M. Brandon Westover

Searching and mining trillions of time series subsequences under dynamic time warping

Most time series data mining algorithms use similarity search as a core subroutine, and thus the time taken for similarity search is the bottleneck for virtually all time series data mining algorithms. The difficulty of scaling search to large datasets largely explains why most academic work on time series data mining has plateaued at considering a few millions of time series objects, while much of industry and science sits on billions of time series objects waiting to be explored. In this work we show that by using a combination of four novel ideas we can search and mine truly massive time series for the first time. We demonstrate the following extremely unintuitive fact; in large datasets we can exactly search under DTW much more quickly than the current state-of-the-art Euclidean distance search algorithms. We demonstrate our work on the largest set of time series experiments ever attempted. In particular, the largest dataset we consider is larger than the combined size of all of the time series datasets considered in all data mining papers ever published. We show that our ideas allow us to solve higher-level time series data mining problem such as motif discovery and clustering at scales that would otherwise be untenable. In addition to mining massive datasets, we will show that our ideas also have implications for real-time monitoring of data streams, allowing us to handle much faster arrival rates and/or use cheaper and lower powered devices than are currently possible.

Sparse Extreme Learning Machine for Classification

Extreme learning machine (ELM) was initially proposed for single-hidden-layer feedforward neural networks (SLFNs). In the hidden layer (feature mapping), nodes are randomly generated independently of training data. Furthermore, a unified ELM was proposed, providing a single framework to simplify and unify different learning methods, such as SLFNs, least square support vector machines, proximal support vector machines, and so on. However, the solution of unified ELM is dense, and thus, usually plenty of storage space and testing time are required for large-scale applications. In this paper, a sparse ELM is proposed as an alternative solution for classification, reducing storage space and testing time. In addition, unified ELM obtains the solution by matrix inversion, whose computational complexity is between quadratic and cubic with respect to the training size. It still requires plenty of training time for large-scale problems, even though it is much faster than many other traditional methods. In this paper, an efficient training algorithm is specifically developed for sparse ELM. The quadratic programming problem involved in sparse ELM is divided into a series of smallest possible sub-problems, each of which are solved analytically. Compared with SVM, sparse ELM obtains better generalization performance with much faster training speed. Compared with unified ELM, sparse ELM achieves similar generalization performance for binary classification applications, and when dealing with large-scale binary classification problems, sparse ELM realizes even faster training speed than unified ELM.

Statin Use Following Intracerebral Hemorrhage: A Decision Analysis

CONTEXT Statins are widely prescribed for primary and secondary prevention of ischemic cardiac and cerebrovascular disease. Although serious adverse effects are uncommon, results from a recent clinical trial suggested increased risk of intracerebral hemorrhage (ICH) associated with statin use. For patients with baseline elevated risk of ICH, it is not known whether this potential adverse effect offsets the cardiovascular and cerebrovascular benefits. OBJECTIVE To address the following clinical question: Given a history of prior ICH, should statin therapy be avoided? DESIGN A Markov decision model was used to evaluate the risks and benefits of statin therapy in patients with prior ICH. MAIN OUTCOME MEASURE Life expectancy, measured as quality-adjusted life-years. We investigated how statin use affects this outcome measure while varying a range of clinical parameters, including hemorrhage location (deep vs lobar), ischemic cardiac and cerebrovascular risks, and magnitude of ICH risk associated with statins. RESULTS Avoiding statins was favored over a wide range of values for many clinical parameters, particularly in survivors of lobar ICH who are at highest risk of ICH recurrence. In survivors of lobar ICH without prior cardiovascular events, avoiding statins yielded a life expectancy gain of 2.2 quality-adjusted life-years compared with statin use. This net benefit persisted even at the lower 95% confidence interval of the relative risk of statin-associated ICH. In patients with lobar ICH who had prior cardiovascular events, the annual recurrence risk of myocardial infarction would have to exceed 90% to favor statin therapy. Avoiding statin therapy was also favored, although by a smaller margin, in both primary and secondary prevention settings for survivors of deep ICH. CONCLUSIONS Avoiding statins should be considered for patients with a history of ICH, particularly those cases with a lobar location.

