The coronavirus disease 2019 (COVID-19) pandemic challenges hospital leaders to make time-sensitive, critical decisions about clinical operations and resource allocations.
To estimate the timing of ...surges in clinical demand and the best- and worst-case scenarios of local COVID-19-induced strain on hospital capacity, and thus inform clinical operations and staffing demands and identify when hospital capacity would be saturated.
Monte Carlo simulation instantiation of a susceptible, infected, removed (SIR) model with a 1-day cycle.
3 hospitals in an academic health system.
All people living in the greater Philadelphia region.
The COVID-19 Hospital Impact Model (CHIME) (http://penn-chime.phl.io) SIR model was used to estimate the time from 23 March 2020 until hospital capacity would probably be exceeded, and the intensity of the surge, including for intensive care unit (ICU) beds and ventilators.
Using patients with COVID-19 alone, CHIME estimated that it would be 31 to 53 days before demand exceeds existing hospital capacity. In best- and worst-case scenarios of surges in the number of patients with COVID-19, the needed total capacity for hospital beds would reach 3131 to 12 650 across the 3 hospitals, including 338 to 1608 ICU beds and 118 to 599 ventilators.
Model parameters were taken directly or derived from published data across heterogeneous populations and practice environments and from the health system's historical data. CHIME does not incorporate more transition states to model infection severity, social networks to model transmission dynamics, or geographic information to account for spatial patterns of human interaction.
Publicly available and designed for hospital operations leaders, this modeling tool can inform preparations for capacity strain during the early days of a pandemic.
University of Pennsylvania Health System and the Palliative and Advanced Illness Research Center.
Develop and implement a machine learning algorithm to predict severe sepsis and septic shock and evaluate the impact on clinical practice and patient outcomes.
Retrospective cohort for algorithm ...derivation and validation, pre-post impact evaluation.
Tertiary teaching hospital system in Philadelphia, PA.
All non-ICU admissions; algorithm derivation July 2011 to June 2014 (n = 162,212); algorithm validation October to December 2015 (n = 10,448); silent versus alert comparison January 2016 to February 2017 (silent n = 22,280; alert n = 32,184).
A random-forest classifier, derived and validated using electronic health record data, was deployed both silently and later with an alert to notify clinical teams of sepsis prediction.
Patients identified for training the algorithm were required to have International Classification of Diseases, 9th Edition codes for severe sepsis or septic shock and a positive blood culture during their hospital encounter with either a lactate greater than 2.2 mmol/L or a systolic blood pressure less than 90 mm Hg. The algorithm demonstrated a sensitivity of 26% and specificity of 98%, with a positive predictive value of 29% and positive likelihood ratio of 13. The alert resulted in a small statistically significant increase in lactate testing and IV fluid administration. There was no significant difference in mortality, discharge disposition, or transfer to ICU, although there was a reduction in time-to-ICU transfer.
Our machine learning algorithm can predict, with low sensitivity but high specificity, the impending occurrence of severe sepsis and septic shock. Algorithm-generated predictive alerts modestly impacted clinical measures. Next steps include describing clinical perception of this tool and optimizing algorithm design and delivery.
This study aimed to develop and validate a claims-based, machine learning algorithm to predict clinical outcomes across both medical and surgical patient populations. This retrospective, ...observational cohort study, used a random 5% sample of 770,777 fee-for-service Medicare beneficiaries with an inpatient hospitalization between 2009-2011. The machine learning algorithms tested included: support vector machine, random forest, multilayer perceptron, extreme gradient boosted tree, and logistic regression. The extreme gradient boosted tree algorithm outperformed the alternatives and was the machine learning method used for the final risk model. Primary outcome was 30-day mortality. Secondary outcomes were: rehospitalization, and any of 23 adverse clinical events occurring within 30 days of the index admission date. The machine learning algorithm performance was evaluated by both the area under the receiver operating curve (AUROC) and Brier Score. The risk model demonstrated high performance for prediction of: 30-day mortality (AUROC = 0.88; Brier Score = 0.06), and 17 of the 23 adverse events (AUROC range: 0.80-0.86; Brier Score range: 0.01-0.05). The risk model demonstrated moderate performance for prediction of: rehospitalization within 30 days (AUROC = 0.73; Brier Score: = 0.07) and six of the 23 adverse events (AUROC range: 0.74-0.79; Brier Score range: 0.01-0.02). The machine learning risk model performed comparably on a second, independent validation dataset, confirming that the risk model was not overfit. We have developed and validated a robust, claims-based, machine learning risk model that is applicable to both medical and surgical patient populations and demonstrates comparable predictive accuracy to existing risk models.
