Gait disorder is a particularly disabling and treatment refractory symptom of Parkinson's disease (PD), contributing to higher fall risk [[1]Pickering R.M. Grimbergen Y.A.M. Rigney U. Ashburn A. Mazibrada G. Wood B. et al.A meta-analysis of six prospective studies of falling in Parkinson's disease.Mov Disord. 2007; 22: 1892-1900https://doi.org/10.1002/mds.21598Crossref PubMed Scopus (401) Google Scholar] and restrictions of daily activities [[2]Bloem B.R. Grimbergen Y.A.M. Cramer M. Willemsen M. Zwinderman A.H. Prospective assessment of falls in Parkinson's disease.J Neurol. 2001; 248: 950-958https://doi.org/10.1007/s004150170047Crossref PubMed Scopus (598) Google Scholar]. Although Parkinsonian gait can improve with deep brain stimulation (DBS) of the subthalamic nucleus (STN) or globus pallidus internus (GPi) [[3]Piper M. Abrams G.M. Marks W.J.J. Deep brain stimulation for the treatment of Parkinson's disease: overview and impact on gait and mobility.NeuroRehabilitation. 2005; 20: 223-232https://doi.org/10.3233/NRE-2005-20308Crossref PubMed Google Scholar], gait is not considered to be adequately treated [[4]Fasano A. Aquino C.C. Krauss J.K. Honey C.R. Bloem B.R. Axial disability and deep brain stimulation in patients with Parkinson disease.Nat Rev Neurol. 2015; 11: 98-110https://doi.org/10.1038/nrneurol.2014.252Crossref PubMed Scopus (161) Google Scholar]. Two "closed-loop" DBS alternatives have been proposed: adaptive DBS (aDBS), where stimulation varies with local field potential (LFP) power (e.g. beta-band 13–30 Hz) [[5]Little S. Beudel M. Zrinzo L. Foltynie T. Limousin P. Hariz M. et al.Bilateral adaptive deep brain stimulation is effective in Parkinson's disease.J Neurol Neurosurg Psychiatry. 2016; 87: 717-721https://doi.org/10.1136/jnnp-2015-310972Crossref PubMed Scopus (187) Google Scholar], and responsive DBS (rDBS), where stimulation is entrained to a specific phase of the pathological movement (e.g. tremor) [[6]Cagnan H. Denison T. Pedrosa D. Little S. Pogosyan A. Cheeran B. et al.Stimulating at the right time: phase-specific deep brain stimulation Stimulating at the right time: phase-specific deep brain stimulation.Brain. 2016; : 132-145https://doi.org/10.1093/brain/aww286Crossref PubMed Scopus (133) Google Scholar]. Two studies have evaluated aDBS effect on gait [[7]Rosa M. Arlotti M. Ardolino G. Cogiamanian F. Marceglia S. Di Fonzo A. et al.Adaptive deep brain stimulation in a freely moving parkinsonian patient.Mov Disord. 2015; 30: 1003-1005https://doi.org/10.1002/mds.26241Crossref PubMed Scopus (144) Google Scholar,[8]Petrucci M.N. Neuville R.S. Afzal M.F. Velisar A. Anidi C.M. Anderson R.W. et al.Neural closed-loop deep brain stimulation for freezing of gait.Brain Stimul. 2020; 13: 1320-1322https://doi.org/10.1016/j.brs.2020.06.018Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar], and none has evaluated rDBS effect on gait. Beta-band power is modulated with the gait cycle [[9]Fischer P. Chen C.C. Chang Y.-J.J. Yeh C.-H.H. Pogosyan A. Herz D.M. et al.Alternating modulation of subthalamic nucleus beta oscillations during stepping.J Neurosci. 2018; 38: 5111-5121https://doi.org/10.1523/JNEUROSCI.3596-17.2018Crossref PubMed Scopus (46) Google Scholar]. We hypothesize that rDBS timed to gait events might allow, or even enhance, this normal, physiological beta modulation, improving gait. We developed and tested a rDBS system delivering short duration pulse trains at specific gait phases in real-time. To assess the accuracy of stimulation delivery, gait phases were aligned with stimulation artifacts collected from a surface EMG electrode on the neck. To measure efficacy of this rDBS system, we assessed spatial and temporal gait metrics. Sixteen PD individuals with bilateral DBS leads (13 STN, 3 GPi) and Medtronic SC, PC, or RC implantable neural stimulators (INSs) were enrolled. All gave informed consent according to a University of Minnesota Institutional Review Board approved protocol. Data from four participants (all STN) were excluded due to technical difficulties (stimulation not delivered at the target gait phase). Detailed participant demographics are in Table S1. Participants walked on an instrumented treadmill (C-Mill, Motek Medical, Netherlands) for one-minute trials. Each trial was under one of