Developing a decision support system based on advances in machine learning is one area for strategic innovation in healthcare. Predicting a patient's progression to septic shock is an active field of ...translational research. The goal of this study was to develop a working model of a clinical decision support system for predicting septic shock in an acute care setting for up to 6 h from the time of admission in an integrated healthcare setting.
Clinical data from Electronic Health Record (EHR), at encounter level, were used to build a predictive model for progression from sepsis to septic shock up to 6 h from the time of admission; that is,
,
, and
from admission. Eight different machine learning algorithms (Random Forest, XGBoost, C5.0, Decision Trees, Boosted Logistic Regression, Support Vector Machine, Logistic Regression, Regularized Logistic, and Bayes Generalized Linear Model) were used for model development. Two adaptive sampling strategies were used to address the class imbalance. Data from two sources (clinical and billing codes) were used to define the case definition (septic shock) using the Centers for Medicare & Medicaid Services (CMS) Sepsis criteria. The model assessment was performed using Area under Receiving Operator Characteristics (AUROC), sensitivity, and specificity. Model predictions for each feature window (1, 3 and 6 h from admission) were consolidated.
Retrospective data from April 2005 to September 2018 were extracted from the EHR, Insurance Claims, Billing, and Laboratory Systems to create a dataset for septic shock detection. The clinical criteria and billing information were used to label patients into two classes-septic shock patients and sepsis patients at three different time points from admission, creating two different case-control cohorts. Data from 45,425 unique in-patient visits were used to build 96 prediction models comparing clinical-based definition versus billing-based information as the gold standard. Of the 24 consolidated models (based on eight machine learning algorithms and three feature windows), four models reached an AUROC greater than 0.9. Overall, all the consolidated models reached an AUROC of at least 0.8820 or higher. Based on the AUROC of 0.9483, the best model was based on Random Forest, with a sensitivity of 83.9% and specificity of 88.1%. The sepsis detection window at 6 h outperformed the 1 and 3-h windows. The sepsis definition based on clinical variables had improved performance when compared to the sepsis definition based on only billing information.
This study corroborated that machine learning models can be developed to predict septic shock using clinical and administrative data. However, the use of clinical information to define septic shock outperformed models developed based on only administrative data. Intelligent decision support tools can be developed and integrated into the EHR and improve clinical outcomes and facilitate the optimization of resources in real-time.
Objective
We sought to evaluate the influence of several well‐documented, readily available risk factors that may influence a psychiatric consultant's decision to admit an emergency department (ED) ...patient reporting suicidal ideation for psychiatric hospitalization.
Methods
We conducted a retrospective study of adult patients presenting to six affiliated EDs within Pennsylvania from January 2015 to June 2017. We identified 533 patients reporting current active suicidal ideation and receiving a complete psychiatric consultation. Socio‐demographic characteristics, psychiatric presentation and history, and disposition were collected. Decision tree analysis was conducted with disposition as the outcome.
Results
Four of 27 variables emerged as most influential to decisionmaking, including psychiatric consultant determination of current suicide risk, patient age, current depressive disorder diagnosis, and patient history of physical violence. Likelihood of admission versus discharge ranged from 97% to 58%, depending on the variables considered. Post hoc analysis indicated that current suicide plan, access to means, lack of social support, and suicide attempt history were significantly associated with psychiatric consultant determination of moderate‐to‐high suicide risk, with small‐to‐medium effect sizes emerging.
Conclusions
Only a handful of variables drive disposition decisions for ED patients reporting current active suicidal ideation, with both high and low fidelity decisions made. Patient suicide risk, determined by considering empirically supported risk factors for suicide attempt and death, contributes the greatest influence on a psychiatric consultant's decision to admit. In line with American College of Emergency Physicians (ACEP) recommendations, this study accentuates the importance of using clinical judgment and adjunct measures to determine patient disposition within this population.
