Search

To develop and evaluate a high-dimensional, data-driven model to identify patients at high risk of clinical deterioration from routinely collected electronic health record (EHR) data.In this single-center, retrospective cohort study, 488 patients with single-ventricle and shunt-dependent congenital heart disease6 months old were admitted to the cardiac intensive care unit before stage 2 palliation between 2014 and 2019. Using machine-learning techniques, we developed the Intensive care Warning Index (I-WIN), which systematically assessed 1028 regularly collected EHR variables (vital signs, medications, laboratory tests, and diagnoses) to identify patients in the cardiac intensive care unit at elevated risk of clinical deterioration. An ensemble of 5 extreme gradient boosting models was developed and validated on 203 cases (130 emergent endotracheal intubations, 34 cardiac arrests requiring cardiopulmonary resuscitation, 10 extracorporeal membrane oxygenation cannulations, and 29 cardiac arrests requiring cardiopulmonary resuscitation onto extracorporeal membrane oxygenation) and 378 control periods from 446 patients.At 4 hours before deterioration, the model achieved an area under the receiver operating characteristic curve of 0.92 (95% confidence interval, 0.84-0.98), 0.881 sensitivity, 0.776 positive predictive value, 0.862 specificity, and 0.571 Brier skill score. Performance remained high at 8 hours before deterioration with 0.815 (0.688-0.921) area under the receiver operating characteristic curve.I-WIN accurately predicted deterioration events in critically-ill infants with high-risk congenital heart disease up to 8 hours before deterioration, potentially allowing clinicians to target interventions. We propose a paradigm shift from conventional expert consensus-based selection of risk factors to a data-driven, machine-learning methodology for risk prediction. With the increased availability of data capture in EHRs, I-WIN can be extended to broader applications in data-rich environments in critical care.

Social determinants of health (SDOH) are non-medical factors that can profoundly impact patient health outcomes. However, SDOH are rarely available in structured electronic health record (EHR) data such as diagnosis codes, and more commonly found in unstructured narrative clinical notes. Hence, identifying social context from unstructured EHR data has become increasingly important. Yet, previous work on using natural language processing to automate extraction of SDOH from text (a) usually focuses on an ad hoc selection of SDOH, and (b) does not use the latest advances in deep learning. Our objective was to advance automatic extraction of SDOH from clinical text by (a) systematically creating a set of SDOH based on standard biomedical and psychiatric ontologies, and (b) training state-of-the-art deep neural networks to extract mentions of these SDOH from clinical notes.A retrospective cohort study.Data were extracted from the Medical Information Mart for Intensive Care (MIMIC-III) database. The corpus comprised 3,504 social related sentences from 2,670 clinical notes.We developed a framework for automated classification of multiple SDOH categories. Our dataset comprised narrative clinical notes under the "Social Work" category in the MIMIC-III Clinical Database. Using standard terminologies, SNOMED-CT and DSM-IV, we systematically curated a set of 13 SDOH categories and created annotation guidelines for these. After manually annotating the 3,504 sentences, we developed and tested three deep neural network (DNN) architectures - convolutional neural network (CNN), long short-term memory (LSTM) network, and the Bidirectional Encoder Representations from Transformers (BERT) - for automated detection of eight SDOH categories. We also compared these DNNs to three baselines models: (1) cTAKES, as well as (2) L2-regularized logistic regression and (3) random forests on bags-of-words. Model evaluation metrics included micro- and macro- F1, and area under the receiver operating characteristic curve (AUC).All three DNN models accurately classified all SDOH categories (minimum micro-F1 = 0.632, minimum macro-AUC = 0.854). Compared to the CNN and LSTM, BERT performed best in most key metrics (micro-F1 = 0.690, macro-AUC = 0.907). The BERT model most effectively identified the "occupational" category (F1 = 0.774, AUC = 0.965) and least effectively identified the "non-SDOH" category (F = 0.491, AUC = 0.788). BERT outperformed cTAKES in distinguishing social vs non-social sentences (BERT F1 = 0.87 vs. cTAKES F1 = 0.06), and outperformed logistic regression (micro-F1 = 0.649, macro-AUC = 0.696) and random forest (micro-F1 = 0.502, macro-AUC = 0.523) trained on bag-of-words.Our study framework with DNN models demonstrated improved performance for efficiently identifying a systematic range of SDOH categories from clinical notes in the EHR. Improved identification of patient SDOH may further improve healthcare outcomes.