five conditions: off-stimulation, continuous stimulation, stimulation triggered on ipsilateral heel-strike (IHS), on contralateral heel-strike (CHS), or on contralateral toe-off (CTO). These were block-randomized in four blocks of five trials each. DBS frequency, contacts, etc. were otherwise as usual, for each participant, except as noted in Table S1. Participants were tested in the overnight off-medication state. Gait events were detected in real-time using force sensitive resistors (FSRs, DC:F01 and Trigno 4-Channel FSR Adapter, Delsys Inc, Natick, MA) and accelerometers (Trigno, Delsys Inc, Natick, MA). A surface EMG electrode on the participant's neck sensed stimulation artifacts. FSRs were placed between the first and second metatarsal and on the heel to capture changes in ground reaction force, which is converted to a voltage. Accelerometers were placed on the shank over the tibialis anterior. Threshold for each FSR was set to 40% of the maximum voltage measured during a short walk on the treadmill. Heel-strike events were detected by the rapid rising segments of the heel FSR voltage and negative acceleration measured on the shank. Toe-off events were detected by the rapid falling segments of metatarsal FSR voltage (Fig. 1A–C). To trigger short duration pulse trains following detection of a heel-strike or toe-off gait phases, the Nexus-D3 interface (Medtronic Inc., Minneapolis, MN) was used to send a stimulation request to the participant's INS after a calculated controlled delay accounting for hardware, software, and communication system delays. The controlled delay was targeted to stimulate at the proper gait phase on the following gait cycle (see details in supplementary). Pulse train durations were between 125 and 135 ms. Accuracy of stimulation timing was measured relative to target gait phase independently detected by the instrumented treadmill (see details in supplementary). Spatial and temporal gait metrics were used to evaluate the stimulation effect. Metrics were calculated using custom Python and MATLAB script from heel-strike and toe-off gait phases detected by the treadmill from center of pressure trajectories. As our sample had only 3 GPi participants, we present all participants' data graphically but limit statistical analysis to STN only. Additionally, participant 3 (shown graphically in orange) was excluded from statistical analysis due to DBS electrodes placed outside the target structure. Statistical analysis of all gait metrics were performed using a repeated measures random intercept mixed model with stimulation condition and treadmill speed as fixed factors and participant as random factor (details in supplement). The rDBS system delivered stimulation accurately at targeted phases (Fig. 1D). Stimulation timings, expressed as gait cycle percentage of the left and right brain (mean ± SD) were: 4.05 ± 9.89% and 52.01 ± 12.56% (IHS), −0.38 ± 11.03% and 52.51 ± 11.43% (CHS), and 16.09 ± 10.37% and 65.93 ± 11.66% (CTO). These correspond with published values for heel-strike and toe-off timing [[10]Kirtley C. Clinical gait analysis: Theory and Practice. 1st Ed. Churchill Livingstone.2006: 219https://doi.org/10.1016/B978-0-443-10009-3.50015-1Crossref Google Scholar]. The effect of stimulation condition was significant for stride length (F(4, 145) = 11.01; P < 0.01) and stride time (F(4,144.9) = 8.45; P < 0.01). Post-hoc analysis revealed this was driven by the continuous stimulation condition. Approximately 4% improvement was observed in all gait metrics when comparing continuous stimulation to off-stimulation (Fig. 1E and F). All phasic stimulation conditions' effects on all gait metrics were negligible and non-significant. Three participants (2 STN and 1 GPi), showed a decreased stride length (−0.5-24.7%) and time (−0.2-24.6%), i.e. worsening of gait, while receiving continuous stimulation compared to off-stimulation (Participants 3, 9, and 10; Fig. 1 E, F). Interestingly, these participants also had improved gait with phasic stimulation, i.e. increased stride length (0.6–4.1%) and time (0.7–4.0%). In conclusion, we demonstrate the feasibility of responsive gait phase triggered DBS and its effect on gait in people with PD during treadmill walking. This is the first study to explore this type of stimulation