No abstract

Reduction of preventable hospital readmissions that result from chronic or acute conditions like stroke, heart failure, myocardial infarction and pneumonia remains a significant challenge for improving the outcomes and decreasing the cost of healthcare delivery in the United States. Patient readmission rates are relatively high for conditions like heart failure (HF) despite the implementation of high-quality healthcare delivery operation guidelines created by regulatory authorities. Multiple predictive models are currently available to evaluate potential 30-day readmission rates of patients. Most of these models are hypothesis driven and repetitively assess the predictive abilities of the same set of biomarkers as predictive features. In this manuscript, we discuss our attempt to develop a data-driven, electronic-medical record-wide (EMR-wide) feature selection approach and subsequent machine learning to predict readmission probabilities. We have assessed a large repertoire of variables from electronic medical records of heart failure patients in a single center. The cohort included 1,068 patients with 178 patients were readmitted within a 30-day interval (16.66% readmission rate). A total of 4,205 variables were extracted from EMR including diagnosis codes (n=1,763), medications (n=1,028), laboratory measurements (n=846), surgical procedures (n=564) and vital signs (n=4). We designed a multistep modeling strategy using the Naïve Bayes algorithm. In the first step, we created individual models to classify the cases (readmitted) and controls (non-readmitted). In the second step, features contributing to predictive risk from independent models were combined into a composite model using a correlation-based feature selection (CFS) method. All models were trained and tested using a 5-fold cross-validation method, with 70% of the cohort used for training and the remaining 30% for testing. Compared to existing predictive models for HF readmission rates (AUCs in the range of 0.6–0.7), results from our EMR-wide predictive model (AUC=0.78; Accuracy=83.19%) and phenome-wide feature selection strategies are encouraging and reveal the utility of such data-driven machine learning. Fine tuning of the model, replication using multi-center cohorts and prospective clinical trial to evaluate the clinical utility would help the adoption of the model as a clinical decision system for evaluating readmission status.

Heritability is essential for understanding the biological causes of disease but requires laborious patient recruitment and phenotype ascertainment. Electronic health records (EHRs) passively capture a wide range of clinically relevant data and provide a resource for studying the heritability of traits that are not typically accessible. EHRs contain next-of-kin information collected via patient emergency contact forms, but until now, these data have gone unused in research. We mined emergency contact data at three academic medical centers and identified 7.4 million familial relationships while maintaining patient privacy. Identified relationships were consistent with genetically derived relatedness. We used EHR data to compute heritability estimates for 500 disease phenotypes. Overall, estimates were consistent with the literature and between sites. Inconsistencies were indicative of limitations and opportunities unique to EHR research. These analyses provide a validation of the use of EHRs for genetics and disease research.

No abstract

In the past ten years, many studies have shown that malignant tissue has been “normalized” in vitro using mechanical signals. We apply the principles of physical oncology (or mechanobiology) in vivo to show the effect of a “constraint field” on tumor growth. The human breast cancer cell line, MDA MB 231, admixed with ferric nanoparticles was grafted subcutaneously in Nude mice. The magnetizable particles rapidly surrounded the growing tumor. Two permanent magnets located on either side of the tumor created a gradient of magnetic field. Magnetic energy is transformed into mechanical energy by the particles acting as “bioactuators”, applying a constraint field and, by consequence, biomechanical stress to the tumor. This biomechanical treatment was applied 2 hours/day during 21 days, from Day 18 to Day 39 following tumor implantation. The study lasted 74 days. Palpable tumor was measured two times a week. There was a significant in vivo difference between the median volume of treated tumors and untreated controls in the mice measured up to D 74 (D 59 + population): (529 [346; 966] mm3 vs 1334 [256; 2106] mm3; p = 0.015), treated mice having smaller tumors. The difference was not statistically significant in the group of mice measured at least to D 59 (D 59 population). On ex vivo examination, the surface of the tumor mass, measured on histologic sections, was less in the treated group, G1, than in the control groups: G2 (nanoparticles, no magnetic field), G3 (magnetic field, no nanoparticles), G4 (no nanoparticles, no magnetic field) in the D 59 population (Median left surface was significantly lower in G1 (5.6 [3.0; 42.4] mm2, p = 0.005) than in G2 (20.8 [4.9; 34.3]), G3 (16.5 [13.2; 23.2]) and G4 (14.8 [1.8; 55.5]); Median right surface was significantly lower in G1 (4.7 [1.9; 29.2] mm2, p = 0.015) than in G2 (25.0 [5.2; 55.0]), G3 (18.0 [14.6; 35.2]) and G4 (12.5 [1.5; 51.8]). There was no statistically significant difference in the day 59+ population. This is the first demonstration of the effect of stress on tumor growth in vivo suggesting that biomechanical intervention may have a high translational potential as a therapy in locally advanced tumors like pancreatic cancer or primary hepatic carcinoma for which no effective therapy is currently available.