Deep learning has been used successfully in supervised classification tasks in order to learn complex patterns. The purpose of the study is to apply this machine learning technique to classifying images of brains with different types of tumors: meningioma, glioma, and pituitary. The image dataset contains 233 patients with a total of 3064 brain images with either meningioma, glioma, or pituitary tumors. The images are T1weighted contrast enhanced MRI (CE-MRI) images of either axial (transverse plane), coronal (frontal plane), or sagittal (lateral plane) planes. This research focuses on the axial images, and expands upon this dataset with the addition of axial images of brains without tumors in order to increase the number of images provided to the neural network. Training neural networks over this data has proven to be accurate in its classifications an average five-fold cross validation of 91.43%.

Study Objectives Polysomnography (PSG) scoring is labor intensive and suffers from variability in inter- and intra-rater reliability. Automated PSG scoring has the potential to reduce the human labor costs and the variability inherent to this task. Deep learning is a form of machine learning that uses neural networks to recognize data patterns by inspecting many examples rather than by following explicit programming. Methods A sleep staging classifier trained using deep learning methods scored PSG data from the Sleep Heart Health Study (SHHS). The training set was composed of 42 560 hours of PSG data from 5213 patients. To capture higher-order data, spectrograms were generated from electroencephalography, electrooculography, and electromyography data and then passed to the neural network. A holdout set of 580 PSGs not included in the training set was used to assess model accuracy and discrimination via weighted F1-score, per-stage accuracy, and Cohen’s kappa (K). Results The optimal neural network model was composed of spectrograms in the input layer feeding into convolutional neural network layers and a long short-term memory layer to achieve a weighted F1-score of 0.87 and K = 0.82. Conclusions The deep learning sleep stage classifier demonstrates excellent accuracy and agreement with expert sleep stage scoring, outperforming human agreement on sleep staging. It achieves comparable or better F1-scores, accuracy, and Cohen’s kappa compared to literature for automated sleep stage scoring of PSG epochs. Accurate automated scoring of other PSG events may eventually allow for fully automated PSG scoring.

BackgroundBiomedical research has traditionally been conducted via surveys and the analysis of medical records. However, these resources are limited in their content, such that non-traditional domains (eg, online forums and social media) have an opportunity to supplement the view of an individual’s health.ObjectiveThe objective of this study was to develop a scalable framework to detect personal health status mentions on Twitter and assess the extent to which such information is disclosed.MethodsWe collected more than 250 million tweets via the Twitter streaming API over a 2-month period in 2014. The corpus was filtered down to approximately 250,000 tweets, stratified across 34 high-impact health issues, based on guidance from the Medical Expenditure Panel Survey. We created a labeled corpus of several thousand tweets via a survey, administered over Amazon Mechanical Turk, that documents when terms correspond to mentions of personal health issues or an alternative (eg, a metaphor). We engineered a scalable classifier for personal health mentions via feature selection and assessed its potential over the health issues. We further investigated the utility of the tweets by determining the extent to which Twitter users disclose personal health status.ResultsOur investigation yielded several notable findings. First, we find that tweets from a small subset of the health issues can train a scalable classifier to detect health mentions. Specifically, training on 2000 tweets from four health issues (cancer, depression, hypertension, and leukemia) yielded a classifier with precision of 0.77 on all 34 health issues. Second, Twitter users disclosed personal health status for all health issues. Notably, personal health status was disclosed over 50% of the time for 11 out of 34 (33%) investigated health issues. Third, the disclosure rate was dependent on the health issue in a statistically significant manner (P<.001). For instance, more than 80% of the tweets about migraines (83/100) and allergies (85/100) communicated personal health status, while only around 10% of the tweets about obesity (13/100) and heart attack (12/100) did so. Fourth, the likelihood that people disclose their own versus other people’s health status was dependent on health issue in a statistically significant manner as well (P<.001). For example, 69% (69/100) of the insomnia tweets disclosed the author’s status, while only 1% (1/100) disclosed another person’s status. By contrast, 1% (1/100) of the Down syndrome tweets disclosed the author’s status, while 21% (21/100) disclosed another person’s status.ConclusionsIt is possible to automatically detect personal health status mentions on Twitter in a scalable manner. These mentions correspond to the health issues of the Twitter users themselves, but also other individuals. Though this study did not investigate the veracity of such statements, we anticipate such information may be useful in supplementing traditional health-related sources for research purposes.

Objective Secure messaging through patient portals is an increasingly popular way that consumers interact with healthcare providers. The increasing burden of secure messaging can affect clinic staffing and workflows. Manual management of portal messages is costly and time consuming. Automated classification of portal messages could potentially expedite message triage and delivery of care. Materials and Methods We developed automated patient portal message classifiers with rule-based and machine learning techniques using bag of words and natural language processing (NLP) approaches. To evaluate classifier performance, we used a gold standard of 3253 portal messages manually categorized using a taxonomy of communication types (i.e., main categories of informational, medical, logistical, social, and other communications, and subcategories including prescriptions, appointments, problems, tests, follow-up, contact information, and acknowledgement). We evaluated our classifiers’ accuracies in identifying individual communication types within portal messages with area under the receiver-operator curve (AUC). Portal messages often contain more than one type of communication. To predict all communication types within single messages, we used the Jaccard Index. We extracted the variables of importance for the random forest classifiers. Results The best performing approaches to classification for the major communication types were: logistic regression for medical communications (AUC: 0.899); basic (rule-based) for informational communications (AUC: 0.842); and random forests for social communications and logistical communications (AUCs: 0.875 and 0.925, respectively). The best performing classification approach of classifiers for individual communication subtypes was random forests for Logistical-Contact Information (AUC: 0.963). The Jaccard Indices by approach were: basic classifier, Jaccard Index: 0.674; Naïve Bayes, Jaccard Index: 0.799; random forests, Jaccard Index: 0.859; and logistic regression, Jaccard Index: 0.861. For medical communications, the most predictive variables were NLP concepts (e.g. Temporal_Concept, which maps to ‘morning’, ‘evening’ and Idea_or_Concept which maps to ‘appointment’ and ‘refill’). For logistical communications, the most predictive variables contained similar numbers of NLP variables and words (e.g. Telephone mapping to ‘phone’, ‘insurance’). For social and informational communications, the most predictive variables were words (e.g. social: ‘thanks’, ‘much’, informational: ‘question’, ‘mean’). Conclusions This study applies automated classification methods to the content of patient portal messages and evaluates the application of NLP techniques on consumer communications in patient portal messages. We demonstrated that random forest and logistic regression approaches accurately classified the content of portal messages, although the best approach to classification varied by communication type. Words were the most predictive variables for classification of most communication types...