scholarly journals AdaCare: Explainable Clinical Health Status Representation Learning via Scale-Adaptive Feature Extraction and Recalibration

2020 ◽  
Vol 34 (01) ◽  
pp. 825-832 ◽  
Author(s):  
Liantao Ma ◽  
Junyi Gao ◽  
Yasha Wang ◽  
Chaohe Zhang ◽  
Jiangtao Wang ◽  
...  
Keyword(s):  
Health Status ◽  
Health Risk ◽  
Clinical Features ◽  
Time Scales ◽  
Prediction Models ◽  
State Of The Art ◽  
Representation Learning ◽  
Superior Performance ◽  
Multiple Time Scales ◽  
Multiple Time

Deep learning-based health status representation learning and clinical prediction have raised much research interest in recent years. Existing models have shown superior performance, but there are still several major issues that have not been fully taken into consideration. First, the historical variation pattern of the biomarker in diverse time scales plays a vital role in indicating the health status, but it has not been explicitly extracted by existing works. Second, key factors that strongly indicate the health risk are different among patients. It is still challenging to adaptively make use of the features for patients in diverse conditions. Third, using prediction models as the black box will limit the reliability in clinical practice. However, none of the existing works can provide satisfying interpretability and meanwhile achieve high prediction performance. In this work, we develop a general health status representation learning model, named AdaCare. It can capture the long and short-term variations of biomarkers as clinical features to depict the health status in multiple time scales. It also models the correlation between clinical features to enhance the ones which strongly indicate the health status and thus can maintain a state-of-the-art performance in terms of prediction accuracy while providing qualitative interpretability. We conduct a health risk prediction experiment on two real-world datasets. Experiment results indicate that AdaCare outperforms state-of-the-art approaches and provides effective interpretability, which is verifiable by clinical experts.

scholarly journals Average Predictability Time. Part II: Seamless Diagnoses of Predictability on Multiple Time Scales

2009 ◽  
Vol 66 (5) ◽  
pp. 1188-1204 ◽  
Author(s):  
Timothy DelSole ◽  
Michael K. Tippett
Keyword(s):  
Time Scales ◽  
State Vector ◽  
Prediction Models ◽  
Decadal Variability ◽  
Linear Trend ◽  
New Method ◽  
Multiple Time Scales ◽  
Climate Change Signal ◽  
Time Averaging ◽  
Multiple Time

Abstract This paper proposes a new method for diagnosing predictability on multiple time scales without time averaging. The method finds components that maximize the average predictability time (APT) of a system, where APT is defined as the integral of the average predictability over all lead times. Basing the predictability measure on the Mahalanobis metric leads to a complete, uncorrelated set of components that can be ordered by their contribution to APT, analogous to the way principal components decompose variance. The components and associated APTs are invariant to nonsingular linear transformations, allowing variables with different units and natural variability to be considered in a single state vector without normalization. For prediction models derived from linear regression, maximizing APT is equivalent to maximizing the sum of squared multiple correlations between the component and the time-lagged state vector. The new method is used to diagnose predictability of 1000-hPa zonal velocity on time scales from 6 h to decades. The leading predictable component is dominated by a linear trend and presumably identifies a climate change signal. The next component is strongly correlated with ENSO indices and hence is identified with seasonal-to-interannual predictability. The third component is related to annular modes and presents decadal variability as well as a trend. The next few components have APTs exceeding 10 days. A reconstruction of the tropical zonal wind field based on the leading seven components reveals eastward propagation of anomalies with time scales consistent with the Madden–Julian oscillation. The remaining components have time scales less than a week and hence are identified with weather predictability. The detection of predictability on these time scales without time averaging is possible by virtue of the fact that predictability on different time scales is characterized by different spatial structures, which can be optimally extracted by suitable projections.

SOME REMARKS ON THE MULTIPLE TIME SCALES OF DIFFUSION IN MINERALS AND MELTS

2018 ◽  
Author(s):  
Yan Liang ◽  
◽  
Daniele J. Cherniak ◽  
Chenguang Sun
Keyword(s):  
Time Scales ◽  
Multiple Time Scales ◽  
Multiple Time

scholarly journals Are We Doing ‘Systems’ Research? An Assessment of Methods for Climate Change Adaptation to Hydrohazards in a Complex World

2019 ◽  
Vol 11 (4) ◽  
pp. 1163 ◽  
Author(s):  
Melissa Bedinger ◽  
Lindsay Beevers ◽  
Lila Collet ◽  
Annie Visser
Keyword(s):  
Climate Change ◽  
Human Nature ◽  
Climate Change Adaptation ◽  
Time Scales ◽  
Spatial Scales ◽  
Local Context ◽  
Multiple Time Scales ◽  
Future Research ◽  
Multiple Time ◽  
Systems Research

