Using Artificial Intelligence With Natural Language Processing to Combine Electronic Health Record's Structured and Free Text Data to Identify Nonvalvular Atrial Fibrillation to Decrease Strokes and Death: Evaluation and Case-Control Study.

BACKGROUND: Nonvalvular atrial fibrillation (NVAF) affects almost 6 million Americans and is a major contributor to stroke but is significantly undiagnosed and undertreated despite explicit guidelines for oral anticoagulation. OBJECTIVE: The aim of this study is to investigate whether the use of semisupervised natural language processing (NLP) of electronic health record's (EHR) free-text information combined with structured EHR data improves NVAF discovery and treatment and perhaps offers a method to prevent thousands of deaths and save billions of dollars. METHODS: We abstracted 96,681 participants from the University of Buffalo faculty practice's EHR. NLP was used to index the notes and compare the ability to identify NVAF, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category (CHA2DS2-VASc), and Hypertension, Abnormal liver/renal function, Stroke history, Bleeding history or predisposition, Labile INR, Elderly, Drug/alcohol usage (HAS-BLED) scores using unstructured data (International Classification of Diseases codes) versus structured and unstructured data from clinical notes. In addition, we analyzed data from 63,296,120 participants in the Optum and Truven databases to determine the NVAF frequency, rates of CHA2DS2­VASc ≥2, and no contraindications to oral anticoagulants, rates of stroke and death in the untreated population, and first year's costs after stroke. RESULTS: The structured-plus-unstructured method would have identified 3,976,056 additional true NVAF cases (P<.001) and improved sensitivity for CHA2DS2-VASc and HAS-BLED scores compared with the structured data alone (P=.002 and P<.001, respectively), causing a 32.1% improvement. For the United States, this method would prevent an estimated 176,537 strokes, save 10,575 lives, and save >US $13.5 billion. CONCLUSIONS: Artificial intelligence-informed bio-surveillance combining NLP of free-text information with structured EHR data improves data completeness, prevents thousands of strokes, and saves lives and funds. This method is applicable to many disorders with profound public health consequences.

Combating Ebola with Repurposed Therapeutics Using the CANDO Platform.

Ebola virus disease (EVD) is extremely virulent with an estimated mortality rate of up to 90%. However, the state-of-the-art treatment for EVD is limited to quarantine and supportive care. The 2014 Ebola epidemic in West Africa, the largest in history, is believed to have caused more than 11,000 fatalities. The countries worst affected are also among the poorest in the world. Given the complexities, time, and resources required for a novel drug development, finding efficient drug discovery pathways is going to be crucial in the fight against future outbreaks. We have developed a Computational Analysis of Novel Drug Opportunities (CANDO) platform based on the hypothesis that drugs function by interacting with multiple protein targets to create a molecular interaction signature that can be exploited for rapid therapeutic repurposing and discovery. We used the CANDO platform to identify and rank FDA-approved drug candidates that bind and inhibit all proteins encoded by the genomes of five different Ebola virus strains. Top ranking drug candidates for EVD treatment generated by CANDO were compared to in vitro screening studies against Ebola virus-like particles (VLPs) by Kouznetsova et al. and genetically engineered Ebola virus and cell viability studies by Johansen et al. to identify drug overlaps between the in virtuale and in vitro studies as putative treatments for future EVD outbreaks. Our results indicate that integrating computational docking predictions on a proteomic scale with results from in vitro screening studies may be used to select and prioritize compounds for further in vivo and clinical testing. This approach will significantly reduce the lead time, risk, cost, and resources required to determine efficacious therapies against future EVD outbreaks.

A randomized, double-blind, parallel-group, fixed-dose, clinical trial of quetiapine at 600 versus 1200 mg/d for patients with treatment-resistant schizophrenia or schizoaffective disorder.

