Impact of sensor data pre-processing strategies and selection of machine learning algorithm on the prediction of metritis events in dairy cattle.

With all the sensor data currently generated at high frequency in dairy farms, there is potential for earlier diagnosis of postpartum diseases compared with traditional monitoring methodologies. Our objectives were 1) to compare the impact of sensor data pre-processing on classifier performance by using multiple time windows before a given metritis event, while considering other cow-level factors and farm-scheduled activities; 2) to compare the performance of random forest (RF), k-nearest neighbors (k-NN), and support vector machine (SVM) classifiers at different decision thresholds using different number of past observations (time-lags) for the detection of behavioral patterns associated with changes in metritis scores; and 3) to compare classifier performance between each one of the five behaviors registered every hour by an ear-tag 3-axis accelerometer (CowManager, Agis Autimatisering, Harmelen, Netherlands). A total of 239 metritis events were created by comparing metritis scores between two consecutive clinical evaluations from cows that were retrospectively selected from a dataset containing sensor data and health information during the first 21 days postpartum from June 2014 to May 2017. Hourly sensor data classified by the accelerometer as either ruminating, eating, not active (including both standing or lying), and two different levels of activity (active and high activity) behaviors corresponding to the 3 days before each metritis event were aggregated every 24-, 12-, 6-, and 3-hour time windows. Multiple time-lags were also used to determine the optimal number of past observations needed for optimal classification. Similarly, different decision thresholds were compared in terms of model performance. Depending on the classifier, algorithm hyperparameters were optimized using grid search (RF, k-NN, SVM) and random search (RF). All behaviors changed throughout the study period and showed distinct daily patterns. From the three algorithms, RF had the highest F1 score followed by k-NN and SVM. Furthermore, sensor data aggregated every 6- or 12-h time windows had the best model performance at multiple time-lags. We concluded that the data from the first 3 days post-partum should be discarded when studying metritis, and either one of the five behaviors measured with CowManager could be used when predicting metritis when sensor data were aggregated every 6- or 12-hour time windows, and using time-lags corresponding to 2-3 days before a given event, depending on the time window used. This study shows how to maximize sensor data in their potential for disease prediction, enhancing the performance of algorithms used in machine learning.

The impact of COVID-19 vaccination on California's return to normalcy.

SARS-CoV-2 has infected nearly 3.7 million and killed 61,722 Californians, as of May 22, 2021. Non-pharmaceutical interventions have been instrumental in mitigating the spread of the coronavirus. However, as we ease restrictions, widespread implementation of COVID-19 vaccines is essential to prevent its resurgence. In this work, we addressed the adequacy and deficiency of vaccine uptake within California and the possibility and severity of resurgence of COVID-19 as restrictions are lifted given the current vaccination rates. We implemented a real-time Bayesian data assimilation approach to provide projections of incident cases and deaths in California following the reopening of its economy on June 15, 2021. We implemented scenarios that vary vaccine uptake prior to reopening, and transmission rates and effective population sizes following the reopening. For comparison purposes, we adopted a baseline scenario using the current vaccination rates, which projects a total 11,429 cases and 429 deaths in a 15-day period after reopening. We used posterior estimates based on CA historical data to provide realistic model parameters after reopening. When the transmission rate is increased after reopening, we projected an increase in cases by 21.8% and deaths by 4.4% above the baseline after reopening. When the effective population is increased after reopening, we observed an increase in cases by 51.8% and deaths by 12.3% above baseline. A 30% reduction in vaccine uptake alone has the potential to increase cases and deaths by 35% and 21.6%, respectively. Conversely, increasing vaccine uptake by 30% could decrease cases and deaths by 26.1% and 17.9%, respectively. As California unfolds its plan to reopen its economy on June 15, 2021, it is critical that social distancing and public behavior changes continue to be promoted, particularly in communities with low vaccine uptake. The Centers for Disease Control and Prevention (CDC) recommendation to ease mask-wearing for fully vaccinated individuals despite major inequities in vaccine uptake in counties across the state highlights some of the logistical challenges that society faces as we enthusiastically phase out of this pandemic.

The Influence of Service Era: Comparing Personality Assessment Inventory (PAI) Scale Scores Within a Posttraumatic Stress Disorder Treatment Clinic (PCT).

Research is mixed on the role of service era in symptom endorsement among Veterans, with differences emerging depending on the instrument evaluated. This study compares Personality Assessment Inventory (PAI) scale scores of VA test-takers who served during the Vietnam, Desert Storm, or Post-9/11 service eras. The sample was collected at a VA Posttraumatic Stress Disorder Clinical Team. Associations between gender and combat exposure were also examined as covariates. Results suggest that Veterans' self-report on the PAI is influenced by service era, even after accounting for gender and combat exposure during deployment. The largest differences were between Vietnam or Post-9/11 Veterans and those from the Gulf War era. Symptom differences typically varied across scales commonly associated with symptoms of trauma exposure/posttraumatic stress disorder. Implications for the clinical use of, and research with, the PAI and other broadband personality assessments within the VA healthcare system and trauma treatment settings are discussed.