Climate change is a product of the Anthropocene, and the human–nature system in which we live. Effective climate change adaptation requires that we acknowledge this complexity. Theoretical literature on sustainability transitions has highlighted this and called for deeper acknowledgment of systems complexity in our research practices. Are we heeding these calls for ‘systems’ research? We used hydrohazards (floods and droughts) as an example research area to explore this question. We first distilled existing challenges for complex human–nature systems into six central concepts: Uncertainty, multiple spatial scales, multiple time scales, multimethod approaches, human–nature dimensions, and interactions. We then performed a systematic assessment of 737 articles to examine patterns in what methods are used and how these cover the complexity concepts. In general, results showed that many papers do not reference any of the complexity concepts, and no existing approach addresses all six. We used the detailed results to guide advancement from theoretical calls for action to specific next steps. Future research priorities include the development of methods for consideration of multiple hazards; for the study of interactions, particularly in linking the short- to medium-term time scales; to reduce data-intensivity; and to better integrate bottom–up and top–down approaches in a way that connects local context with higher-level decision-making. Overall this paper serves to build a shared conceptualisation of human–nature system complexity, map current practice, and navigate a complexity-smart trajectory for future research.

scholarly journals A multistate model incorporating estimation of excess hazards and multiple time scales

2021 ◽  
Vol 40 (9) ◽  
pp. 2139-2154
Author(s):  
Caroline E. Weibull ◽  
Paul C. Lambert ◽  
Sandra Eloranta ◽  
Therese M. L. Andersson ◽  
Paul W. Dickman ◽  
...  
Keyword(s):  
Time Scales ◽  
Multiple Time Scales ◽  
Multiple Time ◽  
Multistate Model

scholarly journals Scale-dependent intrinsic entropies of complex time series

2016 ◽  
Vol 374 (2065) ◽  
pp. 20150204 ◽  
Author(s):  
Jia-Rong Yeh ◽  
Chung-Kang Peng ◽  
Norden E. Huang
Keyword(s):  
Time Series ◽  
Time Scales ◽  
Spatial Scales ◽  
Multiple Time Scales ◽  
Multiple Time ◽  
Complex Time ◽  
Multi Scale ◽  
Mode Decomposition ◽  
Simulated Time ◽  
Simulated Time Series

Multi-scale entropy (MSE) was developed as a measure of complexity for complex time series, and it has been applied widely in recent years. The MSE algorithm is based on the assumption that biological systems possess the ability to adapt and function in an ever-changing environment, and these systems need to operate across multiple temporal and spatial scales, such that their complexity is also multi-scale and hierarchical. Here, we present a systematic approach to apply the empirical mode decomposition algorithm, which can detrend time series on various time scales, prior to analysing a signal’s complexity by measuring the irregularity of its dynamics on multiple time scales. Simulated time series of fractal Gaussian noise and human heartbeat time series were used to study the performance of this new approach. We show that our method can successfully quantify the fractal properties of the simulated time series and can accurately distinguish modulations in human heartbeat time series in health and disease.

scholarly journals Release of a single neurotransmitter from an identified interneuron coherently affects motor output on multiple time scales

2013 ◽  
Vol 109 (9) ◽  
pp. 2327-2334 ◽  
Author(s):  
Andrew M. Dacks ◽  
Klaudiusz R. Weiss
Keyword(s):  
Time Scales ◽  
Motor Program ◽  
Gaba Release ◽  
Motor Output ◽  
Multiple Time Scales ◽  
Multiple Time ◽  
Intrinsic Properties ◽  
Buccal Ganglion ◽  
Motor Programs ◽  
Pattern Characteristic

Neurotransmitters can have diverse effects that occur over multiple time scales often making the consequences of neurotransmission difficult to predict. To explore the consequences of this diversity, we used the buccal ganglion of Aplysia to examine the effects of GABA release by a single interneuron, B40, on the intrinsic properties and motor output of the radula closure neuron B8. B40 induces a picrotoxin-sensitive fast IPSP lasting milliseconds in B8 and a slow EPSP lasting seconds. We found that the excitatory effects of this slow EPSP are also mediated by GABA. Together, these two GABAergic actions structure B8 firing in a pattern characteristic of ingestive programs. Furthermore, we found that repeated B40 stimulation induces a persistent increase in B8 excitability that was occluded in the presence of the GABA B receptor agonist baclofen, suggesting that GABA affects B8 excitability over multiple time scales. The phasing of B8 activity during the feeding motor programs determines the nature of the behavior elicited during that motor program. The persistent increase in B8 excitability induced by B40 biased the activity of B8 during feeding motor programs causing the motor programs to become more ingestive in nature. Thus, a single transmitter released from a single interneuron can have consequences for motor output that are expressed over multiple time scales. Importantly, despite the differences in their signs and temporal characteristics, the three actions of B40 are coherent in that they promote B8 firing patterns that are characteristic of ingestive motor outputs.

Sign in / Sign up

Export Citation Format

Share Document