The value of EHR-based assessment of physician competency: An investigative effort with internal medicine physicians.

OBJECTIVE: The purpose of this study was to investigate the potential of developing an EHR-based model of physician competency, named the Skill Deficiency Evaluation Toolkit for Eliminating Competency-loss Trends (Skill-DETECT), which presents the opportunity to use EHR-based models to inform selection of Continued Medical Education (CME) opportunities specifically targeted at maintaining proficiency. METHODS: The IBM Explorys platform provided outpatient Electronic Health Records (EHRs) representing 76 physicians with over 5000 patients combined. These data were used to develop the Skill-DETECT model, a predictive hybrid model composed of a rule-based model, logistic regression model, and a thresholding model, which predicts cognitive clinical skill deficiencies in internal medicine physicians. A three-phase approach was then used to statistically validate the model performance. RESULTS: Subject Matter Expert (SME) panel reviews resulted in a 100% overall approval rate of the rule based model. Area under the receiver-operating characteristic curves calculated for each logistic regression curve resulted in values between 0.76 and 0.92, which indicated exceptional performance. Normality, skewness, and kurtosis were determined and confirmed that the distribution of values output from the thresholding model were unimodal and peaked, which confirmed effectiveness and generalizability. CONCLUSIONS: The validation has confirmed that the Skill-DETECT model has a strong ability to evaluate EHR data and support the identification of internal medicine cognitive clinical skills that are deficient or are of higher likelihood of becoming deficient and thus require remediation, which will allow both physician and medical organizations to fine tune training efforts.

Improved Mental Acuity Forecasting with an Individualized Quantitative Sleep Model.

Sleep impairment significantly alters human brain structure and cognitive function, but available evidence suggests that adults in developed nations are sleeping less. A growing body of research has sought to use sleep to forecast cognitive performance by modeling the relationship between the two, but has generally focused on vigilance rather than other cognitive constructs affected by sleep, such as reaction time, executive function, and working memory. Previous modeling efforts have also utilized subjective, self-reported sleep durations and were restricted to laboratory environments. In the current effort, we addressed these limitations by employing wearable systems and mobile applications to gather objective sleep information, assess multi-construct cognitive performance, and model/predict changes to mental acuity. Thirty participants were recruited for participation in the study, which lasted 1 week. Using the Fitbit Charge HR and a mobile version of the automated neuropsychological assessment metric called CogGauge, we gathered a series of features and utilized the unified model of performance to predict mental acuity based on sleep records. Our results suggest that individuals poorly rate their sleep duration, supporting the need for objective sleep metrics to model circadian changes to mental acuity. Participant compliance in using the wearable throughout the week and responding to the CogGauge assessments was 80%. Specific biases were identified in temporal metrics across mobile devices and operating systems and were excluded from the mental acuity metric development. Individualized prediction of mental acuity consistently outperformed group modeling. This effort indicates the feasibility of creating an individualized, mobile assessment and prediction of mental acuity, compatible with the majority of current mobile devices.

Gold nanorod translocations and charge measurement through solid-state nanopores.

We study translocations of gold nanoparticles and nanorods through silicon nitride nanopores and present a method for determining the surface charge of nanorods from the magnitude of the ionic current change as nanorods pass through the pore. Positively charged nanorods and spherical nanoparticles with average diameters 10 nm and average nanorod lengths between 44 and 65 nm were translocated through 40 nm thick nanopores with diameters between 19 and 27 nm in 1, 10, or 100 mM KCl solutions. Nanorod passage through the nanopores decreases ion current in larger diameter pores, as in the case of typical Coulter counters, but it increases ion current in smaller diameter nanopores, likely because of the interaction of the nanopore's and nanoparticle's double layers. The presented method predicts a surface charge of 26 mC/m(2) for 44 nm long gold nanorods and 18 mC/m(2) for 65 nm long gold nanorods and facilitates future studies of ligand coverage and surface charge effects in anisotropic particles.

Differentiation of short, single-stranded DNA homopolymers in solid-state nanopores.

In the last two decades, new techniques that monitor ionic current modulations as single molecules pass through a nanoscale pore have enabled numerous single-molecule studies. While biological nanopores have recently shown the ability to resolve single nucleotides within individual DNA molecules, similar developments with solid-state nanopores have lagged, due to challenges both in fabricating stable nanopores of similar dimensions as biological nanopores and in achieving sufficiently low-noise and high-bandwidth recordings. Here we show that small silicon nitride nanopores (0.8- to 2-nm diameter in 5- to 8-nm-thick membranes) can resolve differences between ionic current signals produced by short (30 base) ssDNA homopolymers (poly(dA), poly(dC), poly(dT)), when combined with measurement electronics that allow a signal-to-noise ratio of better than 10 to be achieved at 1-MHz bandwidth. While identifying intramolecular DNA sequences with silicon nitride nanopores will require further improvements in nanopore sensitivity and noise levels, homopolymer differentiation represents an important milestone in the development of solid-state nanopores.

Electrically controlled nanoparticle synthesis inside nanopores.

From their realization just over a decade ago, nanopores in silicon nitride membranes have allowed numerous transport-based single-molecule measurements. Here we report the use of these nanopores as subzeptoliter mixing volumes for the controlled synthesis of metal nanoparticles. Particle synthesis is controlled and monitored through an electric field applied across the nanopore membrane, which is positioned so as to separate electrolyte solutions of a metal precursor and a reducing agent. When the electric field drives reactive ions to the nanopore, a characteristic drop in the ion current is observed, indicating the formation of a nanoparticle inside the nanopore. While traditional chemical synthesis relies on temperature and timing to monitor particle growth, here we observe it in real time by monitoring electrical current. We describe the dynamics of gold particle formation in sub-10 nm diameter silicon nitride pores and the effects of salt concentration and additives on the particle's shape and size. The current versus time signal during particle formation in the nanopore is in excellent agreement with the Richards growth curve, indicating an access-limited growth mechanism.

DNA translocation through graphene nanopores.

We report on DNA translocations through nanopores created in graphene membranes. Devices consist of 1-5 nm thick graphene membranes with electron-beam sculpted nanopores from 5 to 10 nm in diameter. Due to the thin nature of the graphene membranes, we observe larger blocked currents than for traditional solid-state nanopores. However, ionic current noise levels are several orders of magnitude larger than those for silicon nitride nanopores. These fluctuations are reduced with the atomic-layer deposition of 5 nm of titanium dioxide over the device. Unlike traditional solid-state nanopore materials that are insulating, graphene is an excellent electrical conductor. Use of graphene as a membrane material opens the door to a new class of nanopore devices in which electronic sensing and control are performed directly at the pore.