this end, we introduce aperture features that extract gray values by sampling the 3D volume within the cone centered around the associated vertex and its normal vector. Our approach is also capable of estimating a hierarchical pyramid of non rigid deformations for multi-resolution meshes. In the application phase, we use a marginal approach to gradually estimate affine as well as non-rigid transformations. We performed extensive evaluations to highlight the robust performance of our approach on a variety of challenge data as well as clinical data. Additionally, our method has a run time ranging from 0.3 to 2.7 seconds to segment each organ. In addition, we show that the proposed method can be applied to different organs, X-ray based modalities, and scanning protocols without the need of transfer learning. As we learn actions, even unseen reference meshes can be processed as demonstrated in an example with the Visible Human. From this we conclude that our method is robust, and we believe that our method can be successfully applied to many more applications, in particular, in the interventional imaging space.

cs.CV↗

arXiv2018

Double Your Views - Exploiting Symmetry in Transmission Imaging

For a plane symmetric object we can find two views - mirrored at the plane of symmetry - that will yield the exact same image of that object. In consequence, having one image of a plane symmetric object and a calibrated camera, we can automatically have a second, virtual image of that object if the 3-D location of the symmetry plane is known. In this work, we show for the first time that the above concept naturally extends to transmission imaging and present an algorithm to estimate the 3-D symmetry plane from a set of projection domain images based on Grangeat's theorem. We then exploit symmetry to generate a virtual trajectory by mirroring views at the plane of symmetry. If the plane is not perpendicular to the acquired trajectory plane, the virtual and real trajectory will be oblique. The resulting X-shaped trajectory will be data-complete, allowing for the compensation of in-plane motion using epipolar consistency. We evaluate the proposed method on a synthetic symmetric phantom and, in a proof-of-concept study, apply it to a real scan of an anthropomorphic human head phantom.

physics.med-ph↗

arXiv2018

Hyper-Hue and EMAP on Hyperspectral Images for Supervised Layer Decomposition of Old Master Drawings

Old master drawings were mostly created step by step in several layers using different materials. To art historians and restorers, examination of these layers brings various insights into the artistic work process and helps to answer questions about the object, its attribution and its authenticity. However, these layers typically overlap and are oftentimes difficult to differentiate with the unaided eye. For example, a common layer combination is red chalk under ink. In this work, we propose an image processing pipeline that operates on hyperspectral images to separate such layers. Using this pipeline, we show that hyperspectral images enable better layer separation than RGB images, and that spectral focus stacking aids the layer separation. In particular, we propose to use two descriptors in hyperspectral historical document analysis, namely hyper-hue and extended multi-attribute profile (EMAP). Our comparative results with other features underline the efficacy of the three proposed improvements.

cs.CV↗

arXiv2018

Phase-Sensitive Region-of-Interest Computed Tomography

X-Ray Phase-Contrast Imaging (PCI) yields absorption, differential phase, and dark-field images. Computed Tomography (CT) of grating-based PCI can in principle provide high-resolution soft-tissue contrast. Recently, grating-based PCI took several hurdles towards clinical implementation by addressing, for example, acquisition speed, high X-ray energies, and system vibrations. However, a critical impediment in all grating-based systems lies in limits that constrain the grating diameter to few centimeters. In this work, we propose a system and a reconstruction algorithm to circumvent this constraint in a clinically compatible way. We propose to perform a phase-sensitive Region-of-Interest (ROI) CT within a full-field absorption CT. The biggest advantage of this approach is that it allows to correct for phase truncation artifacts, and to obtain quantitative phase values. Our method is robust, and shows high-quality results on simulated data and on a biological mouse sample. This work is a proof of concept showing the potential to use PCI in CT on large specimen, such as humans, in clinical applications.

physics.med-ph↗

arXiv2018

Higgs-boson plus Dijets: Higher-Order Matching for High-Energy Predictions

Several important processes and analyses at the LHC are sensitive to higher-order perturbative corrections beyond what can currently be calculated at fixed order. The formalism of High Energy Jets (HEJ) calculates the corrections systematically enhanced for a large ratio of the centre-of-mass energy to the transverse momentum of the observed jets. These effects are relevant in the analysis of e.g. Higgs-boson production in association with dijets within the cuts devised to enhance the contribution from Vector Boson Fusion (VBF). HEJ obtains an all-order approximation, based on logarithmic corrections which are matched to fixed-order results in the cases where these can be readily evaluated. In this paper we present an improved framework for the matching utilised in HEJ, which for merging of tree-level results is mathematically equivalent to the one used so far. However, by starting from events generated at fixed order and supplementing these with the all-order summation, it is computationally simpler to obtain matching to calculations of high multiplicity. We demonstrate that the impact of the higher-multiplicity matching on predictions is small for the gluon-fusion (GF) contribution of Higgs-boson production in association with dijets in the VBF-region, so perturbative stability against high-multiplicity matching has been achieved within HEJ. We match the improved HEJ prediction to the inclusive next-to-leading order (NLO) cross section and compare to pure NLO in the h->photon photon channel with standard VBF cuts.

hep-ph↗

arXiv2018

Motion Artifact Detection in Confocal Laser Endomicroscopy Images

Confocal Laser Endomicroscopy (CLE), an optical imaging technique allowing non-invasive examination of the mucosa on a (sub)cellular level, has proven to be a valuable diagnostic tool in gastroenterology and shows promising results in various anatomical regions including the oral cavity. Recently, the feasibility of automatic carcinoma detection for CLE images of sufficient quality was shown. However, in real world data sets a high amount of CLE images is corrupted by artifacts. Amongst the most prevalent artifact types are motion-induced image deteriorations. In the scope of this work, algorithmic approaches for the automatic detection of motion artifact-tainted image regions were developed. Hence, this work provides an important step towards clinical applicability of automatic carcinoma detection. Both, conventional machine learning and novel, deep learning-based approaches were assessed. The deep learning-based approach outperforms the conventional approaches, attaining an AUC of 0.90.

cs.CV↗

arXiv2018

Deep Denoising for Hearing Aid Applications

Reduction of unwanted environmental noises is an important feature of today's hearing aids (HA), which is why noise reduction is nowadays included in almost every commercially available device. The majority of these algorithms, however, is restricted to the reduction of stationary noises. In this work, we propose a denoising approach based on a three hidden layer fully connected deep learning network that aims to predict a Wiener filtering gain with an asymmetric input context, enabling real-time applications with high constraints on signal delay. The approach is employing a hearing instrument-grade filter bank and complies with typical hearing aid demands, such as low latency and on-line processing. It can further be well integrated with other algorithms in an existing HA signal processing chain. We can show on a database of real world noise signals that our algorithm is able to outperform a state of the art baseline approach, both using objective metrics and subject tests.

eess.AS↗

arXiv2018

Life of Π

The heavy-quark contribution to the polarisation function $Π$ at higher perturbative orders is presently only known approximately. We scrutinise the accuracy of state-of-the-art approximations at three- and four-loop order. At three loops, we present for the first time a result with arbitrary numerical precision for general kinematics and compare to the best Padé estimate. At four loops, we calculate the fourth (inverse) moment of the non-singlet heavy-quark vacuum polarisation in order to test the prediction for this moment based on Padé approximation.

hep-ph↗