A BS TRACT: Background: Quantitative assessment of severity of ataxia-specific gait impairments from wearable technology could provide sensitive performance outcome measures with high face validity to power clinical trials. Objectives: The aim of this study was to identify a set of gait measures from body-worn inertial sensors that best discriminate between people with prodromal or manifest spinocerebellar ataxia (SCA) and age-matched, healthy control subjects (HC) and determine how these measures relate to disease severity. Methods: One hundred and sixty-three people with SCA (subtypes 1, 2, 3, and 6), 42 people with prodromal SCA, and 96 HC wore 6 inertial sensors while performing a natural pace, 2-minute walk. Areas under the receiver operating characteristic curves (AUC) were compared for 25 gait measures, including standard deviations as variability, to discriminate between ataxic and normal gait. Pearson's correlation coefficient assessed the relationships between the gait measures and severity of ataxia.Results: Increased gait variability was the most discriminative gait feature of SCA; toe-out angle variability (AUC = 0.936; sensitivity = 0.871; specificity = 0.896) and double-support time variability (AUC = 0.932; sensitivity = 0.834; specificity = 0.865) were the most sensitive and specific measures. These variability measures were also significantly correlated with the scale for the assessment and rating of ataxia (SARA) and disease duration. The same gait measures discriminated gait of people with prodromal SCA from the gait of HC (AUC = 0.610, and 0.670, respectively). Conclusions: Wearable inertial sensors provide sensitive and specific measures of excessive gait variability in both manifest and prodromal SCAs that are reliable and related to the severity of the disease, suggesting they may be useful as clinical trial performance outcome measures.

No abstract

Reporting of sports-related concussions (SRCs) has risen dramatically over the last decade, increasing awareness of the need for treatment and prevention of SRCs. To date most prevention studies have focused on equipment and rule changes to sports in order to reduce the risk of injury. However, increased neck strength has been shown to be a predictor of concussion rate. In the TRAIN study, student-athletes will follow a simple neck strengthening program over the course of three years in order to better understand the relationship between neck strength and SRCs. Neck strength of all subjects will be measured at baseline and biannually over the course of the study using a novel protocol. Concussion severity and duration in any subject who incurs an SRC will be evaluated using the Sports Concussion Assessment Tool 5th edition, a questionnaire based tool utilizing several tests that are commonly affected by concussion, and an automated eye tracking algorithm. Neck strength, and improvement of neck strength, will be compared between concussed and non-concussed athletes to determine if neck strength can indeed reduce risk of concussion. Neck strength will also be analyzed taking into account concussion severity and duration to find if a strengthening program can provide a protective factor to athletes. The study population will consist of student-athletes, ages 12–23, from local high schools and colleges. These athletes are involved in a range of both contact and non-contact sports.

Background The manual extraction of valuable data from electronic medical records is cumbersome, error-prone, and inconsistent. By automating extraction in conjunction with standardized terminology, the quality and consistency of data utilized for research and clinical purposes would be substantially improved. Here, we set out to develop and validate a framework to extract pertinent clinical conditions for traumatic brain injury (TBI) from computed tomography (CT) reports. Methods We developed tbiExtractor, which extends pyConTextNLP, a regular expression algorithm using negation detection and contextual features, to create a framework for extracting TBI common data elements from radiology reports. The algorithm inputs radiology reports and outputs a structured summary containing 27 clinical findings with their respective annotations. Development and validation of the algorithm was completed using two physician annotators as the gold standard. Results tbiExtractor displayed high sensitivity (0.92-0.94) and specificity (0.99) when compared to the gold standard. The algorithm also demonstrated a high equivalence (94.6%) with the annotators. A majority of clinical findings (85%) had minimal errors (F1 Score � 0.80). When compared to annotators, tbiExtractor extracted information in significantly less time (0.3 sec vs 1.7 min per report).