Faculty Bibliography

[This corrects the article DOI: 10.1371/journal.pone.0215571.].

BACKGROUND:Because of the strong link between childhood obesity and adulthood obesity comorbidities, and the difficulty in decreasing body mass index (BMI) later in life, effective strategies are needed to address this condition in early childhood. The ability to predict obesity before age five could be a useful tool, allowing prevention strategies to focus on high risk children. The few existing prediction models for obesity in childhood have primarily employed data from longitudinal cohort studies, relying on difficult to collect data that are not readily available to all practitioners. Instead, we utilized real-world unaugmented electronic health record (EHR) data from the first two years of life to predict obesity status at age five, an approach not yet taken in pediatric obesity research. METHODS AND FINDINGS/RESULTS:We trained a variety of machine learning algorithms to perform both binary classification and regression. Following previous studies demonstrating different obesity determinants for boys and girls, we similarly developed separate models for both groups. In each of the separate models for boys and girls we found that weight for length z-score, BMI between 19 and 24 months, and the last BMI measure recorded before age two were the most important features for prediction. The best performing models were able to predict obesity with an Area Under the Receiver Operator Characteristic Curve (AUC) of 81.7% for girls and 76.1% for boys. CONCLUSIONS:We were able to predict obesity at age five using EHR data with an AUC comparable to cohort-based studies, reducing the need for investment in additional data collection. Our results suggest that machine learning approaches for predicting future childhood obesity using EHR data could improve the ability of clinicians and researchers to drive future policy, intervention design, and the decision-making process in a clinical setting.

OBJECTIVES/OBJECTIVE:Black race has been associated with increased 30-day morbidity and mortality following surgery for endometrial cancer. Black women are also less likely to undergo laparoscopy when compared to white women. With the development of improved laparoscopic techniques and equipment, including the robotic platform, we sought to evaluate whether there has been a change in surgical approach for black women, and in turn, improvement in perioperative outcomes. METHODS:Using the American College of Surgeons' National Surgical Quality Improvement Project's database, patients who underwent hysterectomy for endometrial cancer from 2010 to 2015 were identified. Comparative analyses stratified by race and hysterectomy approach were performed to assess the relationship between race and perioperative outcomes. RESULTS:A total of 17,692 patients were identified: of these, 13,720 (77.5%) were white and 1553 (8.8%) were black. Black women were less likely to undergo laparoscopic hysterectomy compared to white women (49.3% vs 71.3%, p<0.0001). Rates of laparoscopy in both races increased over the 6-year period; however these consistently remained lower in black women each year. Black women had higher 30-day postoperative complication rates compared to white women (22.5% vs 13.6%, p<0.0001). When laparoscopic hysterectomies were isolated, there was no difference in postoperative complication rates between black and white women (9.2% vs 7.5%, p=0.1). CONCLUSIONS:Overall black women incur more postoperative complications compared to white women undergoing hysterectomy for endometrial cancer. However, laparoscopy may mitigate this disparity. Efforts should be made to maximize the utilization of minimally invasive surgery for the surgical management of endometrial cancer.

Conventional text classification models make a bag-of-words assumption reducing text into word occurrence counts per document. Recent algorithms such as word2vec are capable of learning semantic meaning and similarity between words in an entirely unsupervised manner using a contextual window and doing so much faster than previous methods. Each word is projected into vector space such that similar meaning words such as "strong" and "powerful" are projected into the same general Euclidean space. Open questions about these embeddings include their utility across classification tasks and the optimal properties and source of documents to construct broadly functional embeddings. In this work, we demonstrate the usefulness of pre-trained embeddings for classification in our task and demonstrate that custom word embeddings, built in the domain and for the tasks, can improve performance over word embeddings learnt on more general data including news articles or Wikipedia.

Natural Language Processing (NLP) holds potential for patient care and clinical research, but a gap exists between promise and reality. While some studies have demonstrated portability of NLP systems across multiple sites, challenges remain. Strategies to mitigate these challenges can strive for complex NLP problems using advanced methods (hard-to-reach fruit), or focus on simple NLP problems using practical methods (low-hanging fruit). This paper investigates a practical strategy for NLP portability using extraction of left ventricular ejection fraction (LVEF) as a use case. We used a tool developed at the Department of Veterans Affair (VA) to extract the LVEF values from free-text echocardiograms in the MIMIC-III database. The approach showed an accuracy of 98.4%, sensitivity of 99.4%, a positive predictive value of 98.7%, and F-score of 99.0%. This experience, in which a simple NLP solution proved highly portable with excellent performance, illustrates the point that simple NLP applications may be easier to disseminate and adapt, and in the short term may prove more useful, than complex applications.

