Search

Supplementary Tables 1-3, Figures 1-6. Supplementary Table 1: Statistical analysis of differences between the study group and the remaining group of patients for several variables. Supplementary Table 2: Primers used for the RT-qPCR assays. Supplementary Table 3: Leave-one-out cross-validation of the angiogenesis-associated and hypoxia gene signatures. Supplementary Figure 1: Average gene expression levels for the top 10-ranked angiogenesis signature calculated based on the microarray data (N = 42 patients) for low (Cluster A1), high (Cluster A2) and medium (Cluster A3) risk patients. Supplementary Figure 2: Average gene expression levels for the top 10-ranked hypoxia response signature calculated based on the microarray data (N = 42 patients). Supplementary Figure 3: Hierarchical clustering of top 10-ranked hypoxia-response genes significantly associated with TS, showing gene expression variation among patients. The heat map represents relative intensity values of gene expression levels. Supplementary Figure 4: Hierarchical clustering of top 10-ranked hypoxia-response genes significantly associated with OS, showing gene expression variation among patients. The heat map represents relative intensity values of gene expression levels. Supplementary Figure 5: Optimized between-group classification (OBC) sensitivity analysis for angiogenesis-associated signature (A) and hypoxia-response signature predictive of TTP under BE. Supplementary Figure 6: Association between the expression of 10-gene hypoxia-response signature predictive of TTP under BE in our study, and PFS of second line sorafenib-treated patients enrolled in the BATTLE-1 trial,1 GEO accession number: GSE33072;2 P = 0.0186, calculated by log-tank test.

Purpose: We aimed to identify gene expression signatures associated with angiogenesis and hypoxia pathways with predictive value for treatment response to bevacizumab/erlotinib (BE) of nonsquamous advanced non–small cell lung cancer (NSCLC) patients.Experimental Design: Whole-genome gene expression profiling was performed on 42 biopsy samples (from SAKK 19/05 trial) using Affymetrix exon arrays, and associations with the following endpoints: time-to-progression (TTP) under therapy, tumor-shrinkage (TS), and overall survival (OS) were investigated. Next, we performed gene set enrichment analyses using genes associated with the angiogenic process and hypoxia response to evaluate their predictive value for patients' outcome.Results: Our analysis revealed that both the angiogenic and hypoxia response signatures were enriched within the genes predictive of BE response, TS, and OS. Higher gene expression levels (GEL) of the 10-gene angiogenesis-associated signature and lower levels of the 10-gene hypoxia response signature predicted improved TTP under BE, 7.1 months versus 2.1 months for low versus high-risk patients (P = 0.005), and median TTP 6.9 months versus 2.9 months (P = 0.016), respectively. The hypoxia response signature associated with higher TS at 12 weeks and improved OS (17.8 months vs. 9.9 months for low vs. high-risk patients, P = 0.001).Conclusions: We were able to identify gene expression signatures derived from the angiogenesis and hypoxia response pathways with predictive value for clinical outcome in advanced nonsquamous NSCLC patients. This could lead to the identification of clinically relevant biomarkers, which will allow for selecting the subset of patients who benefit from the treatment and predict drug response. Clin Cancer Res; 21(23); 5253–63. ©2015 AACR.See related commentary by Cascone and Heymach, p. 5188

In many medical applications, interpretable models with high prediction performance are sought. Often, those models are required to handle semi-structured data like tabular and image data. We show how to apply deep transformation models (DTMs) for distributional regression which fulfill these requirements. DTMs allow the data analyst to specify (deep) neural networks for different input modalities making them applicable to various research questions. Like statistical models, DTMs can provide interpretable effect estimates while achieving the state-of-the-art prediction performance of deep neural networks. In addition, the construction of ensembles of DTMs that retain model structure and interpretability allows quantifying epistemic and aleatoric uncertainty. In this study, we compare several DTMs, including baseline-adjusted models, trained on a semi-structured data set of 407 stroke patients with the aim to predict ordinal functional outcome three months after stroke. We follow statistical principles of model-building to achieve an adequate trade-off between interpretability and flexibility while assessing the relative importance of the involved data modalities. We evaluate the models for an ordinal and dichotomized version of the outcome as used in clinical practice. We show that both, tabular clinical and brain imaging data, are useful for functional outcome prediction, while models based on tabular data only outperform those based on imaging data only. There is no substantial evidence for improved prediction when combining both data modalities. Overall, we highlight that DTMs provide a powerful, interpretable approach to analyzing semi-structured data and that they have the potential to support clinical decision making., Preprint under review