for gait in Parkinson's disease. rDBS-induced changes in gait were not significant for any phase of stimulation. In contrast, continuous stimulation's effect, was robust and significant, which allows us to reject the hypothesis that phasic stimulation's effect is greater or equal to conventional continuous stimulation. However, we cannot exclude some smaller therapeutic effect; sample size estimates for detecting such a hypothetical effect are in the supplemental material. Our results suggest that rDBS may be more effective in patients for whom continuous stimulation is ineffective, e.g. when lead location is suboptimal by conventional criteria. However, more study would be required to substantiate this observation. The authors report no competing interests. We thank all the patients who participated in the study. We thank Medtronic for providing the Nexus-D interface system. We also thank Benjamin Isaacson and his team at Medtronic for their technical assistance. The authors have no conflict of interests or financial disclosures. The following is the Supplementary data to this article: Download .docx (.55 MB) Help with docx files Multimedia component 1 This work was supported by Medtronic (ERP NM-3511), the University of Minnesota Neuromodulation Innovations (MnDrive), NSF IGERT grant DGE-1069104, NIH T32-MH115886, and NIH Udall grant P50 NS098573. Kenneth H. Louie: Conceptualization, Data curation, Formal analysis, Methodology, Software, Writing – original draft, Writing – review & editing. Chiahao Lu: Conceptualization, Data curation, Methodology, Writing – review & editing. Tessneem Abdallah: Data curation, Writing – review & editing. Jacob C. Guzior: Data curation, Writing – review & editing. Emily Twedell: Data curation, Writing – review & editing. Theoden I. Netoff: Supervision, Writing – review & editing. Scott E. Cooper: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Supervision, Writing – review & editing.

Objective 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. Materials and 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). Discussion and Conclusion tbiExtractor is a validated algorithm for extraction of TBI common data elements from radiology reports. This automation reduces the time spent to extract structured data and improves the consistency of data extracted. Lastly, tbiExtractor can be used to stratify subjects into groups based on visible damage by partitioning the annotations of the pertinent clinical conditions on a radiology report.

OBJECTIVE: Traumatic brain injuries (TBIs) are largely underdiagnosed and may have persistent refractory consequences. Current assessments for acute TBI are limited to physical examination and imaging. Biomarkers such as glial fibrillary acidic protein (GFAP), ubiquitin C-terminal hydrolase L1 (UCH-L1), and S100 calcium-binding protein B (S100B) have shown predictive value as indicators of TBI and potential screening tools. METHODS: In total, 37 controls and 118 unique trauma subjects who received a clinically ordered head computed tomography (CT) in the emergency department of a level 1 trauma center were evaluated. Blood samples collected at 0-8 hours (initial) and 12-32 hours (delayed) postinjury were analyzed for GFAP, UCH-L1, and S100B concentrations. These were then compared in CT-negative and CT-positive subjects. RESULTS: Median GFAP, UCH-L1, and S100B concentrations were greater in CT-positive subjects at both timepoints compared with CT-negative subjects. In addition, median UCH-L1 and S100B concentrations were lower at the delayed timepoint, whereas median GFAP concentrations were increased. As predictors of a positive CT of the head, GFAP outperformed UCH-L1 and S100B at both timepoints (initial: 0.89 sensitivity, 0.62 specificity; delayed: 0.94 sensitivity, 0.67 specificity). GFAP alone also outperformed all possible combinations of biomarkers. CONCLUSIONS: GFAP, UCH-L1, and S100B demonstrated utility for rapid prediction of a CT-positive TBI within 0-8 hours of injury. GFAP exhibited the greatest predictive power at 12-32 hours. Furthermore, these results suggest that GFAP alone has greater utility for predicting a positive CT of the head than UCH-L1, S100B, or any combination of the 3.

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.