Hypoxemia is a significant driver of mortality and poor clinical outcomes in conditions such as brain injury and cardiac arrest in critically ill patients, including COVID-19 patients. Given the host of negative clinical outcomes attributed to hypoxemia, identifying patients likely to experience hypoxemia would offer valuable opportunities for early and thus more effective intervention. We present SWIFT (SpO2 Waveform ICU Forecasting Technique), a deep learning model that predicts blood oxygen saturation (SpO2) waveforms 5 and 30 minutes in the future using only prior SpO2 values as inputs. When tested on novel data, SWIFT predicts more than 80% and 60% of hypoxemic events in critically ill and COVID-19 patients, respectively. SWIFT also predicts SpO2 waveforms with average MSE below .0007. SWIFT predicts both occurrence and magnitude of potential hypoxemic events 30 minutes in the future, allowing it to be used to inform clinical interventions, patient triaging, and optimal resource allocation. SWIFT may be used in clinical decision support systems to inform the management of critically ill patients during the COVID-19 pandemic and beyond.

Ectopic beats (EBs) are cellular arrhythmias that can trigger lethal arrhythmias. Simulations using biophysically-detailed cardiac myocyte models can reveal how model parameters influence the probability of these cellular arrhythmias, however such analyses can pose a huge computational burden. Here, we develop a simplified approach in which logistic regression models (LRMs) are used to define a mapping between the parameters of complex cell models and the probability of EBs (P(EB)). As an example, in this study, we build an LRM for P(EB) as a function of the initial value of diastolic cytosolic Ca2+ concentration ([Ca2+]iini), the initial state of sarcoplasmic reticulum (SR) Ca2+ load ([Ca2+]SRini), and kinetic parameters of the inward rectifier K+ current (IK1) and ryanodine receptor (RyR). This approach, which we refer to as arrhythmia sensitivity analysis, allows for evaluation of the relationship between these arrhythmic event probabilities and their associated parameters. This LRM is also used to demonstrate how uncertainties in experimentally measured values determine the uncertainty in P(EB). In a study of the role of [Ca2+]SRini uncertainty, we show a special property of the uncertainty in P(EB), where with increasing [Ca2+]SRini uncertainty, P(EB) uncertainty first increases and then decreases. Lastly, we demonstrate that IK1 suppression, at the level that occurs in heart failure myocytes, increases P(EB).

Ectopic heartbeats can trigger reentrant arrhythmias, leading to ventricular fibrillation and sudden cardiac death. Such events have been attributed to perturbed Ca2+ handling in cardiac myocytes leading to spontaneous Ca2+ release and delayed afterdepolarizations (DADs). However, the ways in which perturbation of specific molecular mechanisms alters the probability of ectopic beats is not understood. We present a multiscale model of cardiac tissue incorporating a biophysically detailed three-dimensional model of the ventricular myocyte. This model reproduces realistic Ca2+ waves and DADs driven by stochastic Ca2+ release channel (RyR) gating and is used to study mechanisms of DAD variability. In agreement with previous experimental and modeling studies, key factors influencing the distribution of DAD amplitude and timing include cytosolic and sarcoplasmic reticulum Ca2+ concentrations, inwardly rectifying potassium current (IK1) density, and gap junction conductance. The cardiac tissue model is used to investigate how random RyR gating gives rise to probabilistic triggered activity in a one-dimensional myocyte tissue model. A novel spatial-average filtering method for estimating the probability of extreme (i.e. rare, high-amplitude) stochastic events from a limited set of spontaneous Ca2+ release profiles is presented. These events occur when randomly organized clusters of cells exhibit synchronized, high amplitude Ca2+ release flux. It is shown how reduced IK1 density and gap junction coupling, as observed in heart failure, increase the probability of extreme DADs by multiple orders of magnitude. This method enables prediction of arrhythmia likelihood and its modulation by alterations of other cellular mechanisms.