Quetiapine is often prescribed at doses higher than those approved by regulatory authorities, with limited evidence from controlled trials. The objective of this study was to assess the safety, tolerability, and efficacy of high-dose quetiapine (1200 mg/d) compared with a standard dose of 600 mg/d among patients with Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, schizophrenia or schizoaffective disorder hospitalized at 2 state-operated psychiatric facilities. In order to be eligible for randomization, subjects were required to prospectively fail to demonstrate an initial therapeutic response during a 4-week run-in phase with quetiapine at 600 mg/d (immediate release and dosed twice a day). Lack of an adequate initial response was defined a 15% or lower decrease in the Positive and Negative Syndrome Scale total score. Patients were then randomized to either continue quetiapine at 600 mg/d for an additional 8 weeks or to receive 1200 mg/d quetiapine instead. No significant differences were observed between the high dose (n = 29) and standard dose (n = 31) groups in change from baseline to endpoint on extrapyramidal symptoms, electrocardiographic changes, or most laboratory measures between groups. There was a significant difference between groups for triglycerides (P = 0.035), and post hoc tests revealed a decrease in triglycerides from baseline (mean [SD], 162.7 [59.3] mg/dL) to endpoint (mean [SD], 134.8 [62.7] mg/dL) for the 600 mg/d group (P = 0.019). The mean change in the Positive and Negative Syndrome Scale total score did not differ between groups. In conclusion, quetiapine at 1200 mg/d, although reasonably tolerated, did not confer any advantages over quetiapine at 600 mg/d among patients who had failed to demonstrate an adequate response to a prospective 4-week trial of 600 mg/d.

ChIP-chip analysis of neurexins and other candidate genes for addiction and neuropsychiatric disorders.

Several addiction susceptibility genes have been mapped by linkage and genomewide association. However, functional alleles associated with disease risk have not been identified, with a few possible exceptions. In addition, little is known about the cis- and trans-acting factors involved in regulating their expression. To address these issues, we used a ChIP-chip approach to identify regulatory elements in fetal-brain- targeting genes implicated in addiction and other neuropsychiatric conditions. Our data point to a number of putative regulatory elements, several of which, we show, are functionally significant. Many established or putative regulatory elements map near-disease-associated SNPs. These regions would be of interest to survey for patient-specific functional variants involved in disease susceptibility.

Aborted sudden cardiac death (SCD) in a patient with hypertrophic cardiomyopathy (HCM) with low-risk factors for SCD.

Remarkable advancements have been made in understanding the pathophysiology of hypertrophic cardiomyopathy (HCM), since the first implantable cardioverter defibrillator (ICD) was placed in a human, more than 25 years ago to prevent sudden cardiac death (SCD). ICD has become the cornerstone in the management of HCM, with an ability to change the natural course of this complex disease. American College of Cardiology/American Heart Association guidelines have been charted out to help risk stratify asymptomatic or minimally symptomatic adults with HCM, in order to prevent SCD in such individuals. Our patient with left ventricular outflow tract gradient <30 mm Hg, left ventricular wall thickness < 30 mm, negative medical history of syncope and no family history of SCD, would have been at low risk for SCD.

Resolution of ischemia with ß-blocker in a patient with anomalous right coronary artery.

Anomalous origin of right coronary artery (RCA) has been associated with angina, myocardial infarction and sudden death. Often these patients are referred for surgery. However, long-term ß-blocker therapy instead of surgery has been used in some cases with success. The authors report a case of anomalous RCA where resolution of ischemia with ß-blocker was objectively demonstrated on serial stress testing.

Cardiac rupture in takotsubo cardiomyopathy: a systematic review.