Accurate risk assessment is essential for the success of population screening programs in breast cancer. Models with high sensitivity and specificity would enable programs to target more elaborate ...screening efforts to high-risk populations, while minimizing overtreatment for the rest. Artificial intelligence (AI)-based risk models have demonstrated a significant advance over risk models used today in clinical practice. However, the responsible deployment of novel AI requires careful validation across diverse populations. To this end, we validate our AI-based model, Mirai, across globally diverse screening populations.
We collected screening mammograms and pathology-confirmed breast cancer outcomes from Massachusetts General Hospital, USA; Novant, USA; Emory, USA; Maccabi-Assuta, Israel; Karolinska, Sweden; Chang Gung Memorial Hospital, Taiwan; and Barretos, Brazil. We evaluated Uno's concordance index for Mirai in predicting risk of breast cancer at one to five years from the mammogram.
A total of 128,793 mammograms from 62,185 patients were collected across the seven sites, of which 3,815 were followed by a cancer diagnosis within 5 years. Mirai obtained concordance indices of 0.75 (95% CI, 0.72 to 0.78), 0.75 (95% CI, 0.70 to 0.80), 0.77 (95% CI, 0.75 to 0.79), 0.77 (95% CI, 0.73 to 0.81), 0.81 (95% CI, 0.79 to 0.82), 0.79 (95% CI, 0.76 to 0.83), and 0.84 (95% CI, 0.81 to 0.88) at Massachusetts General Hospital, Novant, Emory, Maccabi-Assuta, Karolinska, Chang Gung Memorial Hospital, and Barretos, respectively.
Mirai, a mammography-based risk model, maintained its accuracy across globally diverse test sets from seven hospitals across five countries. This is the broadest validation to date of an AI-based breast cancer model and suggests that the technology can offer broad and equitable improvements in care.
Care deferral is the phenomenon where patients defer or are unable to receive healthcare services, such as seeing doctors, medications or planned surgery. Care deferral can be the result of patient ...decisions, service availability, service limitations, or restrictions due to cost. Continual care deferral in populations may lead to a decline in population health and compound health issues leading to higher social and financial costs in the long term. Consequently, identification of patients who may be at risk of deferring care is important towards improving population health and reducing care total costs. Additionally, minority and vulnerable populations are at a greater risk of care deferral due to socioeconomic factors. In this paper, we (a) address the problem of predicting care deferral for well-care visits; (b) observe that social determinants of health are relevant explanatory factors towards predicting care deferral, and (c) compute how fair the models are with respect to demographics, socioeconomic factors and selected comorbidities. Many health systems currently use rules-based techniques to retroactively identify patients who previously deferred care. The objective of this model is to identify patients at risk of deferring care and allow the health system to prevent care deferrals through direct outreach or social determinant mediation.
The average survival time for untreated primary carcinoma of the lung, from the onset of symptoms, is 8.2 months 1. Lung cancer is the second most common cancer in both men and women. Early detection ...of malignancy and treatment of lung nodules increases chances of survival. In a medium to large sized health system, hundreds of chest, thorax, and abdominal CT scans are performed daily. Many of these scans are looking for problems with organs other than the lung. The radiologist, while reading the CT scans, makes notes about findings on pulmonary nodules - an incidental finding. Additional follow-up is often delayed or missed all together, as the pulmonary nodule was not the chief complaint for the visit or the reason for the imaging study. Geisinger's Close-the-Loop clinical program uses Natural Language Processing (NLP) of radiology notes to isolate pulmonary nodule mentions and the RADS score associated with the nodule to determine the urgency of follow-up recommendations. Close-the-Loop care managers, using information output from the NLP, contact patients for follow-up visits and treatments. The NLP algorithm is complicated by variations in style, content, context, and negations of the text being analyzed by the machine. The other complication arises with processing large number of notes efficiently and in a timely manner. Geisinger's lung nodule NLP solution analyzed millions of radiology notes and found thousands of patients with need for follow-up care with a validated accuracy of 90%.