In the heart, electrical stimulation of cardiac myocytes increases the open probability of sarcolemmal voltage-sensitive Ca2+ channels and flux of Ca2+ into the cells. This increases Ca2+ binding to ligand-gated channels known as ryanodine receptors (RyR2). Their openings cause cell-wide release of Ca2+, which in turn causes muscle contraction and the generation of the mechanical force required to pump blood. In resting myocytes, RyR2s can also open spontaneously giving rise to spatially-confined Ca2+ release events known as "sparks." RyR2s are organized in a lattice to form clusters in the junctional sarcoplasmic reticulum membrane. Our recent work has shown that the spatial arrangement of RyR2s within clusters strongly influences the frequency of Ca2+ sparks. We showed that the probability of a Ca2+ spark occurring when a single RyR2 in the cluster opens spontaneously can be predicted from the precise spatial arrangements of the RyR2s. Thus, "function" follows from "structure." This probability is related to the maximum eigenvalue (λ1) of the adjacency matrix of the RyR2 cluster lattice. In this work, we develop a theoretical framework for understanding this relationship. We present a stochastic contact network model of the Ca2+ spark initiation process. We show that λ1 determines a stability threshold for the formation of Ca2+ sparks in terms of the RyR2 gating transition rates. We recapitulate these results by applying the model to realistic RyR2 cluster structures informed by super-resolution stimulated emission depletion (STED) microscopy. Eigendecomposition of the linearized mean-field contact network model reveals functional subdomains within RyR2 clusters with distinct sensitivities to Ca2+. This work provides novel perspectives on the cardiac Ca2+ release process and a general method for inferring the functional properties of transmembrane receptor clusters from their structure.

A method for automated location of shape differences in diseased anatomical structures via high resolution biomedical atlases annotated with labels from formal ontologies is described. In particular, a high resolution magnetic resonance image of the myocardium of the human left ventricle was segmented and annotated with structural terms from an extracted subset of the Foundational Model of Anatomy ontology. The atlas was registered to the end systole template of a previous study of left ventricular remodeling in cardiomyopathy using a diffeomorphic registration algorithm. The previous study used thresholding and visual inspection to locate a region of statistical significance which distinguished patients with ischemic cardiomyopathy from those with nonischemic cardiomyopathy. Using semantic technologies and the deformed annotated atlas, this location was more precisely found. Although this study used only a cardiac atlas, it provides a proof-of-concept that ontologically labeled biomedical atlases of any anatomical structure can be used to automate location-based inferences.

Loss of mitochondrial function is a fundamental determinant of cell injury and death. In heart cells under metabolic stress, we have previously described how the abrupt collapse or oscillation of the mitochondrial energy state is synchronized across the mitochondrial network by local interactions dependent upon reactive oxygen species (ROS). Here, we develop a mathematical model of ROS-induced ROS release (RIRR) based on reaction-diffusion (RD-RIRR) in one- and two-dimensional mitochondrial networks. The nodes of the RD-RIRR network are comprised of models of individual mitochondria that include a mechanism of ROS-dependent oscillation based on the interplay between ROS production, transport, and scavenging; and incorporating the tricarboxylic acid (TCA) cycle, oxidative phosphorylation, and Ca(2+) handling. Local mitochondrial interaction is mediated by superoxide (O2.-) diffusion and the O2.(-)-dependent activation of an inner membrane anion channel (IMAC). In a 2D network composed of 500 mitochondria, model simulations reveal DeltaPsi(m) depolarization waves similar to those observed when isolated guinea pig cardiomyocytes are subjected to a localized laser-flash or antioxidant depletion. The sensitivity of the propagation rate of the depolarization wave to O(2.-) diffusion, production, and scavenging in the reaction-diffusion model is similar to that observed experimentally. In addition, we present novel experimental evidence, obtained in permeabilized cardiomyocytes, confirming that DeltaPsi(m) depolarization is mediated specifically by O2.-). The present work demonstrates that the observed emergent macroscopic properties of the mitochondrial network can be reproduced in a reaction-diffusion model of RIRR. Moreover, the findings have uncovered a novel aspect of the synchronization mechanism, which is that clusters of mitochondria that are oscillating can entrain mitochondria that would otherwise display stable dynamics. The work identifies the fundamental mechanisms leading from the failure of individual organelles to the whole cell, thus it has important implications for understanding cell death during the progression of heart disease.

BACKGROUND: There is an urgent need for new prognostic markers of breast cancer metastases to ensure that newly diagnosed patients receive appropriate therapy. Recent studies have demonstrated the potential value of gene expression signatures in assessing the risk of developing distant metastases. However, due to the small sample sizes of individual studies, the overlap among signatures is almost zero and their predictive power is often limited. Integrating microarray data from multiple studies in order to increase sample size is therefore a promising approach to the development of more robust prognostic tests. RESULTS: In this study, by using a highly stable data aggregation procedure based on expression comparisons, we have integrated three independent microarray gene expression data sets for breast cancer and identified a structured prognostic signature consisting of 112 genes organized into 80 pair-wise expression comparisons. A classical likelihood ratio test based on these comparisons, essentially weighted voting, achieves 88.6% sensitivity and 54.6% specificity in an independent external test set of 154 samples. The test is highly informative in assessing the risk of developing distant metastases within five years (hazard ratio 9.3 with 95% CI 2.9-29.9). CONCLUSION: Rank-based features provide a stable way to integrate patient data from separate microarray studies due to invariance to data normalization, and such features can be combined into a useful predictor of distant metastases in breast cancer within a statistical modeling framework which begins to capture gene-gene interactions. Upon further confirmation on large-scale independent data, such prognostic signatures and tests could provide a powerful tool to guide adjuvant systemic treatment that could greatly reduce the cost of breast cancer treatment, both in terms of toxic side effects and health care expenditures.