Study design Retrospective, mono-centric cohort research study. Objectives The purpose of this study is to validate a novel artificial intelligence (AI)-based algorithm against human-generated ground truth for radiographic parameters of adolescent idiopathic scoliosis (AIS). Methods An AI-algorithm was developed that is capable of detecting anatomical structures of interest (clavicles, cervical, thoracic, lumbar spine and sacrum) and calculate essential radiographic parameters in AP spine X-rays fully automatically. The evaluated parameters included T1-tilt, clavicle angle (CA), coronal balance (CB), lumbar modifier, and Cobb angles in the proximal thoracic (C-PT), thoracic, and thoracolumbar regions. Measurements from 2 experienced physicians on 100 preoperative AP full spine X-rays of AIS patients were used as ground truth and to evaluate inter-rater and intra-rater reliability. The agreement between human raters and AI was compared by means of single measure Intra-class Correlation Coefficients (ICC; absolute agreement; >.75 rated as excellent), mean error and additional statistical metrics. Results The comparison between human raters resulted in excellent ICC values for intra- (range: .97-1) and inter-rater (.85-.99) reliability. The algorithm was able to determine all parameters in 100% of images with excellent ICC values (.78-.98). Consistently with the human raters, ICC values were typically smallest for C-PT (eg, rater 1A vs AI: .78, mean error: 4.7°) and largest for CB (.96, -.5 mm) as well as CA (.98, .2°). Conclusions The AI-algorithm shows excellent reliability and agreement with human raters for coronal parameters in preoperative full spine images. The reliability and speed offered by the AI-algorithm could contribute to the efficient analysis of large datasets (eg, registry studies) and measurements in clinical practice.

Rheumatoid arthritis is an autoimmune disease that causes chronic inflammation of synovial joints, often resulting in irreversible structural damage. The activity of the disease is evaluated by clinical examinations, laboratory tests, and patient self-assessment. The long-term course of the disease is assessed with radiographs of hands and feet. The evaluation of the X-ray images performed by trained medical staff requires several minutes per patient. We demonstrate that deep convolutional neural networks can be leveraged for a fully automated, fast, and reproducible scoring of X-ray images of patients with rheumatoid arthritis. A comparison of the predictions of different human experts and our deep learning system shows that there is no significant difference in the performance of human experts and our deep learning model.

Outcomes with a natural order commonly occur in prediction tasks and often the available input data are a mixture of complex data like images and tabular predictors. Deep Learning (DL) models are state-of-the-art for image classification tasks but frequently treat ordinal outcomes as unordered and lack interpretability. In contrast, classical ordinal regression models consider the outcome's order and yield interpretable predictor effects but are limited to tabular data. We present ordinal neural network transformation models (ONTRAMs), which unite DL with classical ordinal regression approaches. ONTRAMs are a special case of transformation models and trade off flexibility and interpretability by additively decomposing the transformation function into terms for image and tabular data using jointly trained neural networks. The performance of the most flexible ONTRAM is by definition equivalent to a standard multi-class DL model trained with cross-entropy while being faster in training when facing ordinal outcomes. Lastly, we discuss how to interpret model components for both tabular and image data on two publicly available datasets., 41 pages (incl. appendix, figures and literature), 11 figures in main text, 4 figures in appendix

We present a deep transformation model for probabilistic regression. Deep learning is known for outstandingly accurate predictions on complex data but in regression tasks, it is predominantly used to just predict a single number. This ignores the non-deterministic character of most tasks. Especially if crucial decisions are based on the predictions, like in medical applications, it is essential to quantify the prediction uncertainty. The presented deep learning transformation model estimates the whole conditional probability distribution, which is the most thorough way to capture uncertainty about the outcome. We combine ideas from a statistical transformation model (most likely transformation) with recent transformation models from deep learning (normalizing flows) to predict complex outcome distributions. The core of the method is a parameterized transformation function which can be trained with the usual maximum likelihood framework using gradient descent. The method can be combined with existing deep learning architectures. For small machine learning benchmark datasets, we report state of the art performance for most dataset and partly even outperform it. Our method works for complex input data, which we demonstrate by employing a CNN architecture on image data.

At present, the majority of the proposed Deep Learning (DL) methods provide point predictions without quantifying the models uncertainty. However, a quantification of the reliability of automated image analysis is essential, in particular in medicine when physicians rely on the results for making critical treatment decisions. In this work, we provide an entire framework to diagnose ischemic stroke patients incorporating Bayesian uncertainty into the analysis procedure. We present a Bayesian Convolutional Neural Network (CNN) yielding a probability for a stroke lesion on 2D Magnetic Resonance (MR) images with corresponding uncertainty information about the reliability of the prediction. For patient-level diagnoses, different aggregation methods are proposed and evaluated, which combine the single image-level predictions. Those methods take advantage of the uncertainty in image predictions and report model uncertainty at the patient-level. In a cohort of 511 patients, our Bayesian CNN achieved an accuracy of 95.33% at the image-level representing a significant improvement of 2% over a non-Bayesian counterpart. The best patient aggregation method yielded 95.89% of accuracy. Integrating uncertainty information about image predictions in aggregation models resulted in higher uncertainty measures to false patient classifications, which enabled to filter critical patient diagnoses that are supposed to be closer examined by a medical doctor. We therefore recommend using Bayesian approaches not only for improved image-level prediction and uncertainty estimation but also for the detection of uncertain aggregations at the patient-level., Comment: 21 pages, 13 figures