BACKGROUND AND OBJECTIVES: Missed visits are common in office-based buprenorphine treatment (OBOT). The feasibility of text message (TM) appointment reminders among OBOT patients is unknown. METHODS: This 6-month prospective cohort study provided TM reminders to OBOT program patients (N = 93). A feasibility survey was completed following delivery of TM reminders and at 6 months. RESULTS: Respondents reported that the reminders should be provided to all OBOT patients (100%) and helped them to adhere to their scheduled appointment (97%). At 6 months, there were no reports of intrusion to their privacy or disruption of daily activities due to the TM reminders. Most participants reported that the TM reminders were helpful in adhering to scheduled appointments (95%), that the reminders should be offered to all clinic patients (95%), and favored receiving only TM reminders rather than telephone reminders (95%). Barriers to adhering to scheduled appointment times included transportation difficulties (34%), not being able to take time off from school or work (31%), long clinic wait-times (9%), being hospitalized or sick (8%), feeling sad or depressed (6%), and child care (6%). CONCLUSIONS: This study demonstrated the acceptability and feasibility of TM appointment reminders in OBOT. Older age and longer duration in buprenorphine treatment did not diminish interest in receiving the TM intervention. Although OBOT patients expressed concern regarding the privacy of TM content sent from their providers, privacy issues were uncommon among this cohort. Scientific Significance Findings from this study highlighted patient barriers to adherence to scheduled appointments. These barriers included transportation difficulties (34%), not being able to take time off from school or work (31%), long clinic lines (9%), and other factors that may confound the effect of future TM appointment reminder interventions. Further research is also required to assess 1) the level of system changes required to integrate TM appointment reminder tools with already existing electronic medical records and appointment records software; 2) acceptability among clinicians and administrators; and 3) financial and resource constraints to healthcare systems. (Am J Addict 2017;XX:1-6).

BACKGROUND: At NYU, as part of a comprehensive objective structured clinical skills exam, experienced medical educators judge clinical knowledge, decision-making, and clinical reasoning skills of trainees based on their patient notes. Despite being rubric-driven, this task requires tremendous time and effort to establish consistent scoring, delaying and limiting individualized feedback. We conducted pilot machine learning text classification studies to establish if accurate automated scoring of clinical notes is possible. METHODS: As a use case, we tested 100 student written clinical notes from7 standardized patient cases (Vision Loss, Tel Diarrhea, Difficulty Sleeping, Shoulder Pain, Failure To Thrive, Abdominal, Pain, Palpitations) that had been scored for quality of clinical reasoning by faculty on a 1-4 scale. In order to assess performance of NLP strategies to categorize students in meaningful groups we dichotomized students based on their faculty given scores by case into "failing" (score of 1, 5-18 students per case) and "passing" (score 2,3,4). We treated each task as a binary classification task in a text classification pipeline. First, we treated each note as a bag of tokens and weight each token with term frequency-inverse document frequency (TFIDF) a numerical statistic that reflects howimportant aword is to a document. We then applied 3 different classification algorithms (random forests, support vector machines, and Bayesian logistic regression) and measured discriminatory performance using Area Under Curve (AUC) in a cross validation evaluation design. RESULTS: TFDIF performed with AUCs between 0.669 and 0.905. Logistic regression provided the highestAUC in four cases: Difficulty Sleeping (0.905), Shoulder Pain (0.618), Failure To Thrive (0.717) and Abdominal Pain (0.892). As we observed the highest AUCs in Difficulty Sleeping and Abdominal Pain cases, we have begun to refine the algorithm for these two cases by identifying the importance features that lead faculty to give students to a higher grade and improve the accuracy of NLP based scoring. Promising features include the presence and sequence of certainwords in the problem representation, sentence length in the management section, ranking of the differential diagnosis, sequence between key words (e.g. rule out appendicitis), and evidence of "thinkingness" or what many call semantic qualifiers. CONCLUSIONS: With additional effort to build targeted case specific classifiers for clinical content and reasoning, a validated machine-learning model may achieve partial or full automation of grading of the notes. This work, which builds on decades of clinical decision-making and critical reasoning research, may provide medical trainees with more and potentially better feedback; facilitating learning of clinical reasoning, freeing faculty to coach this process, and in the long run impacting healthcare quality and patient safety

This article reviews examples of big data analyses in health care with a focus on radiology. We review the defining characteristics of big data, the use of natural language processing, traditional and novel data sources, and large clinical data repositories available for research. This article aims to invoke novel research ideas through a combination of examples of analyses and domain knowledge.

RATIONALE AND OBJECTIVES: This study aimed to assess the performance of a text classification machine-learning model in predicting highly cited articles within the recent radiological literature and to identify the model's most influential article features. MATERIALS AND METHODS: We downloaded from PubMed the title, abstract, and medical subject heading terms for 10,065 articles published in 25 general radiology journals in 2012 and 2013. Three machine-learning models were applied to predict the top 10% of included articles in terms of the number of citations to the article in 2014 (reflecting the 2-year time window in conventional impact factor calculations). The model having the highest area under the curve was selected to derive a list of article features (words) predicting high citation volume, which was iteratively reduced to identify the smallest possible core feature list maintaining predictive power. Overall themes were qualitatively assigned to the core features. RESULTS: The regularized logistic regression (Bayesian binary regression) model had highest performance, achieving an area under the curve of 0.814 in predicting articles in the top 10% of citation volume. We reduced the initial 14,083 features to 210 features that maintain predictivity. These features corresponded with topics relating to various imaging techniques (eg, diffusion-weighted magnetic resonance imaging, hyperpolarized magnetic resonance imaging, dual-energy computed tomography, computed tomography reconstruction algorithms, tomosynthesis, elastography, and computer-aided diagnosis), particular pathologies (prostate cancer; thyroid nodules; hepatic adenoma, hepatocellular carcinoma, non-alcoholic fatty liver disease), and other topics (radiation dose, electroporation, education, general oncology, gadolinium, statistics). CONCLUSIONS: Machine learning can be successfully applied to create specific feature-based models for predicting articles likely to achieve high influence within the radiological literature.