BACKGROUND: There is a continuing need to develop molecular diagnostic tools which complement histopathologic examination to increase the accuracy of cancer diagnosis. DNA microarrays provide a means for measuring gene expression signatures which can then be used as components of genomic-based diagnostic tests to determine the presence of cancer. RESULTS: In this study, we collect and integrate ~1500 microarray gene expression profiles from 26 published cancer data sets across 21 major human cancer types. We then apply a statistical method, referred to as the Top-Scoring Pair of Groups (TSPG) classifier, and a repeated random sampling strategy to the integrated training data sets and identify a common cancer signature consisting of 46 genes. These 46 genes are naturally divided into two distinct groups; those in one group are typically expressed less than those in the other group for cancer tissues. Given a new expression profile, the classifier discriminates cancer from normal tissues by ranking the expression values of the 46 genes in the cancer signature and comparing the average ranks of the two groups. This signature is then validated by applying this decision rule to independent test data. CONCLUSION: By combining the TSPG method and repeated random sampling, a robust common cancer signature has been identified from large-scale microarray data integration. Upon further validation, this signature may be useful as a robust and objective diagnostic test for cancer.

This paper proposes a method to match diffusion tensor magnetic resonance images (DT-MRI) through the large deformation diffeomorphic metric mapping of vector fields, focusing on the fiber orientations, considered as unit vector fields on the image volume. We study a suitable action of diffeomorphisms on such vector fields, and provide an extension of the large deformation diffeomorphic metric-mapping framework to this type of dataset, resulting in optimizing for geodesies on the space of diffeomorphisms connecting two images. Two different distance function of vector fields are considered. Existence of the minimizers under smoothness assumptions on the compared vector fields is proved, and coarse to fine hierarchical strategies are detailed, to reduce both ambiguities and computation load. This is illustrated by numerical experiments on DT-MRI heart and brain images

MOTIVATION: Various studies have shown that cancer tissue samples can be successfully detected and classified by their gene expression patterns using machine learning approaches. One of the challenges in applying these techniques for classifying gene expression data is to extract accurate, readily interpretable rules providing biological insight as to how classification is performed. Current methods generate classifiers that are accurate but difficult to interpret. This is the trade-off between credibility and comprehensibility of the classifiers. Here, we introduce a new classifier in order to address these problems. It is referred to as k-TSP (k-Top Scoring Pairs) and is based on the concept of 'relative expression reversals'. This method generates simple and accurate decision rules that only involve a small number of gene-to-gene expression comparisons, thereby facilitating follow-up studies. RESULTS: In this study, we have compared our approach to other machine learning techniques for class prediction in 19 binary and multi-class gene expression datasets involving human cancers. The k-TSP classifier performs as efficiently as Prediction Analysis of Microarray and support vector machine, and outperforms other learning methods (decision trees, k-nearest neighbour and naïve Bayes). Our approach is easy to interpret as the classifier involves only a small number of informative genes. For these reasons, we consider the k-TSP method to be a useful tool for cancer classification from microarray gene expression data. AVAILABILITY: The software and datasets are available at http://www.ccbm.jhu.edu CONTACT: [email protected].

MOTIVATION: DNA microarray data analysis has been used previously to identify marker genes which discriminate cancer from normal samples. However, due to the limited sample size of each study, there are few common markers among different studies of the same cancer. With the rapid accumulation of microarray data, it is of great interest to integrate inter-study microarray data to increase sample size, which could lead to the discovery of more reliable markers. RESULTS: We present a novel, simple method of integrating different microarray datasets to identify marker genes and apply the method to prostate cancer datasets. In this study, by applying a new statistical method, referred to as the top-scoring pair (TSP) classifier, we have identified a pair of robust marker genes (HPN and STAT6) by integrating microarray datasets from three different prostate cancer studies. Cross-platform validation shows that the TSP classifier built from the marker gene pair, which simply compares relative expression values, achieves high accuracy, sensitivity and specificity on independent datasets generated using various array platforms. Our findings suggest a new model for the discovery of marker genes from accumulated microarray data and demonstrate how the great wealth of microarray data can be exploited to increase the power of statistical analysis. CONTACT: [email protected].

