Predicting continuous ground reaction forces from accelerometers during uphill and downhill running: A recurrent neural network solution

BackgroundGround reaction forces (GRFs) are important for understanding the biomechanics of human movement but the measurement of GRFs is generally limited to a laboratory environment. Wearable devices like accelerometers have been used to measure biomechanical variables outside the laboratory environment, but they cannot directly measure GRFs. Previous studies have used neural networks to predict the entire GRF waveform during the stance phase from wearable device data, but these networks require normalization of GRFs to the duration of a step or stance phase, resulting in a loss of the GRF waveform’s temporal component. Additionally, previous studies have predicted GRF waveforms during level-ground, but not uphill or downhill running. A method of predicting the normal (perpendicular to running surface) GRF waveform using wearable devices across a range of running speeds and slopes, while maintaining the GRF waveform’s temporal component, could allow researchers and clinicians to predict kinetic and kinematic variables outside the laboratory environment.PurposeWe sought to develop a recurrent neural network capable of predicting normal GRF waveforms across a range of running speeds and slopes using data from accelerometers located on the sacrum and shoe.Methods19 subjects completed 30-s running trials on a force-measuring treadmill at five slopes (0°, ±5°, ±10°) and three speeds (2.5, 3.33, and 4.17 m/s) per slope. One biaxial accelerometer was adhered to the sacrum and two uniaxial accelerometers were adhered to the right shoe during all trials. Accelerometers on the shoe were used to classify foot strike patterns as rearfoot, midfoot, or forefoot, and sacral acceleration data were divided into overlapping 12-ms windows, allowing the neural network to iteratively predict the experimentally-measured normal GRF waveform frame-by-frame. The mean, SD, and range of accelerometer data for each 12-ms window were included as neural network input features, along with the subject’s body mass, height, running speed, slope, and percentage of a trial’s steps classified as a rearfoot, midfoot, or forefoot strike. We assessed the accuracy and generalizability of the neural network using leave-one-subject-out cross validation, which provided an ensemble of Root Mean Square Error (RMSE) and relative RMSE (rRMSE) values comparing the normal GRF waveform predicted by the neural network to the normal GRF waveform measured by the force-measuring treadmill. Additionally, we calculated the mean absolute percent error (MAPE) of step frequency, contact time, normal impulse, normal GRF active peak, and loading rate between the predicted and measured GRF waveforms.ResultsThe average ± SD RMSE was 0.16 ± 0.04 BW and rRMSE was 6.4 ± 1.5% for neural network predictions of each subject’s normal GRF waveform compared to measured GRF waveforms across all conditions. RMSE values were lower during slow uphill running (2.5 m/s, +10°; 0.13 ± 0.07 BW) compared to fast downhill running (4.17 m/s, −10°; 0.20 ± 0.05 BW). The MAPE ± SD for step frequency was 0.1 ± 0.1%, contact time was 4.9 ± 4.0%, normal impulse was 6.4 ± 6.9%, normal GRF active peak was 8.5 ± 8.2%, and loading rate was 27.6 ± 36.1%.ConclusionsWe developed a recurrent neural network that uses accelerometer data to predict the continuous normal GRF waveform across a range of running speeds and slopes. The neural network does not require preliminary identification of the stance phase, maintains the temporal component of the GRF waveform, can be applied to up- and downhill running, and facilitates the prediction of kinetic and kinematic variables outside the laboratory environment. This represents a substantial step towards accurately quantifying and monitoring the external loads experienced by the body when running outdoors.

Production and characterization of semi-quinolone antibiotic produced by Streptomyces griseorubens

This work is an attempt to overcome antimicrobial resistance problem which dispersed worldwide inparticular developing countries due to misuse of antibiotics. Actinobacteria were isolated and screenedagainst selected resistant Gram-negative bacteria to detect the powerful antibacterial activity.Identification of the most potent actinobacterial isolate has been carried out using classical and geneticalmethods. Antibacterial compound has been extracted, purified and characterized using accurate and morespecific techniques and instruments. Among forty actinobacterial isolates, only twenty-two isolates couldinhibit the growth of Gram-negative bacteria. The most potent isolate Eg-7 was identified as S.griseorubens, which has a typical 16S rRNA gene. The antibacterial compound was extracted using ethylacetate, and separated by High Performance Liquid Chromatography using methanol and water as amobile phase. Five active peaks were displayed and retained in the range 40 – 45 min, but the last threepeaks were retained at 41.90, 43.43, and 44.54 min, respectively. The crude extract was analyzed byliquid chromatography mass spectrum, where the active peak was displayed at 721.325 m/z. Theantibacterial compound was purified using flash column chromatography and gel filtration columnchromatography. The active fraction was analyzed by Infra-Red spectrum, where a broad absorption at3338 cm-1 was displayed. Molecular formula of an antibacterial compound was determined by massspectrum as C35H26N6O4. Nuclear magnetic resonance analysis was carried out for an antibacterialcompound. These results suggest that a new antibacterial compound that similar quinolone could beproduced by S. griseorubens and exhibited a higher activity against Gram-negative bacteria.

