Scalable and accurate deep learning with electronic health records.

Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient's record. We propose a representation of patients' entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24 h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93-0.94), 30-day unplanned readmission (AUROC 0.75-0.76), prolonged length of stay (AUROC 0.85-0.86), and all of a patient's final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient's chart.

Double-blind randomized controlled trial of low-level laser therapy in carpal tunnel syndrome.

Several studies have suggested that low-level laser therapy (LLLT) is effective in patients with carpal tunnel syndrome (CTS). In a double-blind randomized controlled trial of LLLT, 15 CTS patients, 34 to 67 years of age, were randomly assigned to either the control group (n = 8) or treatment group (n =7). Both groups were treated three times per week for 5 weeks. Those in the treatment group received 860 nm galium/aluminum/arsenide laser at a dosage of 6 J/cm2 over the carpal tunnel, whereas those in the control group were treated with sham laser. The primary outcome measure was the Levine Carpal Tunnel Syndrome Questionnaire, and the secondary outcome measures were electrophysiological data and the Purdue pegboard test. All patients completed the study without adverse effects. There was a significant symptomatic improvement in both the control (P = 0.034) and treatment (P =0.043) groups. However, there was no significant difference in any of the outcome measures between the two groups. Thus, LLLT is no more effective in the reduction of symptoms of CTS than is sham treatment.