Small molecules play a pivotal role in regulating physiological processes and serve as biomarkers to uncover pathological conditions and the effects of therapeutic treatments. However, it remains a ...significant challenge to detect small molecules given the size as compared to macromolecules. Recently, the newly emerging plasmonic immunoassays based on surface‐enhanced Raman scattering (SERS) offer great promise to deliver extraordinary sensitivity. Nevertheless, they are limited by the intrinsic SERS intensity fluctuations associated with the SERS uncertainty principle. The single transducer that relies on the intensity change is also prone to false signals. Additionally, the prevailing sandwich immunoassay format proves less effective towards detecting small molecules. To circumvent these critical issues, a dual‐modal single‐antibody approach that synergizes both the intensity and shift of the peak‐based immunoassay with Raman enhancement, coined as the INSPIRE assay, is developed for small molecules detection. With two independent transduction mechanisms, it allows better prediction of analyte concentration and attenuation of signal artifacts, providing a new and robust strategy for molecular analysis. With a proof‐of‐concept demonstration for detection of free T4 and testosterone in serum matrix, the authors envision that the INSPIRE assay could be expanded for a wide spectrum of applications in biomedical diagnosis, discovery of new biopharmaceuticals, food safety, and environmental monitoring.
A dual‐modal single‐antibody plasmonic spectro‐immunoassay is developed for the detection of small molecules. By harnessing both the intensity and frequency shift in surface‐enhanced Raman scattering, the newly developed spectro‐immunoassay allows better prediction of analyte concentration and attenuation of signal artifacts, providing a new and robust strategy for molecular analysis.
Enhancing light‐matter interactions is fundamental to the advancement of nanophotonics and optoelectronics. Yet, light diffraction on dielectric platforms and energy loss on plasmonic metallic ...systems present an undesirable trade‐off between coherent energy exchange and incoherent energy damping. Through judicious structural design, both light confinement and energy loss issues could be potentially and simultaneously addressed by creating bound states in the continuum (BICs) where light is ideally decoupled from the radiative continuum. Herein, the authors present a general framework based on the two‐coupled resonances to first conceptualize and then numerically demonstrate a type of quasi‐BICs that can be achieved through the interference between two bare resonance modes and is characterized by the considerably narrowed spectral line shape even on lossy metallic nanostructures. The ubiquity of the proposed framework further allows the paradigm to be extended for the realization of plexcitonic quasi‐BICs on the same metallic systems. Owing to the topological nature, both plasmonic and plexcitonic quasi‐BICs display strong mode robustness against parameters variation, thereby providing an attractive platform to unlock the potential of the coupled plasmon‐exciton systems for manipulation of the photophysical properties of condensed phases.
A general framework based on two‐coupled resonances is presented to conceptualize a type of quasi‐bound states in the continuum (quasi‐BICs). Through judicious structural design, both plasmonic and plexcitonic quasi‐BICs are numerically demonstrated on lossy metallic systems in the visible wavelength range. This study provides a promising platform for studying coupled plasmon‐exciton systems for manipulating photophysical properties of condensed phases.