Deep learning (DL) methods have gained considerable attention since 2014. In this chapter we briefly review the state of the art in DL and then give several examples of applications from diverse areas of application. We will focus on convolutional neural networks (CNNs), which have since the seminal work of Krizhevsky et al. (ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems 25, pp. 1097–1105, 2012) revolutionized image classification and even started surpassing human performance on some benchmark data sets (Ciresan et al., Multi-column deep neural network for traffic sign classification, 2012a; He et al., Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification. CoRR, Vol. 1502.01852, 2015a). While deep neural networks have become popular primarily for image classification tasks, they can also be successfully applied to other areas and problems with some local structure in the data. We will first present a classical application of CNNs on image-like data, in particular, phenotype classification of cells based on their morphology, and then extend the task to clustering voices based on their spectrograms. Next, we will describe DL applications to semantic segmentation of newspaper pages into their corresponding articles based on clues in the pixels, and outlier detection in a predictive maintenance setting. We conclude by giving advice on how to work with DL having limited resources (e.g., training data).

Deep convolutional neural networks show outstanding performance in image-based phenotype classification given that all existing phenotypes are presented during the training of the network. However, in real-world high-content screening (HCS) experiments, it is often impossible to know all phenotypes in advance. Moreover, novel phenotype discovery itself can be an HCS outcome of interest. This aspect of HCS is not yet covered by classical deep learning approaches. When presenting an image with a novel phenotype to a trained network, it fails to indicate a novelty discovery but assigns the image to a wrong phenotype. To tackle this problem and address the need for novelty detection, we use a recently developed Bayesian approach for deep neural networks called Monte Carlo (MC) dropout to define different uncertainty measures for each phenotype prediction. With real HCS data, we show that these uncertainty measures allow us to identify novel or unclear phenotypes. In addition, we also found that the MC dropout method results in a significant improvement of classification accuracy. The proposed procedure used in our HCS case study can be easily transferred to any existing network architecture and will be beneficial in terms of accuracy and novelty detection.

Deep neural networks have become a veritable alternative to classic speaker recognition and clustering methods in recent years. However, while the speech signal clearly is a time series, and despite the body of literature on the benefits of prosodic (suprasegmental) features, identifying voices has usually not been approached with sequence learning methods. Only recently has a recurrent neural network (RNN) been successfully applied to this task, while the use of convolutional neural networks (CNNs) (that are not able to capture arbitrary time dependencies, unlike RNNs) still prevails. In this paper, we show the effectiveness of RNNs for speaker recognition by improving state of the art speaker clustering performance and robustness on the classic TIMIT benchmark. We provide arguments why RNNs are superior by experimentally showing a “sweet spot” of the segment length for successfully capturing prosodic information that has been theoretically predicted in previous work.

We propose a novel end-to-end neural network architecture that, once trained, directly outputs a probabilistic clustering of a batch of input examples in one pass. It estimates a distribution over the number of clusters $k$, and for each $1 \leq k \leq k_\mathrm{max}$, a distribution over the individual cluster assignment for each data point. The network is trained in advance in a supervised fashion on separate data to learn grouping by any perceptual similarity criterion based on pairwise labels (same/different group). It can then be applied to different data containing different groups. We demonstrate promising performance on high-dimensional data like images (COIL-100) and speech (TIMIT). We call this ``learning to cluster'' and show its conceptual difference to deep metric learning, semi-supervise clustering and other related approaches while having the advantage of performing learnable clustering fully end-to-end., Comment: Accepted for publication on ANNPR 2018

Transparent, light yellow single crystals of the barium chloride nitridotantalate(V) Ba3Cl2TaN3 (hexagonal, P63cm; a = 1507.39(8) pm, c = 632.98(3) pm, c/a = 0.420; Z = 6) are obtained by the reaction of the barium azide chloride Ba(N3)Cl or its ternary decomposition products with the container wall of arc-welded tantalum ampoules at 900 °C after four days. Its crystal structure displays one-dimensional chains 1∞{[TaN2/2N2/1]4–} consisting of corner-sharing [TaN4]7– tetrahedra (d(Ta–N) = 191–200 pm), which run along [001] and arrange as hexagonal rod packing. Charge compensation of these parallel 1∞{[TaN3]4–} strands occurs via Ba2+ cations and Cl– anions erecting a three-dimensional network 3∞{[Ba3Cl2]4+} (d(Ba–Cl) = 318–339 pm) with narrow channels apt to embed the chains. Several Ba–N contacts ranging between 264 and 319 pm secure the interconnection between both structural arrays and complete the coordination numbers of the Ba2+ cations to seven or eight.