Exploration and modulation of brain network interactions with noninvasive brain stimulation in combination with neuroimaging

Much recent work in systems neuroscience has focused on how dynamic interactions between different cortical regions underlie complex brain functions such as motor coordination, language and emotional regulation. Various studies using neuroimaging and neurophysiologic techniques have suggested that in many neuropsychiatric disorders, these dynamic brain networks are dysregulated. Here we review the utility of combined noninvasive brain stimulation and neuroimaging approaches towards greater understanding of dynamic brain networks in health and disease. Brain stimulation techniques, such as transcranial magnetic stimulation and transcranial direct current stimulation, use electromagnetic principles to alter brain activity noninvasively, and induce focal but also network effects beyond the stimulation site. When combined with brain imaging techniques such as functional magnetic resonance imaging, positron emission tomography and electroencephalography, these brain stimulation techniques enable a causal assessment of the interaction between different network components, and their respective functional roles. The same techniques can also be applied to explore hypotheses regarding the changes in functional connectivity that occur during task performance and in various disease states such as stroke, depression and schizophrenia. Finally, in diseases characterized by pathologic alterations in either the excitability within a single region or in the activity of distributed networks, such techniques provide a potential mechanism to alter cortical network function and architectures in a beneficial manner.

Effects of sevoflurane and propofol on frontal electroencephalogram power and coherence.

Background:The neural mechanisms of anesthetic vapors have not been studied in depth. However, modeling and experimental studies on the intravenous anesthetic propofol indicate that potentiation of &ggr;-aminobutyric acid receptors leads to a state of thalamocortical synchrony, observed as coherent frontal alpha oscillations, associated with unconsciousness. Sevoflurane, an ether derivative, also potentiates &ggr;-aminobutyric acid receptors. However, in humans, sevoflurane-induced coherent frontal alpha oscillations have not been well detailed. Methods:To study the electroencephalogram dynamics induced by sevoflurane, the authors identified age- and sex-matched patients in which sevoflurane (n = 30) or propofol (n = 30) was used as the sole agent for maintenance of general anesthesia during routine surgery. The authors compared the electroencephalogram signatures of sevoflurane with that of propofol using time-varying spectral and coherence methods. Results:Sevoflurane general anesthesia is characterized by alpha oscillations with maximum power and coherence at approximately 10 Hz, (mean ± SD; peak power, 4.3 ± 3.5 dB; peak coherence, 0.73 ± 0.1). These alpha oscillations are similar to those observed during propofol general anesthesia, which also has maximum power and coherence at approximately 10 Hz (peak power, 2.1 ± 4.3 dB; peak coherence, 0.71 ± 0.1). However, sevoflurane also exhibited a distinct theta coherence signature (peak frequency, 4.9 ± 0.6 Hz; peak coherence, 0.58 ± 0.1). Slow oscillations were observed in both cases, with no significant difference in power or coherence. Conclusions:The study results indicate that sevoflurane, like propofol, induces coherent frontal alpha oscillations and slow oscillations in humans to sustain the anesthesia-induced unconscious state. These results suggest a shared molecular and systems-level mechanism for the unconscious state induced by these drugs.

Interrater agreement for Critical Care EEG Terminology

The interpretation of critical care electroencephalography (EEG) studies is challenging because of the presence of many periodic and rhythmic patterns of uncertain clinical significance. Defining the clinical significance of these patterns requires standardized terminology with high interrater agreement (IRA). We sought to evaluate IRA for the final, published American Clinical Neurophysiology Society (ACNS)–approved version of the critical care EEG terminology (2012 version). Our evaluation included terms not assessed previously and incorporated raters with a broad range of EEG reading experience.

The continuum of spreading depolarizations in acute cortical lesion development: Examining Leão's legacy.

A modern understanding of how cerebral cortical lesions develop after acute brain injury is based on Aristides Leão’s historic discoveries of spreading depression and asphyxial/anoxic depolarization. Treated as separate entities for decades, we now appreciate that these events define a continuum of spreading mass depolarizations, a concept that is central to understanding their pathologic effects. Within minutes of acute severe ischemia, the onset of persistent depolarization triggers the breakdown of ion homeostasis and development of cytotoxic edema. These persistent changes are diagnosed as diffusion restriction in magnetic resonance imaging and define the ischemic core. In delayed lesion growth, transient spreading depolarizations arise spontaneously in the ischemic penumbra and induce further persistent depolarization and excitotoxic damage, progressively expanding the ischemic core. The causal role of these waves in lesion development has been proven by real-time monitoring of electrophysiology, blood flow, and cytotoxic edema. The spreading depolarization continuum further applies to other models of acute cortical lesions, suggesting that it is a universal principle of cortical lesion development. These pathophysiologic concepts establish a working hypothesis for translation to human disease, where complex patterns of depolarizations are observed in acute brain injury and appear to mediate and signal ongoing secondary damage.

