Predictors of shorter- and longer-term mortality after COVID-19 presentation among dialysis patients: parallel use of machine learning models in Latin and North American countries.

BACKGROUND: We developed machine learning models to understand the predictors of shorter-, intermediate-, and longer-term mortality among hemodialysis (HD) patients affected by COVID-19 in four countries in the Americas. METHODS: We used data from adult HD patients treated at regional institutions of a global provider in Latin America (LatAm) and North America who contracted COVID-19 in 2020 before SARS-CoV-2 vaccines were available. Using 93 commonly captured variables, we developed machine learning models that predicted the likelihood of death overall, as well as during 0-14, 15-30, > 30 days after COVID-19 presentation and identified the importance of predictors. XGBoost models were built in parallel using the same programming with a 60%:20%:20% random split for training, validation, & testing data for the datasets from LatAm (Argentina, Columbia, Ecuador) and North America (United States) countries. RESULTS: Among HD patients with COVID-19, 28.8% (1,001/3,473) died in LatAm and 20.5% (4,426/21,624) died in North America. Mortality occurred earlier in LatAm versus North America; 15.0% and 7.3% of patients died within 0-14 days, 7.9% and 4.6% of patients died within 15-30 days, and 5.9% and 8.6% of patients died > 30 days after COVID-19 presentation, respectively. Area under curve ranged from 0.73 to 0.83 across prediction models in both regions. Top predictors of death after COVID-19 consistently included older age, longer vintage, markers of poor nutrition and more inflammation in both regions at all timepoints. Unique patient attributes (higher BMI, male sex) were top predictors of mortality during 0-14 and 15-30 days after COVID-19, yet not mortality > 30 days after presentation. CONCLUSIONS: Findings showed distinct profiles of mortality in COVID-19 in LatAm and North America throughout 2020. Mortality rate was higher within 0-14 and 15-30 days after COVID-19 in LatAm, while mortality rate was higher in North America > 30 days after presentation. Nonetheless, a remarkable proportion of HD patients died > 30 days after COVID-19 presentation in both regions. We were able to develop a series of suitable prognostic prediction models and establish the top predictors of death in COVID-19 during shorter-, intermediate-, and longer-term follow up periods.

The Firing Rate Speed Code of Entorhinal Speed Cells Differs across Behaviorally Relevant Time Scales and Does Not Depend on Medial Septum Inputs.

The firing rate of speed cells, a dedicated subpopulation of neurons in the medial entorhinal cortex (MEC), is correlated with running speed. This correlation has been interpreted as a speed code used in various computational models for path integration. These models consider firing rate to be linearly tuned by running speed in real-time. However, estimation of firing rates requires integration of spiking events over time, setting constraints on the temporal accuracy of the proposed speed code. We therefore tested whether the proposed speed code by firing rate is accurate at short time scales using data obtained from open-field recordings in male rats and mice. We applied a novel filtering approach differentiating between speed codes at multiple time scales ranging from deciseconds to minutes. In addition, we determined the optimal integration time window for firing-rate estimation using a general likelihood framework and calculated the integration time window that maximizes the correlation between firing rate and running speed. Data show that these time windows are on the order of seconds, setting constraints on real-time speed coding by firing rate. We further show that optogenetic inhibition of either cholinergic, GABAergic, or glutamatergic neurons in the medial septum/diagonal band of Broca does not affect modulation of firing rates by running speed at each time scale tested. These results are relevant for models of path integration and for our understanding of how behavioral activity states may modulate firing rates and likely information processing in the MEC.SIGNIFICANCE STATEMENT Path integration is the most basic form of navigation relying on self-motion cues. Models of path integration use medial septum/diagonal band of Broca (MSDB)-dependent MEC grid-cell firing patterns as the neurophysiological substrate of path integration. These models use a linear speed code by firing rate, but do not consider temporal constraints of integration over time for firing-rate estimation. We show that firing-rate estimation for speed cells requires integration over seconds. Using optogenetics, we show that modulation of firing rates by running speed is independent of MSDB inputs. These results enhance our understanding of path integration mechanisms and the role of the MSDB for information processing in the MEC.

Neural correlates of long-term object memory in the mouse anterior cingulate cortex.

Damage to the hippocampal formation results in a profound temporally graded retrograde amnesia, implying that it is necessary for memory acquisition but not its long-term storage. It is therefore thought that memories are transferred from the hippocampus to the cortex for long-term storage in a process called systems consolidation (Dudai and Morris, 2000). Where in the cortex this occurs remains an open question. Recent work (Frankland et al., 2005; Vetere et al., 2011) suggests the anterior cingulate cortex (ACC) as a likely candidate area, but there is little direct electrophysiological evidence to support this claim. Previously, we demonstrated object-associated firing correlates in caudal ACC during tests of recognition memory and described evidence of neuronal responses to where an object had been following a brief delay. However, long-term memory requires evidence of more durable representations. Here we examined the activity of ACC neurons while testing for long-term memory of an absent object. Mice explored two objects in an arena and then were returned 6 h later with one of the objects removed. Mice continued to explore where the object had been, demonstrating memory for that object. Remarkably, some ACC neurons continued to respond where the object had been, while others developed new responses in the absent object's location. The incidence of absent-object responses by ACC neurons was greatly increased with increased familiarization to the objects, and such responses were still evident 1 month later. These data strongly suggest that the ACC contains neural correlates of consolidated object/place association memory.

