Automated extraction of heart rate variability from magnetoencephalography signals.

Magnetoencephalography (MEG) measures magnetic fluctuations in the brain generated by neural processes, some of which, such as cardiac signals, are generally removed as artifacts and discarded. However, heart rate variability (HRV) has long been regarded as a biomarker related to autonomic function, suggesting the cardiac signal in MEG contains valuable information that can provide supplemental health information about a patient. To enable access to these ancillary HRV data, we created an automated extraction tool capable of capturing HRV directly from raw MEG data with artificial intelligence. Five scans were conducted with simultaneous MEG and electrocardiogram (ECG) acquisition, which provides a ground truth metric for assessing our algorithms and data processing pipeline. In addition to directly comparing R-peaks between the MEG and ECG signals, this work explores the variation of the corresponding HRV output in time, frequency, and non-linear domains. After removing outlier intervals and aligning the ECG and derived cardiac MEG signals, the RMSE between the RR-intervals of each was RMSE1 = 2 ms, RMSE2 = 2 ms, RMSE3 = 8 ms, RMSE4 = 4 ms, RMSE5 = 13 ms. The findings indicate that cardiac artifacts from MEG data carry sufficient signal to approximate an individual's HRV metrics.

Relationship between Aggressiveness, Self-Confidence, and Perceived Coach Support and Head Impact Exposure in Youth Football.

This study evaluated head impact exposure (HIE) metrics in relation to individual-level determinants of HIE. Youth (n = 13) and high school (n = 21) football players were instrumented with the Head Impact Telemetry (HIT) system during one season. Players completed the Trait-Robustness of Self-Confidence Inventory (TROSCI), Sports Climate Questionnaire (SCQ), and Competitive Aggressiveness and Anger Scale (CAAS), measuring self-confidence, perceived coach support, and competitive aggressiveness, respectively. Relationships between HIE metrics (number of impacts, median and 95th percentile accelerations, and risk-weighted exposure (RWE)) and survey scores were evaluated using linear regression analysis. For middle school athletes, TROSCI scores were significantly negatively associated with the number of competition impacts and the mean number of impacts per player per competition. SCQ scores were significantly positively associated with median linear acceleration during practice. CAAS scores were not significantly associated with biomechanical metrics at either level of play. Perceived coach support and self-confidence might influence HIE among middle school football players. Football athletes' competitive aggressiveness may have less influence their HIE than other factors.

Evaluation of head impact exposure measured from youth football game plays.

OBJECTIVE: There is a growing body of literature informing efforts to improve the safety of football; however, research relating on-field activity to head impacts in youth football is limited. Therefore, the objective of this study was to compare head impact exposure (HIE) measured in game plays among 3 youth football teams. METHODS: Head impact and video data were collected from athletes (ages 10-13 years) participating on 3 youth football teams. Video analysis was performed to verify head impacts and assign each to a specific play type. Each play was categorized as a down, punt, kickoff, field goal, or false start. Kickoffs and punts were classified as special teams. Downs were classified as running, passing, or other. HIE was quantified by play type in terms of mean, median, and 95th percentile linear and rotational acceleration. Mixed-effects models were used to assess differences in acceleration among play types. Contact occurring on special teams plays was evaluated using a standardized video abstraction form. RESULTS: A total of 3003 head impacts over 27.5 games were analyzed and paired with detailed video coding of plays. Most head impacts were attributed to running (79.6%), followed by passing (14.0%), and special teams (6.4%) plays. The 95th percentile linear acceleration measured during each play type was 52.6g, 50.7g, and 65.5g, respectively. Special teams had significantly greater mean linear acceleration than running and passing plays (both p = 0.03). The most common kick result on special teams was a deep kick, of which 85% were attempted to be returned. No special teams plays resulted in a touchback, and one resulted in a fair catch. One-third of all special teams plays and 92% of all nonreturned kicks resulted in athletes diving toward the ball. CONCLUSIONS: The results demonstrate a trend toward higher head impact magnitudes on special teams than for running and passing plays, but a greater number of impacts were measured during running plays. Deep kicks were most common on special teams, and many returned and nonreturned kicks resulted in athletes diving toward the ball. These results support policy changes to youth special teams plays, including modifying the yard line the ball is kicked from and coaching proper return technique. Further investigation into biomechanical exposure measured during game impact scenarios is needed to inform policy relevant to the youth level.

Comparison of head impact exposure in practice drills among multiple youth football teams.

Objective: Limiting contact in football practice can reduce the number of head impacts a player receives, but further research is needed to inform the modification of optimal drills that mitigate head impact exposure (HIE) while the player develops the skills needed to safely play the game. This study aimed to compare HIE in practice drills among 6 youth football teams and to evaluate the effect of a team on HIE. Methods: On-field head impact data were collected from athletes (ages 10­13 years) playing on 6 local youth football teams (teams A­F) during all practices using the Head Impact Telemetry System. Video was recorded and analyzed to verify and assign impacts to a specific drill. Drills were identified as follows: dummy/sled tackling, half install, install, install walk through, multiplayer tackle, Oklahoma, one-on-one, open field tackling, other, passing, position skill work, scrimmage, special teams, tackling drill stations, and technique. HIE was quantified in terms of impacts per player per minute (ppm) and peak linear and rotational head acceleration. Generalized linear models were used to assess differences in head impact magnitude and frequency among drills as well as among teams within the most common drills. Results: Among 67 athlete-seasons, a total of 14,718 impacts during contact practices were collected and evaluated in this study. Among all 6 teams, the mean linear (p < 0.0001) and rotational (p < 0.0001) acceleration varied significantly among all drills. Open field tackling had significantly (p < 0.001) higher mean linear acceleration than all other drills. Multiplayer tackle had the highest mean impact rate (0.35 ppm). Significant variations in linear acceleration and impact rate were observed among teams within specific drills. Team A had the highest mean linear acceleration in install, one-on-one, and open field tackling and the highest mean impact rate in Oklahoma and position skill work. Although team A spent the greatest proportion of their practice on minimal- or no-player versus player contact drills (27%) compared to other teams, they had the highest median (20.2g) and 95th percentile (56.4g) linear acceleration in practice. Conclusions: Full-speed tackling and blocking drills resulted in the highest HIE. Reducing time spent on contact drills relative to minimal or no contact drills may not lower overall HIE. Instead, interventions such as reducing the speed of players engaged in contact, correcting tackling technique, and progressing to contact may reduce HIE more effectively.

Machine Learning Applications to Resting-State Functional MR Imaging Analysis.

Machine learning is one of the most exciting and rapidly expanding fields within computer science. Academic and commercial research entities are investing in machine learning methods, especially in personalized medicine via patient-level classification. There is great promise that machine learning methods combined with resting state functional MR imaging will aid in diagnosis of disease and guide potential treatment for conditions thought to be impossible to identify based on imaging alone, such as psychiatric disorders. We discuss machine learning methods and explore recent advances.