In the COVID-19 pandemic, a rigorous testing scheme was crucial. However, tests can be time-consuming and expensive. A machine learning-based diagnostic tool for audio recordings could enable widespread testing at low costs. In order to achieve comparability between such algorithms, the DiCOVA challenge was created. It is based on the Coswara dataset offering the recording categories cough, speech, breath and vowel phonation. Recording durations vary greatly, ranging from one second to over a minute. A base model is pre-trained on random, short time intervals. Subsequently, a Multiple Instance Learning (MIL) model based on self-attention is incorporated to make collective predictions for multiple time segments within each audio recording, taking advantage of longer durations. In order to compete in the fusion category of the DiCOVA challenge, we utilize a linear regression approach among other fusion methods to combine predictions from the most successful models associated with each sound modality. The application of the MIL approach significantly improves generalizability, leading to an AUC ROC score of 86.6% in the fusion category. By incorporating previously unused data, including the sound modality 'sustained vowel phonation' and patient metadata, we were able to significantly improve our previous results reaching a score of 92.2%.

While model architectures and training strategies have become more generic and flexible with respect to different data modalities over the past years, a persistent limitation lies in the assumption of fixed quantities and arrangements of input features. This limitation becomes particularly relevant in scenarios where the attributes captured during data acquisition vary across different samples. In this work, we aim at effectively leveraging data with varying features, without the need to constrain the input space to the intersection of potential feature sets or to expand it to their union. We propose a novel architecture that can directly process data without the necessity of aligned feature modalities by learning a general embedding space that captures the relationship between features across data samples with varying sets of features. This is achieved via a set-transformer architecture augmented by feature-encoder layers, thereby enabling the learning of a shared latent feature space from data originating from heterogeneous feature spaces. The advantages of the model are demonstrated for automatic cancer cell detection in acute myeloid leukemia in flow cytometry data, where the features measured during acquisition often vary between samples. Our proposed architecture’s capacity to operate seamlessly across incongruent feature spaces is particularly relevant in this context, where data scarcity arises from the low prevalence of the disease. The code is available for research purposes at https://github.com/lisaweijler/FATE.

In this study, we proposed a computer-aided diagnosis (CADx) framework under dual-energy spectral CT (DECT), which operates directly on the transmission data in the pre-log domain, called CADxDE, to explore the spectral information for lesion diagnosis. The CADxDE includes material identification and machine learning (ML) based CADx. Benefits from DECT's capability of performing virtual monoenergetic imaging with the identified materials, the responses of different tissue types (e.g., muscle, water, and fat) in lesions at each energy can be explored by ML for CADx. Without losing essential factors in the DECT scan, a pre-log domain model-based iterative reconstruction is adopted to obtain decomposed material images, which are then used to generate the virtual monoenergetic images (VMIs) at selected n energies. While these VMIs have the same anatomy, their contrast distribution patterns contain rich information along with the n energies for tissue characterization. Thus, a corresponding ML-based CADx is developed to exploit the energy-enhanced tissue features for differentiating malignant from benign lesions. Specifically, an original image-driven multi-channel three-dimensional convolutional neural network (CNN) and extracted lesion feature-based ML CADx methods are developed to show the feasibility of CADxDE. Results from three pathologically proven clinical datasets showed 4.01% to 14.25% higher AUC (area under the receiver operating characteristic curve) scores than the scores of both the conventional DECT data (high and low energy spectrum separately) and the conventional CT data. The mean gain >9.13% in AUC scores indicated that the energy spectral-enhanced tissue features from CADxDE have great potential to improve lesion diagnosis performance.

