Effectiveness of a Cloud-Based EHR Clinical Decision Support Program for Body Mass Index (BMI) Screening and Follow-up.

The effectiveness of a clinical decision support (CDS) program encouraging clinicians to record patient's Body Mass Index (BMI) and document appropriate follow-up plans is evaluated. Test (4,987 practices, 33,445 clinicians) and control groups (881 practices, 6,316 clinicians) were selected using stratified random sampling. Three CDS alerts for BMI screening and follow-up based on evidence based clinical quality guidelines were displayed at the point of care in a cloud-based EHR. The effectiveness of the CDS program was measured over 4 months by tracking recorded BMI and documented follow-up plans. Over the program, BMI recording increased minimally and documentation of follow-up plans increased 5-fold (p=0.05) compared to the control group. The overweight test group patients (18- 64yo) gained less weight (p=0.06) than the control group and underweight patients gained more weight (p<0.01) during the program period. Outcome studies with longer follow-up periods are needed to further confirm positive outcomes.

Syntactic parsing of clinical text: guideline and corpus development with handling ill-formed sentences.

OBJECTIVE: To develop, evaluate, and share: (1) syntactic parsing guidelines for clinical text, with a new approach to handling ill-formed sentences; and (2) a clinical Treebank annotated according to the guidelines. To document the process and findings for readers with similar interest. METHODS: Using random samples from a shared natural language processing challenge dataset, we developed a handbook of domain-customized syntactic parsing guidelines based on iterative annotation and adjudication between two institutions. Special considerations were incorporated into the guidelines for handling ill-formed sentences, which are common in clinical text. Intra- and inter-annotator agreement rates were used to evaluate consistency in following the guidelines. Quantitative and qualitative properties of the annotated Treebank, as well as its use to retrain a statistical parser, were reported. RESULTS: A supplement to the Penn Treebank II guidelines was developed for annotating clinical sentences. After three iterations of annotation and adjudication on 450 sentences, the annotators reached an F-measure agreement rate of 0.930 (while intra-annotator rate was 0.948) on a final independent set. A total of 1100 sentences from progress notes were annotated that demonstrated domain-specific linguistic features. A statistical parser retrained with combined general English (mainly news text) annotations and our annotations achieved an accuracy of 0.811 (higher than models trained purely with either general or clinical sentences alone). Both the guidelines and syntactic annotations are made available at https://sourceforge.net/projects/medicaltreebank. CONCLUSIONS: We developed guidelines for parsing clinical text and annotated a corpus accordingly. The high intra- and inter-annotator agreement rates showed decent consistency in following the guidelines. The corpus was shown to be useful in retraining a statistical parser that achieved moderate accuracy.

Part-of-speech tagging for clinical text: wall or bridge between institutions?

Part-of-speech (POS) tagging is a fundamental step required by various NLP systems. The training of a POS tagger relies on sufficient quality annotations. However, the annotation process is both knowledge-intensive and time-consuming in the clinical domain. A promising solution appears to be for institutions to share their annotation efforts, and yet there is little research on associated issues. We performed experiments to understand how POS tagging performance would be affected by using a pre-trained tagger versus raw training data across different institutions. We manually annotated a set of clinical notes at Kaiser Permanente Southern California (KPSC) and a set from the University of Pittsburg Medical Center (UPMC), and trained/tested POS taggers with intra- and inter-institution settings. The cTAKES POS tagger was also included in the comparison to represent a tagger partially trained from the notes of a third institution, Mayo Clinic at Rochester. Intra-institution 5-fold cross-validation estimated an accuracy of 0.953 and 0.945 on the KPSC and UPMC notes respectively. Trained purely on KPSC notes, the accuracy was 0.897 when tested on UPMC notes. Trained purely on UPMC notes, the accuracy was 0.904 when tested on KPSC notes. Applying the cTAKES tagger pre-trained with Mayo Clinic's notes, the accuracy was 0.881 on KPSC notes and 0.883 on UPMC notes. After adding UPMC annotations to KPSC training data, the average accuracy on tested KPSC notes increased to 0.965. After adding KPSC annotations to UPMC training data, the average accuracy on tested UPMC notes increased to 0.953. The results indicated: first, the performance of pre-trained POS taggers dropped about 5% when applied directly across the institutions; second, mixing annotations from another institution following the same guideline increased tagging accuracy for about 1%. Our findings suggest that institutions can benefit more from sharing raw annotations but less from sharing pre-trained models for the POS tagging task. We believe the study could also provide general insights on cross-institution data sharing for other types of NLP tasks.

Ontogeny of skeletal maturation in the juvenile rat.

Systemic regulation of the cellular processes that produce endochondral elongation and endochondral mineralization during postnatal skeletal maturation are not completely understood. In particular, a mechanism coupling the decline of cellular activity in the bone microenvironment to the onset of sexual maturity remains elusive. The purpose of this study was to empirically integrate the dynamic progression of bone mineral accrual and endochondral elongation as a function of animal age in growing male and female Sprague-Dawley rats. We used serial dual-energy X-ray absorptiometry (DXA) and radiography to study the temporal progression of bone growth and mineral accrual from weaning to adulthood. We observed that skeletal maturation proceeds in a pattern adequately described by the Gompertz function. During this period of growth, we found that serum markers of osteoblastic bone formation declined with age, while osteoclastic bone resorption activity remained unchanged. We also report a slight lag in the age at inflection in the rate of bone mineral accrual relative to the rate of tibial elongation and that both endochondral processes eventually come to asymptotic equilibrium by approximately 20 weeks of age. In addition, we studied tibial growth plate histomorphometry at select time points through 1 year of age. We report that, despite the histologic persistence of physeal cartilage, very little proliferative or elongative activity was measured in this tissue beyond 20 weeks of age. Taken together, these data provide insight to the temporal coordination of postnatal endochondral growth processes.

Combination radioprotectors maintain proliferation better than single agents by decreasing early parathyroid hormone-related protein changes after growth plate irradiation.

Our hypothesis was that combinations of radioprotectors would be more effective than individual agents in minimizing the effects of radiation on the growth plate after single-fraction hind-limb irradiation of Sprague-Dawley rats. At 2 days postirradiation, the decrease in parathyroid hormone-related protein and parathyroid hormone receptor 1 expression in the irradiated growth plate transitional and hypertrophic zones was reversed in both of the combination groups but persisted in the groups treated with the individual drugs. By 2 weeks, positive findings unique to the combination-treatment animals included greater mean proliferation in the irradiated growth plate than on the contralateral side, smaller limb length discrepancies, reversal of the increased overall matrix area fraction, and reversal of the usual deficiency in Indian hedgehog staining in the irradiated hypertrophic zone. While all treatments had a positive effect in reversing the decrease in B-cell leukemia 2 protein and coincident increase in Bax previously observed 2 weeks postirradiation, the two combination groups had a more robust effect. Combinations of radioprotectors may achieve their beneficial additive effects in the growth plate by decreasing the usual early drop in parathyroid hormone-related protein and parathyroid hormone receptor 1 after irradiation, resulting in a cascade of parathyroid hormone-related protein-mediated events.