Predicting SARS-CoV-2 infection among hemodialysis patients using multimodal data.

Background: The coronavirus disease 2019 (COVID-19) pandemic has created more devastation among dialysis patients than among the general population. Patient-level prediction models for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection are crucial for the early identification of patients to prevent and mitigate outbreaks within dialysis clinics. As the COVID-19 pandemic evolves, it is unclear whether or not previously built prediction models are still sufficiently effective. Methods: We developed a machine learning (XGBoost) model to predict during the incubation period a SARS-CoV-2 infection that is subsequently diagnosed after 3 or more days. We used data from multiple sources, including demographic, clinical, treatment, laboratory, and vaccination information from a national network of hemodialysis clinics, socioeconomic information from the Census Bureau, and county-level COVID-19 infection and mortality information from state and local health agencies. We created prediction models and evaluated their performances on a rolling basis to investigate the evolution of prediction power and risk factors. Result: From April 2020 to August 2020, our machine learning model achieved an area under the receiver operating characteristic curve (AUROC) of 0.75, an improvement of over 0.07 from a previously developed machine learning model published by Kidney360 in 2021. As the pandemic evolved, the prediction performance deteriorated and fluctuated more, with the lowest AUROC of 0.6 in December 2021 and January 2022. Over the whole study period, that is, from April 2020 to February 2022, fixing the false-positive rate at 20%, our model was able to detect 40% of the positive patients. We found that features derived from local infection information reported by the Centers for Disease Control and Prevention (CDC) were the most important predictors, and vaccination status was a useful predictor as well. Whether or not a patient lives in a nursing home was an effective predictor before vaccination, but became less predictive after vaccination. Conclusion: As found in our study, the dynamics of the prediction model are frequently changing as the pandemic evolves. County-level infection information and vaccination information are crucial for the success of early COVID-19 prediction models. Our results show that the proposed model can effectively identify SARS-CoV-2 infections during the incubation period. Prospective studies are warranted to explore the application of such prediction models in daily clinical practice.

Machine Learning for Prediction of Patients on Hemodialysis with an Undetected SARS-CoV-2 Infection.

Background: We developed a machine learning (ML) model that predicts the risk of a patient on hemodialysis (HD) having an undetected SARS-CoV-2 infection that is identified after the following ≥3 days. Methods: As part of a healthcare operations effort, we used patient data from a national network of dialysis clinics (February-September 2020) to develop an ML model (XGBoost) that uses 81 variables to predict the likelihood of an adult patient on HD having an undetected SARS-CoV-2 infection that is identified in the subsequent ≥3 days. We used a 60%:20%:20% randomized split of COVID-19-positive samples for the training, validation, and testing datasets. Results: We used a select cohort of 40,490 patients on HD to build the ML model (11,166 patients who were COVID-19 positive and 29,324 patients who were unaffected controls). The prevalence of COVID-19 in the cohort (28% COVID-19 positive) was by design higher than the HD population. The prevalence of COVID-19 was set to 10% in the testing dataset to estimate the prevalence observed in the national HD population. The threshold for classifying observations as positive or negative was set at 0.80 to minimize false positives. Precision for the model was 0.52, the recall was 0.07, and the lift was 5.3 in the testing dataset. Area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) for the model was 0.68 and 0.24 in the testing dataset, respectively. Top predictors of a patient on HD having a SARS-CoV-2 infection were the change in interdialytic weight gain from the previous month, mean pre-HD body temperature in the prior week, and the change in post-HD heart rate from the previous month. Conclusions: The developed ML model appears suitable for predicting patients on HD at risk of having COVID-19 at least 3 days before there would be a clinical suspicion of the disease.

A longitudinal study of event related potentials and correlations with psychosocial functioning and clinical features in first episode psychosis patients.