Recent technological advances in high-content screening instrumentation have increased its ease of use and throughput, expanding the application of high-content screening to the early stages of drug discovery. However, high-content screens produce complex data sets, presenting a challenge for both extraction and interpretation of meaningful information. This shifts the high-content screening process bottleneck from the experimental to the analytical stage. In this article, the authors discuss different approaches of data analysis, using a phenotypic neurite outgrowth screen as an example. Distance measurements and hierarchical clustering methods lead to a profound understanding of different high-content screening readouts. In addition, the authors introduce a hit selection procedure based on machine learning methods and demonstrate that this method increases the hit verification rate significantly (up to a factor of 5), compared to conventional hit selection based on single readouts only. (Journal of Biomolecular Screening 2007:1042-1049)

Published online: 12 February 2016 Deep learning methods are currently outperforming traditional state-of-the-art computer vision algorithms in diverse applications and recently even surpassed human performance in object recognition. Here we demonstrate the potential of deep learning methods to high-content screening-based phenotype classification. We trained a deep learning classifier in the form of convolutional neural networks with approximately 40,000 publicly available single-cell images from samples treated with compounds from four classes known to lead to different phenotypes. The input data consisted of multichannel images. The construction of appropriate feature definitions was part of the training and carried out by the convolutional network, without the need for expert knowledge or handcrafted features. We compare our results against the recent state-of-the-art pipeline in which predefined features are extracted from each cell using specialized software and then fed into various machine learning algorithms (support vector machine, Fisher linear discriminant, random forest) for classification. The performance of all classification approaches is evaluated on an untouched test image set with known phenotype classes. Compared to the best reference machine learning algorithm, the misclassification rate is reduced from 8.9% to 6.6%.

Purpose: We aimed to identify gene expression signatures associated with angiogenesis and hypoxia pathways with predictive value for treatment response to bevacizumab/erlotinib (BE) of nonsquamous advanced non–small cell lung cancer (NSCLC) patients. Experimental Design: Whole-genome gene expression profiling was performed on 42 biopsy samples (from SAKK 19/05 trial) using Affymetrix exon arrays, and associations with the following endpoints: time-to-progression (TTP) under therapy, tumor-shrinkage (TS), and overall survival (OS) were investigated. Next, we performed gene set enrichment analyses using genes associated with the angiogenic process and hypoxia response to evaluate their predictive value for patients' outcome. Results: Our analysis revealed that both the angiogenic and hypoxia response signatures were enriched within the genes predictive of BE response, TS, and OS. Higher gene expression levels (GEL) of the 10-gene angiogenesis-associated signature and lower levels of the 10-gene hypoxia response signature predicted improved TTP under BE, 7.1 months versus 2.1 months for low versus high-risk patients (P = 0.005), and median TTP 6.9 months versus 2.9 months (P = 0.016), respectively. The hypoxia response signature associated with higher TS at 12 weeks and improved OS (17.8 months vs. 9.9 months for low vs. high-risk patients, P = 0.001). Conclusions: We were able to identify gene expression signatures derived from the angiogenesis and hypoxia response pathways with predictive value for clinical outcome in advanced nonsquamous NSCLC patients. This could lead to the identification of clinically relevant biomarkers, which will allow for selecting the subset of patients who benefit from the treatment and predict drug response. Clin Cancer Res; 21(23); 5253–63. ©2015 AACR. See related commentary by Cascone and Heymach, p. 5188

A semi-microscopic description of ionic transport in polyethylene oxide (PEO)-type electrolytes is presented, which is based on discrete stochastic moves of individual molecular units. Diffusion coefficients for ions and for the center of mass motion of chains are calculated by Monte Carlo simulation as a function of various model parameters, with emphasis on the incorporation of pressure effects. Within an even more coarse-grained description of diffusion in an athermal system of pointlike particles and chains we provide a quantitative verification of the concept of dynamic percolation (DP).

Tracer-diffusion of small molecules through dense systems of chain polymers is studied within an athermal lattice model, where hard core interactions are taken into account by means of the site exclusion principle. An approximate mapping of this problem onto dynamic percolation theory is proposed. This method is shown to yield quantitative results for the tracer correlation factor of the molecules as a function of density and chain length provided the non-Poisson character of temporal renewals in the disorder configurations is properly taken into account.