We present an approach to detecting small cell populations in flow cytometry (FCM) samples based on the combination of unsupervised manifold embedding and supervised random forest classification. Each sample consists of hundred thousands to a million cells where each cell typically corresponds to a measurement vector with 10 to 50 dimensions. The difficulty of the task is that clusters of measurement vectors formed in the data space according to standard clustering criteria often do not correspond to biologically meaningful sub-populations of cells, due to strong variations in shape and size of their distributions. In many cases the relevant population consists of less than 100 scattered events out of millions of events, where supervised approaches perform better than unsupervised clustering. The aim of this paper is to demonstrate that the performance of the standard supervised classifier can be improved significantly by combining it with a preceding unsupervised learning step involving the Uniform Manifold Approximation and Projection (UMAP). We present an experimental evaluation on FCM data from children suffering from Acute Lymphoblastic Leukemia (ALL) showing that the improvement particularly occurs in difficult samples where the size of the relevant population of leukemic cells is low in relation to other sub-populations. We show that the positive effect of the UMAP becomes more noticeable for smaller training sets. Further, the experiments indicate that in this situation the algorithm also outperforms other baseline methods based on Gaussian Mixture Models.

Acute Myeloid Leukaemia (AML) is a rare type of childhood acute leukaemia. During treatment, the assessment of the number of cancer cells is particularly important to determine treatment response and consequently adapt the treatment scheme if necessary. Minimal Residual Disease (MRD) is a diagnostic measure based on Flow CytoMetry (FCM) data that captures the amount of blasts in a blood sample and is a clinical tool for planning patients' individual therapy, which requires reliable blast identification. In this work we propose a novel semi-supervised learning approach, which is acquired whenever large amounts of unlabeled data and only a small amount of annotated data is available. The proposed semi-supervised learning approach is based on Wasserstein Generative Adversarial Network (WGAN) latent space embeddings learned in an unsupervised fashion and a simple Fully connected Neural Network (FNN) trained on labeled data leveraging the learned embedding. We apply our proposed learning approach for semi-supervised classification of blasts vs. non-blasts. We compare our approach with two baseline approaches, 1) semi-supervised learning based on Principal Component Analysis (PCA) embedding, and 2) a deep FNN that is trained only on the annotated data without leveraging an embedding. Results suggest that our proposed semi-supervised WGAN embedding outperforms semi-supervised learning based on PCA embeddings and if only small amounts of annotated data is available it even outperforms an FNN classifier.

Minimal Residual Disease (MRD) is a powerful risk-stratification marker in the treatment of primary and relapsed childhood Acute Lymphoblastic Leukemia. Flow Cytometry is a fast and sensitive method to detect MRD, however, the interpretation of these multi-parametric data demands intensive operator training and experience.

We propose a user-driven method for the segmentation of neuroblastoma nuclei in microscopic fluorescence images involving the gradient energy tensor . Multispectral fluorescence images contain intensity and spatial information about antigene expression, fluorescence in situ hybridization (FISH) signals and nucleus morphology. The latter serves as basis for the detection of single cells and the calculation of shape features, which are used to validate the segmentation and to reject false detections. Accurate segmentation is difficult due to varying staining intensities and aggregated cells. It requires several (meta-) parameters, which have a strong influence on the segmentation results and have to be selected carefully for each sample (or group of similar samples) by user interactions. Because our method is designed for clinicians and biologists, who may have only limited image processing background, an interactive parameter selection step allows the implicit tuning of parameter values. With this simple but intuitive method, segmentation results with high precision for a large number of cells can be achieved by minimal user interaction. The strategy was validated on handsegmented datasets of three neuroblastoma cell lines.

Simulations are characterized by long running calculations and complex data handling tasks accompanied by non-trivial data dependencies. The workflow technology helps to automate and steer such simulations. Quality of Data frameworks are used to determine the goodness of simulation data, e.g., they analyze the accuracy of input data with regards to the usability within numerical solvers. In this paper, we present generic approaches using evaluated Quality of Data to steer simulation workflows. This allows for ensuring that the predefined requirements such as a precise final result or a short execution time will be met even after the execution of simulation workflow has been started. We discuss mechanisms for steering a simulation on all relevant levels - workflow, service, algorithms, and define a unifying approach to control such workflows. To realize Quality of Data-driven workflows, we present an architecture realizing the presented approach and a WS-Policy-based language to describe Quality of Data requirements and capabilities.