BACKGROUND: Deficits in auditory event-related potentials (ERPs), brain responses to stimuli indexing different cognitive processes, have been demonstrated widely in chronic schizophrenia (SZ) patients though much less is known about these responses across the early course of psychosis. The present study examined multiple ERP components in first episode psychosis (FEP) patients longitudinally and investigated the relationships between ERPs, psychosocial functioning, and clinical features over time. METHODS: N1, P2, P3a, and P3b ERPs were elicited using a three-stimulus (novelty) auditory oddball paradigm. FEP patients included SZ-spectrum and psychotic bipolar disorder (BD) diagnoses. Data were collected from 41 patients at baseline, 20 patients at 12-month follow-up, 14 at 24-month follow-up, and 29 healthy control subjects. RESULTS: N1 and P2 ERPs were intact across the early stages of psychosis. Baseline P2 was significantly larger in BD than SZ patients. Reduced P3a and P3b ERPs were found in patients followed longitudinally and are stable over time. ERPs tracked distinct aspects of symptomology and medication, though specific associations were inconsistent across time. Baseline P3a amplitude predicted later psychosocial functioning. The pattern of correlations between ERP components in patients differed from controls. DISCUSSION: Baseline P3a ERP, and PANSS general score were significant and independent predictors of later MCAS functioning at 12-month. Overall, individuals with worse functioning and greater symptomology produced smaller amplitudes. Our results highlight the heterogeneity within the FEP population. Correlation patterns among ERPs are similar between patients and controls. P3a and P3b amplitudes appear to link with higher-order cognitive and psychosocial functioning.

Reconciling the different faces of hippocampal theta: The role of theta oscillations in cognitive, emotional and innate behaviors.

The theta oscillation (5-10Hz) is a prominent behavior-specific brain rhythm. This review summarizes studies showing the multifaceted role of theta rhythm in cognitive functions, including spatial coding, time coding and memory, exploratory locomotion and anxiety-related behaviors. We describe how activity of hippocampal theta rhythm generators - medial septum, nucleus incertus and entorhinal cortex, links theta with specific behaviors. We review evidence for functions of the theta-rhythmic signaling to subcortical targets, including lateral septum. Further, we describe functional associations of theta oscillation properties - phase, frequency and amplitude - with memory, locomotion and anxiety, and outline how manipulations of these features, using optogenetics or pharmacology, affect associative and innate behaviors. We discuss work linking cognition to the slope of the theta frequency to running speed regression, and emotion-sensitivity (anxiolysis) to its y-intercept. Finally, we describe parallel emergence of theta oscillations, theta-mediated neuronal activity and behaviors during development. This review highlights a complex interplay of neuronal circuits and synchronization features, which enables an adaptive regulation of multiple behaviors by theta-rhythmic signaling.

Systemic administration of two different anxiolytic drugs decreases local field potential theta frequency in the medial entorhinal cortex without affecting grid cell firing fields.

Neurons coding spatial location (grid cells) are found in medial entorhinal cortex (MEC) and demonstrate increasing size of firing fields and spacing between fields (grid scale) along the dorsoventral axis. This change in grid scale correlates with differences in theta frequency, a 6-10Hz rhythm in the local field potential (LFP) and rhythmic firing of cells. A relationship between theta frequency and grid scale can be found when examining grid cells recorded in different locations along the dorsoventral axis of MEC. When describing the relationship between theta frequency and grid scale, it is important to account for the strong positive correlation between theta frequency and running speed. Plotting LFP theta frequency across running speeds dissociates two components of this relationship: slope and intercept of the linear fit. Change in theta frequency through a change in the slope component has been modeled and shown experimentally to affect grid scale, but the prediction that change in the intercept component would not affect grid scale has not been tested experimentally. This prediction about the relationship of intercept to grid scale is the primary hypothesis tested in the experiments presented here. All known anxiolytic drugs decrease hippocampal theta frequency despite their differing mechanisms of action. Specifically, anxiolytics decrease the intercept of the theta frequency-running speed relationship in the hippocampus. Here we demonstrate that anxiolytics decrease the intercept of the theta frequency-running speed relationship in the MEC, similar to hippocampus, and the decrease in frequency through this change in intercept does not affect grid scale.

Cholinergic modulation of cognitive processing: insights drawn from computational models.

Acetylcholine plays an important role in cognitive function, as shown by pharmacological manipulations that impact working memory, attention, episodic memory, and spatial memory function. Acetylcholine also shows striking modulatory influences on the cellular physiology of hippocampal and cortical neurons. Modeling of neural circuits provides a framework for understanding how the cognitive functions may arise from the influence of acetylcholine on neural and network dynamics. We review the influences of cholinergic manipulations on behavioral performance in working memory, attention, episodic memory, and spatial memory tasks, the physiological effects of acetylcholine on neural and circuit dynamics, and the computational models that provide insight into the functional relationships between the physiology and behavior. Specifically, we discuss the important role of acetylcholine in governing mechanisms of active maintenance in working memory tasks and in regulating network dynamics important for effective processing of stimuli in attention and episodic memory tasks. We also propose that theta rhythm plays a crucial role as an intermediary between the physiological influences of acetylcholine and behavior in episodic and spatial memory tasks. We conclude with a synthesis of the existing modeling work and highlight future directions that are likely to be rewarding given the existing state of the literature for both empiricists and modelers.