In large-deformation diffeomorphic metric mapping (LDDMM), the diffeomorphic matching of images are modeled as evolution in time, or a flow, of an associated smooth velocity vector field v controlling the evolution. The initial momentum parameterizes the whole geodesic and encodes the shape and form of the target image. Thus, methods such as principal component analysis (PCA) of the initial momentum leads to analysis of anatomical shape and form in target images without being restricted to small-deformation assumption in the analysis of linear displacements. We apply this approach to a study of dementia of the Alzheimer type (DAT). The left hippocampus in the DAT group shows significant shape abnormality while the right hippocampus shows similar pattern of abnormality. Further, PCA of the initial momentum leads to correct classification of 12 out of 18 DAT subjects and 22 out of 26 control subjects.

UNLABELLED: CaGE is a Cardiac Gene Expression knowledgebase we have developed to facilitate the analysis of genes important to human cardiac function. CaGE integrates the functionality of the LocusLink database with data from several human cardiac expression libraries, phenotypic data from OMIM and data from large-scale microarray gene expression studies to create a knowledgebase of gene expression in human cardiac tissue. The knowledgebase is fully searchable via the web using several intuitive query interfaces. Results can be displayed in several concise easy to navigate formats. AVAILABILITY: CaGE is located at http://www.cage.wbmei.jhu.edu

WEAVE (Workbench Environment for Analysis and Visual Exploration) is an environment for creating interactive visualization applications. WEAVE differs from previous systems in that it provides transparent linking between custom 3D visualizations and multidimensional statistical representations, and provides interactive color brushing between all visualizations. The authors demonstrate how WEAVE can be used to rapidly prototype a biomedical application, weaving together simulation data, measurement data, and 3D anatomical data concerning the propagation of excitation in the heart. These linked statistical and custom three-dimensional visualizations of the heart can allow scientists to more effectively study the correspondence of structure and behavior.

The rod photocurrent contains two noise components that may limit the detectability of flash intensity increments. The limits imposed by the low- and high-frequency noise components were assessed by computing the performance of an optimal detector of increments in flash intensity. The limits imposed by these noise components depend on the interval of observation of the photocurrent signal. When the entire photocurrent signal, lasting 3 or more seconds, is observed, the low-frequency component of the photocurrent noise (attributed to the quantal noise of the incoming light, as well as random isomerizations of enzymes within the phototransduction cascade) is the most significant limitation on detectability. When only the first 380 ms or less is observed, the high-frequency component of the noise (due to the thermal isomerizations of the cGMP-gated channel) presents a significant limit on the detectability of flashes.

Large-scale, biophysically detailed network models of the SA node and atrium are developed on the CM-5 parallel supercomputer. The models are used to investigate three factors controlling the site of generation and propagation of the cardiac pacemaker potential: magnitude of cell-to-cell gap junction coupling in the SA node; spatial variation of membrane current density in SA node; and interdigitation of SA node and atrial tissue.>

Chopper cells in the anteroventral cochlear nucleus of the cat maintain a robust rate-place representation of vowel spectra over a broad range of stimulus levels. This representation resembles that of low threshold, high spontaneous rate primary auditory nerve fibers at low stimulus levels, and that of high threshold, low spontaneous rate auditory-nerve fibers at high stimulus levels. This has led to the hypothesis that chopper cells in the anteroventral cochlear nucleus selectively process inputs from different spontaneous rate populations of primary auditory-nerve fibers at different stimulus levels. We present a computational model, making use of shunting inhibition, for how this level dependent processing may be performed within the chopper cell dendritic tree. We show that this model (1) implements level-dependent selective processing, (2) reproduces detailed features of real chopper cell post-stimulus-time histograms, and (3) reproduces nonmonotonic rate versus level functions in response to single tones measured.

This paper describes a high quality 16 kb/s subband coder using quadrature mirror filters (QMF). In this algorithm, the partitioning of the speech band into subbands is performed via a bank of eight bandpass filters whose coefficients are derived from a QMF tree structure. Due to the special properties of the QMF, no significant spectral distortions are introduced in the band-splitting and reconstruction processes. To enable a more efficient encoding of the subband waveforms, an adaptive bit assignment scheme is incorporated which allocates the number of quantizer bits according to the distribution of subband energies.