This study entails the use of meta-analytic techniques to calculate and analyze 18 independent and 52 nonindependent effect sizes across 17 published studies of psychological assessment as a ...therapeutic intervention. In this sample of studies, which involves 1,496 participants, a significant overall Cohen's
d
effect size of 0.423 (95% CI 0.321, 0.525) was found, whereby 66% of treatment group means fell above the control and comparison group means. When categorical variables were taken into account, significant treatment group effects were found for therapy process variables (
d
= 1.117, 0.679, 1.555), therapy outcomes (
d
= 0.367, 0.256, 0.478), and combined process/outcome variables (
d
= 0.547, 0.193, 0.901). These findings appear to be robust on the basis of fail-safe
N
calculations. Taken together, they suggest that psychological assessment procedures-when combined with personalized, collaborative, and highly involving test feedback-have positive, clinically meaningful effects on treatment, especially regarding treatment processes. They also have important implications for assessment practice, training, and policy making, as well as future research, which are discussed in the conclusion of the article.
Functional disorders of the thyroid remain a global challenge and have profound impacts on human health. Serving as the barometer for thyroid function, thyroid‐stimulating hormone (TSH) is considered ...the single most useful test of thyroid function. However, the prevailing TSH immunoassays rely on two types of antibodies in a sandwich format. The requirement of repeated incubation and washing further complicates the issue, making it unable to meet the requirements of the shifting public health landscape that demands rapid, sensitive, and low‐cost TSH tests. Herein, a systematic study is performed to investigate the clinical translational potential of a single antibody‐based biosensing platform for the TSH test. The biosensing platform leverages Raman spectral variations induced by the interaction between a TSH antigen and a Raman molecule‐conjugated TSH antibody. In conjunction with machine learning, it allows TSH concentrations in various patient samples to be predicted with high accuracy and precision, which is robust against substrate‐to‐substrate, intra‐substrate, and day‐to‐day variations. It is envisioned that the simplicity and generalizability of this single‐antibody immunoassay coupled with the demonstrated performance in patient samples pave the way for it to be widely applied in clinical settings for low‐cost detection of hormones, other molecular biomarkers, DNA, RNA, and pathogens.
A single antibody‐based biosensing platform is presented for detecting thyroid‐stimulating hormone (TSH) in patient samples. The biosensing platform leverages the Raman spectral variations induced by the interaction between a TSH antigen and a Raman molecule‐conjugated TSH antibody. In conjunction with machine learning, it allows TSH concentrations in various patient samples to be predicted with high accuracy and precision.
While immunoassays are pivotal to medical diagnosis and bioanalytical chemistry, the current landscape of public health has catalyzed an important shift in the requirements of immunoassays that ...demand innovative solutions. For example, rapid, label‐free, and low‐cost screening of a given analyte is required to inform the best countermeasures to combat infectious diseases in a timely manner. Yet, the current design of immunoassays cannot accommodate such requirements as constraint by accumulative challenges, such as repeated incubation and washing, and the need of two types of antibodies in the sandwich format. To provide a potential solution, herein, a plasmonic Raman immunoassay with single‐antibody, multivariate regression, and shift‐of‐peak strategies, coined as the PRISM assay, for serum biomarkers detection, is reported. The PRISM assay relies on Raman reporter‐antibody conjugates to capture analytes on a plasmonic substrate. The ensuing nanomechanical perturbations to vibration of Raman reporters induce subtle but characteristic spectral changes that encode rich information related to the captured analytes. By fusing Raman spectroscopy and chemometric analysis, both Raman frequency shift‐ and multivariate regression models for sensitive detection of biomarkers are developed. The PRISM assay is expected to find a wide range of applications in clinical diagnosis, food safety surveillance, and environmental monitoring.
Leveraging nanomechanical perturbations in Raman spectro‐immunoassays, a versatile serum biomarker detection platform is designed. The platform allows for the detection of a wide range of biomarkers spanning a spectrum of molecular weights in various sample matrices.