In recent years scientific workflows have been used for conducting data-intensive and long running simulations. Such simulation workflows have processed and produced different types of data whose quality has a strong influence on the final outcome of simulations. Therefore being able to monitor and analyze quality of this data during workflow execution is of paramount importance, as detection of quality problems will enable us to control the execution of simulations efficiently. Unfortunately, existing scientific workflow execution systems do not support the monitoring and analysis of quality of data for multi-scale or multi-domain simulations. In this paper, we examine how quality of data can be comprehensively measured within workflows and how the measured quality can be used to control and adapt running workflows. We present a quality of data measurement process and describe a quality of data monitoring and analysis framework that integrates this measurement process into a workflow management system.

Multi-material components, which contain metal parts surrounded by plastic materials, are highly interesting for inspection using industrial 3D X-ray computed tomography (3DXCT). Examples of this application scenario are connectors or housings with metal inlays in the electronic or automotive industry. A major problem of this type of components is the presence of metal, which causes streaking artifacts and distorts the surrounding media in the reconstructed volume. Streaking artifacts and dark-band artifacts around metal components significantly influence the material characterization (especially for the plastic components). In specific cases these artifacts even prevent a further analysis. Due to the nature and the different characteristics of artifacts, the development of an efficient artifact-reduction technique in reconstruction-space is rather complicated. In this paper we present a projection-space pipeline for metal-artifacts reduction. The proposed technique first segments the metal in the spatial domain of the reconstructed volume in order to separate it from the other materials. Then metal parts are forward-projected on the set of projections in a way that metal-projection regions are treated as voids. Subsequently the voids, which are left by the removed metal, are interpolated in the 2D projections. Finally, the metal is inserted back into the reconstructed 3D volume during the fusion stage. We present a visual analysis tool, allowing for interactive parameter estimation of the metal segmentation. The results of the proposed artifact-reduction technique are demonstrated on a test part as well as on real world components. For these specimens we achieve a significant reduction of metal artifacts, allowing an enhanced material characterization.

Industrial cone-beam X-Ray computed tomography (CT) systems often face problems due to artifacts caused by a bad placement of the specimen on the rotary plate. This paper presents a visual-analysis tool for CT systems, which provides a simulation-based preview and estimates artifacts and deviations of a specimen's placement using the corresponding 3D geometrical surface model as input. The presented tool identifies potentially good or bad placements of a specimen and regions of a specimen, which cause the major portion of artefacts. The tool can be used for a preliminary analysis of the specimen before CT scanning, in order to determine the optimal way of placing the object. The analysis includes: penetration lengths, placement stability and an investigation in Radon space. Novel visualization techniques are applied to the simulation data. A stability widget is presented for determining the placement parameters' robustness. The performance and the comparison of results provided by the tool compared with real world data is demonstrated using two specimens.

A fast AAM search algorithm based on canonical correlation analysis (CCA-AAM) is introduced. It efficiently models the dependency between texture residuals and model parameters during search. Experiments show that CCA-AAMs, while requiring similar implementation effort, consistently outperform standard search with regard to convergence speed by a factor of four.

We present a method allowing a significant speed-up of the eigen-detection method (detection based on principle component analysis). We derive a formula for an upper bound on the class-conditional probability (or equivalently a lower bound on the Mahalanobis distance) on which detection is based. Often, the lower bound of Mahalanobis distance (MD) reaches a preset threshold after computation of only a few eigen-projections. In this case the computation of MD can be immediately terminated. Regardless of the precise value of MD, the detection hypothesis (object from class /spl Omega/ is detected) can be rejected. While provably obtaining results identical to the standard technique, we achieved a two- to three-fold speed-up in face detection experiments on images from the CMU database.