Recording, analysis, and interpretation of spreading depolarizations in neurointensive care: Review and recommendations of the COSBID research group

Spreading depolarizations (SD) are waves of abrupt, near-complete breakdown of neuronal transmembrane ion gradients, are the largest possible pathophysiologic disruption of viable cerebral gray matter, and are a crucial mechanism of lesion development. Spreading depolarizations are increasingly recorded during multimodal neuromonitoring in neurocritical care as a causal biomarker providing a diagnostic summary measure of metabolic failure and excitotoxic injury. Focal ischemia causes spreading depolarization within minutes. Further spreading depolarizations arise for hours to days due to energy supply-demand mismatch in viable tissue. Spreading depolarizations exacerbate neuronal injury through prolonged ionic breakdown and spreading depolarization-related hypoperfusion (spreading ischemia). Local duration of the depolarization indicates local tissue energy status and risk of injury. Regional electrocorticographic monitoring affords even remote detection of injury because spreading depolarizations propagate widely from ischemic or metabolically stressed zones; characteristic patterns, including temporal clusters of spreading depolarizations and persistent depression of spontaneous cortical activity, can be recognized and quantified. Here, we describe the experimental basis for interpreting these patterns and illustrate their translation to human disease. We further provide consensus recommendations for electrocorticographic methods to record, classify, and score spreading depolarizations and associated spreading depressions. These methods offer distinct advantages over other neuromonitoring modalities and allow for future refinement through less invasive and more automated approaches.

A comparison of propofol- and dexmedetomidine-induced electroencephalogram dynamics using spectral and coherence analysis.

Background:Electroencephalogram patterns observed during sedation with dexmedetomidine appear similar to those observed during general anesthesia with propofol. This is evident with the occurrence of slow (0.1 to 1 Hz), delta (1 to 4 Hz), propofol-induced alpha (8 to 12 Hz), and dexmedetomidine-induced spindle (12 to 16 Hz) oscillations. However, these drugs have different molecular mechanisms and behavioral properties and are likely accompanied by distinguishing neural circuit dynamics. Methods:The authors measured 64-channel electroencephalogram under dexmedetomidine (n = 9) and propofol (n = 8) in healthy volunteers, 18 to 36 yr of age. The authors administered dexmedetomidine with a 1-µg/kg loading bolus over 10 min, followed by a 0.7 µg kg−1 h−1 infusion. For propofol, the authors used a computer-controlled infusion to target the effect-site concentration gradually from 0 to 5 &mgr;g/ml. Volunteers listened to auditory stimuli and responded by button press to determine unconsciousness. The authors analyzed the electroencephalogram using multitaper spectral and coherence analysis. Results:Dexmedetomidine was characterized by spindles with maximum power and coherence at approximately 13 Hz (mean ± SD; power, −10.8 ± 3.6 dB; coherence, 0.8 ± 0.08), whereas propofol was characterized with frontal alpha oscillations with peak frequency at approximately 11 Hz (power, 1.1 ± 4.5 dB; coherence, 0.9 ± 0.05). Notably, slow oscillation power during a general anesthetic state under propofol (power, 13.2 ± 2.4 dB) was much larger than during sedative states under both propofol (power, −2.5 ± 3.5 dB) and dexmedetomidine (power, −0.4 ± 3.1 dB). Conclusion:The results indicate that dexmedetomidine and propofol place patients into different brain states and suggest that propofol enables a deeper state of unconsciousness by inducing large-amplitude slow oscillations that produce prolonged states of neuronal silence.

Absence of early epileptiform abnormalities predicts lack of seizures on continuous EEG

Objective: To determine whether the absence of early epileptiform abnormalities predicts absence of later seizures on continuous EEG monitoring of hospitalized patients. Methods: We retrospectively reviewed 242 consecutive patients without a prior generalized convulsive seizure or active epilepsy who underwent continuous EEG monitoring lasting at least 18 hours for detection of nonconvulsive seizures or evaluation of unexplained altered mental status. The findings on the initial 30-minute screening EEG, subsequent continuous EEG recordings, and baseline clinical data were analyzed. We identified early EEG findings associated with absence of seizures on subsequent continuous EEG. Results: Seizures were detected in 70 (29%) patients. A total of 52 patients had their first seizure in the initial 30 minutes of continuous EEG monitoring. Of the remaining 190 patients, 63 had epileptiform discharges on their initial EEG, 24 had triphasic waves, while 103 had no epileptiform abnormalities. Seizures were later detected in 22% (n = 14) of studies with epileptiform discharges on their initial EEG, vs 3% (n = 3) of the studies without epileptiform abnormalities on initial EEG (p < 0.001). In the 3 patients without epileptiform abnormalities on initial EEG but with subsequent seizures, the first epileptiform discharge or electrographic seizure occurred within the first 4 hours of recording. Conclusions: In patients without epileptiform abnormalities during the first 4 hours of recording, no seizures were subsequently detected. Therefore, EEG features early in the recording may indicate a low risk for seizures, and help determine whether extended monitoring is necessary.