BACKGROUND: Takotsubo cardiomyopathy (TSC) and its complications, such as cardiac rupture (CR), are increasingly being reported in the literature. CR is associated with rapid clinical decline and is uniformly fatal if not surgically repaired. To identify patients who developed CR we performed an analysis of all available indexed cases in the literature and compared them with a control group of patients with TSC without rupture. HYPOTHESIS: Takotsubo cardiomyopathy patients with cardiac rupture do not differ significantly from those without rupture. METHODS: MEDLINE (2009) was searched for all TSC case reports with CR. Eleven case reports were identified. Using a random sampling method, we selected 12 case reports of TSC without rupture (control). We included our patient with TSC with rupture as the 12th case of TSC cohort with CR (CR group). Demographic and clinical characteristics were compared between CR group and control. RESULTS: All patients in the TSC group with rupture were female and were significantly older than controls. TSC group with rupture had significantly higher frequency of ST elevation in lead II and absence of T-wave inversion in lead V5 on hospital admission than controls. Mean ejection fraction, systolic blood pressure, and double product, a measure of oxygen demand, was significantly higher in the rupture group compared to controls. The CR group was associated with less frequent use of ß-blocker as compared to controls. CONCLUSIONS: CR as a complication of TSC could be more common than recognized. Higher double product and ejection fraction suggest higher fluctuation of intracardiac pressure and may cause CR in TSC. Use of ß blockers in TSC may provide protection against CR.

Outcome evaluation of a structured educational wellness program in patients with severe mental illness.

OBJECTIVE: Obesity is increasing at an alarming rate in the United States, as is the obesity rate in patients with schizophrenia. Our study retrospectively evaluated the effectiveness of the Solutions for Wellness and Team Solutions programs, 2 structured educational patient programs, and evaluated the effects on obesity and other metabolic markers in a large, naturalistic inpatient sample. METHOD: Between September 18, 2006, and September 15, 2007, 275 inpatients with DSM-IV-TR-diagnosed chronic mental illness admitted to a tertiary care psychiatric facility were included in the 36-week comprehensive and manualized educational program for healthy lifestyles for patients with chronic mental illness incorporating psychoeducational small-group curricula. Patients were tested before and after each of three 12-week group periods by 30 knowledge-assessment questions, and metabolic markers were recorded at baseline, midpoint, and endpoint. RESULTS: Of the 275 included inpatients, 50.5% completed more than 5 modules, 20.4% completed less than or equal to 2 or fewer modules, and 5.1% completed all 11 modules. Significant increases in scores were observed for 7 of the 11 modules in the knowledge assessments (P < .001). Eighty-seven patients (43.72%) had a body mass index (BMI) >/=30 (indicating obesity) at the start of the program. There was a significant mean weight loss of 4.88 lb (P = .035) together with a significant decrease in mean BMI (P = .045). Patients with diabetes showed a reduction in mean weight of 5.98 lb. Significant reductions were observed in glucose and triglyceride levels (both P < .05). Patients with impaired glucose tolerance showed a significantly greater decrease in glucose level (P = .000). Sixty-nine patients (25.46%) met criteria for metabolic syndrome at baseline, and this number was reduced to 53 patients (19.56%) at endpoint; this decrease was significant (P = .027). Regarding relationship of change in knowledge after completion of the modules and metabolic changes, we found a significant correlation between reduction in weight and change in Fitness and Exercise score (r = 0.62, P = .001) and a significant correlation between the change score on Nutrition/Healthy Lifestyles and change in glucose values (r = 0.56, P = .001). CONCLUSIONS: We found that a structured wellness program using a psychoeducational curriculum can be successfully implemented in a large, naturalistic psychiatric setting with unselected, chronically mentally ill inpatients. Results may help both clinicians and hospital managers to implement similar programs or to include successful components in existing programs for psychiatric patients.

Interface for electronic data capture systems for clinical trials by optimal utilization of available hospital resources.

We describe Clinical Trials System (CTS), an innovative EDC system utilizing data from existing hospital-based electronic databases that supports information gathering and storing for various clinical trials. The complexities of designing electronic clinical trials systems and their ideal features are outlined. CTS optimally utilizes existing electronic databases in a well-organized and easy-to-reference format. CTS is currently incorporated within a large psychiatric center, allowing easy sharing of information and data among multidisciplinary clinical and research teams.