Esophageal squamous cell carcinoma (ESCC) is the most common histological subtype of esophageal cancer in India. Cigarette smoking and chewing tobacco are known risk factors associated with ESCC. ...However, genomic alterations associated with ESCC in India are not well-characterized. In this study, we carried out exome sequencing to characterize the mutational landscape of ESCC tumors from subjects with a varied history of tobacco usage. Whole exome sequence analysis of ESCC from an Indian cohort revealed several genes that were mutated or had copy number changes. ESCC from tobacco chewers had a higher frequency of C:G > A:T transversions and 2-fold enrichment for mutation signature 4 compared to smokers and non-users of tobacco. Genes, such as
, and
were found to be frequently mutated in Indian cohort. Mutually exclusive mutation patterns were observed in
-
-
-
, and
-
gene pairs. Recurrent amplifications were observed in 3q22-3q29, 11q13.3-q13.4, 7q22.1-q31.1, and 8q24 regions. Approximately 53% of tumors had genomic alterations in
making this pathway a promising candidate for targeted therapy. In conclusion, we observe enrichment of mutation signature 4 in ESCC tumors from patients with a history of tobacco chewing. This is likely due to direct exposure of esophagus to tobacco carcinogens when it is chewed and swallowed. Genomic alterations were frequently observed in PIK3CA-AKT pathway members independent of the history of tobacco usage. PIK3CA pathway can be potentially targeted in ESCC which currently has no effective targeted therapeutic options.
Though smoking remains one of the established risk factors of esophageal squamous cell carcinoma, there is limited data on molecular alterations associated with cigarette smoke exposure in esophageal ...cells. To investigate molecular alterations associated with chronic exposure to cigarette smoke, non-neoplastic human esophageal epithelial cells were treated with cigarette smoke condensate (CSC) for up to 8 months. Chronic treatment with CSC increased cell proliferation and invasive ability of non-neoplastic esophageal cells. Whole exome sequence analysis of CSC treated cells revealed several mutations and copy number variations. This included loss of high mobility group nucleosomal binding domain 2 (HMGN2) and a missense variant in mediator complex subunit 1 (MED1). Both these genes play an important role in DNA repair. Global proteomic and phosphoproteomic profiling of CSC treated cells lead to the identification of 38 differentially expressed and 171 differentially phosphorylated proteins. Bioinformatics analysis of differentially expressed proteins and phosphoproteins revealed that most of these proteins are associated with DNA damage response pathway. Proteomics data revealed decreased expression of HMGN2 and hypophosphorylation of MED1. Exogenous expression of HMGN2 and MED1 lead to decreased proliferative and invasive ability of smoke exposed cells. Immunohistochemical labeling of HMGN2 in primary ESCC tumor tissue sections (from smokers) showed no detectable expression while strong to moderate staining of HMGN2 was observed in normal esophageal tissues. Our data suggests that cigarette smoke perturbs expression of proteins associated with DNA damage response pathways which might play a vital role in development of ESCC.
Tobacco usage is a known risk factor associated with development of oral cancer. It is mainly consumed in two different forms (smoking and chewing) that vary in their composition and methods of ...intake. Despite being the leading cause of oral cancer, molecular alterations induced by tobacco are poorly understood. We therefore sought to investigate the adverse effects of cigarette smoke/chewing tobacco exposure in oral keratinocytes (OKF6/TERT1). OKF6/TERT1 cells acquired oncogenic phenotype after treating with cigarette smoke/chewing tobacco for a period of 8 months. We employed whole exome sequencing (WES) and quantitative proteomics to investigate the molecular alterations in oral keratinocytes chronically exposed to smoke/ chewing tobacco. Exome sequencing revealed distinct mutational spectrum and copy number alterations in smoke/ chewing tobacco treated cells. We also observed differences in proteomic alterations. Proteins downstream of MAPK1 and EGFR were dysregulated in smoke and chewing tobacco exposed cells, respectively. This study can serve as a reference for fundamental damages on oral cells as a consequence of exposure to different forms of tobacco.