Wastewater-Based Epidemiology for COVID-19: Handling qPCR Nondetects and Comparing Spatially Granular Wastewater and Clinical Data Trends.

Wastewater-based epidemiology (WBE) is a useful complement to clinical testing for managing COVID-19. While community-scale wastewater and clinical data frequently correlate, less is known about subcommunity relationships between the two data types. Moreover, nondetects in qPCR wastewater data are typically handled through methods known to bias results, overlooking perhaps better alternatives. We address these knowledge gaps using data collected from September 2020-June 2021 in Davis, California (USA). We hypothesize that coupling the expectation maximization (EM) algorithm with the Markov Chain Monte Carlo (MCMC) method could improve estimation of "missing" values in wastewater qPCR data. We test this hypothesis by applying EM-MCMC to city wastewater treatment plant data and comparing output to more conventional nondetect handling methods. Dissimilarities in results (i) underscore the importance of specifying nondetect handling method in reporting and (ii) suggest that using EM-MCMC may yield better agreement between community-scale clinical and wastewater data. We also present a novel framework for spatially aligning clinical data with wastewater data collected upstream of a treatment plant (i.e., distributed across a sewershed). Applying the framework to data from Davis reveals reasonable agreement between wastewater and clinical data at highly granular spatial scales-further underscoring the public-health value of WBE.

Can auxiliary indicators improve COVID-19 forecasting and hotspot prediction?

Short-term forecasts of traditional streams from public health reporting (such as cases, hospitalizations, and deaths) are a key input to public health decision-making during a pandemic. Since early 2020, our research group has worked with data partners to collect, curate, and make publicly available numerous real-time COVID-19 indicators, providing multiple views of pandemic activity in the United States. This paper studies the utility of five such indicators-derived from deidentified medical insurance claims, self-reported symptoms from online surveys, and COVID-related Google search activity-from a forecasting perspective. For each indicator, we ask whether its inclusion in an autoregressive (AR) model leads to improved predictive accuracy relative to the same model excluding it. Such an AR model, without external features, is already competitive with many top COVID-19 forecasting models in use today. Our analysis reveals that 1) inclusion of each of these five indicators improves on the overall predictive accuracy of the AR model; 2) predictive gains are in general most pronounced during times in which COVID cases are trending in "flat" or "down" directions; and 3) one indicator, based on Google searches, seems to be particularly helpful during "up" trends.

Adaptive nonparametric regression with the K-nearest neighbour fused lasso.

The fused lasso, also known as total-variation denoising, is a locally adaptive function estimator over a regular grid of design points. In this article, we extend the fused lasso to settings in which the points do not occur on a regular grid, leading to a method for nonparametric regression. This approach, which we call the [Formula: see text]-nearest-neighbours fused lasso, involves computing the [Formula: see text]-nearest-neighbours graph of the design points and then performing the fused lasso over this graph. We show that this procedure has a number of theoretical advantages over competing methods: specifically, it inherits local adaptivity from its connection to the fused lasso, and it inherits manifold adaptivity from its connection to the [Formula: see text]-nearest-neighbours approach. In a simulation study and an application to flu data, we show that excellent results are obtained. For completeness, we also study an estimator that makes use of an [Formula: see text]-graph rather than a [Formula: see text]-nearest-neighbours graph and contrast it with the [Formula: see text]-nearest-neighbours fused lasso.

Proteogenomic Analysis of Surgically Resected Lung Adenocarcinoma.

INTRODUCTION: Despite apparently complete surgical resection, approximately half of resected early-stage lung cancer patients relapse and die of their disease. Adjuvant chemotherapy reduces this risk by only 5% to 8%. Thus, there is a need for better identifying who benefits from adjuvant therapy, the drivers of relapse, and novel targets in this setting. METHODS: RNA sequencing and liquid chromatography/liquid chromatography-mass spectrometry proteomics data were generated from 51 surgically resected non-small cell lung tumors with known recurrence status. RESULTS: We present a rationale and framework for the incorporation of high-content RNA and protein measurements into integrative biomarkers and show the potential of this approach for predicting risk of recurrence in a group of lung adenocarcinomas. In addition, we characterize the relationship between mRNA and protein measurements in lung adenocarcinoma and show that it is outcome specific. CONCLUSIONS: Our results suggest that mRNA and protein data possess independent biological and clinical importance, which can be leveraged to create higher-powered expression biomarkers.