Clinical concern for acute coronary syndrome (ACS) is one of emergency medicine's most common patient encounters. This study aims to develop an ensemble learning-driven framework as a diagnostic support tool to prevent misdiagnosis. We obtained extensive clinical electronic health data on patient encounters with clinical concerns for ACS from a large urban emergency department (ED) between January 2017 and August 2020. We applied an analytical framework equipped with many well-developed algorithms to improve the data quality by addressing missing values, dimensionality reduction, and data imbalance. We trained ensemble learning algorithms to classify patients with ACS or non-ACS etiologies of their symptoms. We used performance evaluation metrics such as accuracy, sensitivity, precision, F1-score, and the area under the receiver operating characteristic (AUROC) to measure the model's performance. The analysis included 31,228 patients, of whom 563 (1.8%) had ACS and 30,665 (98.2%) had alternative diagnoses. Eleven features, including systolic blood pressure, brain natriuretic peptide, chronic heart disease, coronary artery disease, creatinine, glucose, heart attack, heart rate, nephrotic syndrome, red cell distribution width, and troponin level, are reported as significantly contributing risk factors. The proposed framework successfully classifies these cohorts with sensitivity and AUROC as high as 86.3% and 93.3%. Our proposed model's accuracy, precision, specificity, Matthew's correlation coefficient, and F1-score were 85.7%, 86.3%, 93%, 80%, and 86.3%, respectively. Our proposed framework can identify early patients with ACS through further refinement and validation.

The coronavirus disease (COVID-19) outbreak has become a global public health threat. The influx of COVID-19 patients has prolonged the length of stay (LOS) in the emergency department (ED) in the United States. Our objective is to develop a reliable prediction model for COVID-19 patient ED LOS and identify clinical factors, such as age and comorbidities, associated with LOS within a “4-hour target.” Data were collected from an urban, demographically diverse hospital in Detroit for all COVID-19 patients’ ED presentations from March 16 to December 29, 2020. We trained four machine learning models, namely logistic regression (LR), gradient boosting (GB), decision tree (DT), and random forest (RF), across different data processing stages to predict COVID-19 patients with an ED LOS of less than or greater than 4 hours. The analysis is inclusive of 3,301 COVID-19 patients with known ED LOS, and 16 significant clinical factors were incorporated. The GB model outperformed the baseline classifier (LR) and tree-based classifiers (DT and RF) with an accuracy of 85% and F1-score of 0.88 for predicting ED LOS in the testing data. No significant accuracy gains were achieved through further splitting. This study identified key independent factors from a combination of patient demographics, comorbidities, and ED operational data that predicted ED stay in patients with prolonged COVID-19. The prediction framework can serve as a decision-support tool to improve ED and hospital resource planning and inform patients about better ED LOS estimations.

Prostate cancer (PCa), the second leading cause of cancer death in American men, is a relatively slow-growing malignancy with multiple early treatment options. Yet, a significant number of low-risk PCa patients are over-diagnosed and over-treated with significant and long-term quality of life effects. Further, there is ever increasing evidence of metastasis and higher mortality when hormone-sensitive or castration-resistant PCa tumors are treated indistinctively. Hence, the critical need is to discover clinically-relevant and actionable PCa biomarkers by better understanding the biology of PCa. In this paper, we have discovered novel biomarkers of PCa tumors through cross-cancer learning by leveraging the pathological and molecular similarities in the DNA repair pathways of ovarian, prostate, and breast cancer tumors. Cross-cancer disease learning enriches the study population and identifies genetic/phenotypic commonalities that are important across diseases with pathological and molecular similarities. Our results show that ADIRF, SLC2A5, C3orf86, HSPA1B are among the most significant PCa biomarkers, while MTRNR2L1, EEPD1, TEPP and VN1R2 are jointly important biomarkers across prostate, breast and ovarian cancers. Our validation results have further shown that the discovered biomarkers can predict the disease state better than any randomly selected subset of differentially expressed prostate cancer genes.

The COVID-19 pandemic has heightened the existing concern about the uncertainty surrounding patient arrival and the overutilization of resources in emergency departments (EDs). The prediction of variations in patient arrivals is vital for managing limited healthcare resources and facilitating data-driven resource planning. The objective of this study was to forecast ED patient arrivals during a pandemic over different time horizons. A secondary objective was to compare the performance of different forecasting models in predicting ED patient arrivals. We included all ED patient encounters at an urban teaching hospital between January 2019 and December 2020. We divided the data into training and testing datasets and applied univariate and multivariable forecasting models to predict daily ED visits. The influence of COVID-19 lockdown and climatic factors were included in the multivariable models. The model evaluation consisted of the root mean square error (RMSE) and mean absolute error (MAE) over different forecasting horizons. Our exploratory analysis illustrated that monthly and weekly patterns impact daily demand for care. The Holt–Winters approach outperformed all other univariate and multivariable forecasting models for short-term predictions, while the Long Short-Term Memory approach performed best in extended predictions. The developed forecasting models are able to accurately predict ED patient arrivals and peaks during a surge when tested on two years of data from a high-volume urban ED. These short- and long-term prediction models can potentially enhance ED and hospital resource planning.

Analysis of 3D medical imaging data has been a large topic of focus in the area of Machine Learning/Artificial Intelligence, though little work has been done in algorithmic (particularly unsupervised) analysis of neonatal brain MRI’s. A myriad of conditions can manifest at an early age, including neonatal encephalopathy (NE), which can result in lifelong physical consequences. As such, there is a dire need for better biomarkers of NE and other conditions. The objective of the study is to improve identification of anomalies and prognostication of neonatal MRI brain scans. We introduce a framework designed to support the analysis and assessment of neonatal MRI brain scans, the results of which can be used as an aid to neuroradiologists. We explored the efficacy of the framework through iterations of several deep convolutional Autoencoder (AE) unsupervised modeling architectures designed to learn normalcy of the neonatal brain structure. We tested this framework on the developing human connectome project (dHCP) dataset with 97 patients that were previously categorized by severity. Our framework demonstrated the model’s ability to identify and distinguish subtle morphological signatures present in brain structures. Normal and abnormal neonatal brain scans can be distinguished with reasonable accuracy, correctly categorizing them in up to 83% of cases. Most critically, new brain anomalies originally missed during the radiological reading were identified and corroborated by a neuroradiologist. This framework and our modeling approach demonstrate an ability to improve prognostication of neonatal brain conditions and are able to localize new anomalies.