With the development of Internet of Things and the increasingly rampant malicious network activities, higher requirements are put forward for security to detect malicious behavior and prevent attackers from obtaining sensitive data in the smart home environment. In this paper, an intrusion detection system is proposed to detect and classify abnormal behavior in the smart home environment. The two-layer feature processing method based on random forest and principal component analysis can reduce the loss of data information and is suitable for massive data. The three-layer detection model can detect four common attacks with binary classifiers and effectively improve the accuracy. The experimental evaluation of the proposed model is conducted using the real smart home traffic dataset and achieves a classification accuracy of 95.90%. The experimental results show that our model has a good performance in detecting and classifying malicious attacks in the smart home.

This work describes a sensitive electrochemical biosensor for detection of carbamate pesticides based on immobilization of acetylcholinesterase (AChE) on the electrochemically inducing porous graphene oxide network (e-pGON) which is prepared by scanning the GO modified electrode using successive cyclic voltammetry method. The e-pGON effectively promotes the electron transfer rate and facilitates the access of substrates to the active centers. The as-prepared biosensor shows high affinity to acetylthiocholine (ATCl) with a Michaelis-Menten constant value of 0.45 mmol L −1 . Under optimum conditions, the inhibition of carbaryl is proportional to its concentration ranging from 0.3 to 6.1 ng/mL. The detection limit is 0.15 ng/mL. The developed biosensor exhibits good performance such as reproducibility and stability, thus providing a promising tool for the analysis of enzyme inhibitors.

The rapid development of data analysis technologies and the easily accessible datasets have enabled the construction of a comprehensive analytics model, which can facilitate the decision makings involved in services. Meanwhile, the individual privacy preservation is of great necessity. Decision tree is a common method in medical prediction and diagnose, known for its simplicity of understanding and interpreting. However, the process of building a decision tree might cause individual privacy disclosure. Differential privacy provides a rigorous mathematical definition of privacy by controlling the risk of privacy leakage in a manageable range while maintaining the statistical characteristics. In this paper, we propose a Differentially Private Greedy Decision Forest with high utility (DPGDF) to build a privacy-preserving decision forest. In DPGDF, we design a novel budget allocation strategy that allows the nodes in greater depth get more privacy budgets in the decision tree construction process, which can, to some extent, mitigate the problem of excessive noises introduced to the leaf nodes. To aggregate multiple trees into a forest, we propose a selective aggregation method based on the prediction accuracy of the decision forest. In addition, we develop an iterative method to speed up the process of selective aggregation. Finally, we experimentally prove that the proposed DPGDF can achieve a better performance on two practical datasets compared with other algorithms.

In many electronic applications, the surface properties of hydrogen-terminated diamond films determine the eventual performance of the electronic device. In this work, diamond films were grown by hot-filament chemical vapor deposition method and hydrogenated films were obtained by hydrogen plasma treatment. X-ray photoelectron spectroscopy analysis was carried out to evaluate the film surface after annealing treatment. The C1s XPS spectrum showed the C1s spectrum changed after annealing treatment. In air atmosphere, hydrogen decreased and oxygen absorption increased. The interaction on the surface changed drastically after annealed in argon atmosphere and phase transition would happen when the annealing temperature increased to 600 °C.

High quality freestanding diamond (FSD) films with smooth nucleation surfaces were grown by microwave plasma chemical vapor deposition (MPCVD) method. A p-type hydrogenated surface conductivity of FSD film was obtained by using hydrogen plasma treatment. The annealing process in vacuum on the p-type behavior of FSD nucleation surfaces was investigated by Hall effect measurement. H-terminated diamond phototransistors were fabricated and the results suggest that they may be ideally suited for ultraviolet (UV) switching applications.