INFLUENCE OF BIOSTIMULANTS ON THE IMMUNE PROTECTION OF CHICKENS

Among the animals of different species chickens react in greater numbers and more noticeable to a va-riety of growth biostimulators. There are great improvements on their general state, growth spurts and development of internal organs is quicker when they are injected with small portions. Such stimulation has a great influence not only on growth and development of chickens in their first period of life but also on health and productivity later on. The most active peak of reaction is when chicken is two months old. Their internal organs, especially the digestive system, develop earlier, their genitals appear earlier and they begin egg-laying much earlier too, when chickens are being fed those biostimulants. Slaughter meat yield becomes more and quality of meat improves with the influence of stimulators. A major disease pre-vention and healing effect can be reached, since most of the biostimulants raise immune system and re-sistance of the organism. It can be the only thing to justify their usage on animals. Tests show that the most typical growth spurt of birds is from fifteen to twenty percent in normal conditions. Growth spurts are also accompanied with the rise of resistance to different infections and activation of different physio-logical processes. Growth spurts can be twice or more than written here, but those spurts are usually short-timed and often accompanied with the dysfunction of different organs

Locomotor Adaptations During RaceRunning in People With Neurological Motor Disorders

The aim of this study was to examine strategies to absorb impact shock during RaceRunning in participants with neurological motor disorders. For this purpose, 8 RaceRunning athletes (4 male and 4 female) voluntarily took part in the study. Each participant performed a series of 100-m sprints with a RaceRunning bike. Acceleration of the tibia and head was measured with 2 inertial measurement units and used to calculate foot-impact shock measures. Results showed that RaceRunning pattern was characterized by a lack of impact peak in foot–ground contact time and the existence of an active peak after foot contact. Due to the ergonomic properties of the RaceRunning bike, shock is attenuated throughout the stance phase. In conclusion, the results revealed that RaceRunning athletes with neurological motor disorders are capable of absorbing impact shock during assisted RaceRunning using a strategy that mimics runners without disabilities.

No immediate effects of highly cushioned shoes on basic running biomechanics

The aim of the study was to investigate the effects of highly cushioned shoes on running biomechanics. Sixteen recreational runners (8 males, 8 females) participated and ran at a self-selected pace across the force platform in the research laboratory wearing either the standard or highly cushioned shoes, in randomized order. Impact peak (IP), loading rate to IP (LR), active peak (AP), contact time (CT), strike index (SI), running velocity, and knee and ankle kinematics at initial contact (IC) and AP were recorded during the running trials. Overall, there was no effect of footwear on IP, LR, AP, CT and velocity (p>.05) with small effect sizes (ES<0.2). The highly cushioned shoes resulted in a more anterior foot strike pattern, based on the slightly higher SI (p=.03, ES=0.5), although the runners demonstrated a rearfoot strike pattern regardless of shoe condition. No kinematic differences were observed at IC or AP, across shoe conditions (p>.05). Our results indicate that the highly cushioned shoes did not show immediate changes in running biomechanics.

2017 Roger A. Mann Award Winner

Category: Midfoot/Forefoot Introduction/Purpose: Hallux rigidus (HR) is a very common symptomatic problem affecting one in 40 patients over the age of 50 years. A variety of treatment options exist and, as is common in surgery, grading systems are used to assess severity of the condition and aid in the guidance of treatment. The most commonly used grading system for HR uses radiographic images, great toe range of motion and clinical symptoms. This study examines the relationship of radiographic and motion findings to observed intra-operative cartilage loss in patients with HR and explores hallux rigidus grade and cartilage loss as predictive variables for treatment outcomes. Methods: A prospective, randomized non-inferiority study examining outcomes of arthrodesis compared to hemiarthroplasty of the first metatarsal phalangeal joint (Cartiva®) was performed.2 All randomized and treated patients were included in this study. Patients underwent pre-operative clinical examination, including measures of joint motion, radiographic assessment and HR grade. Operatively, observations of cartilage loss on the metatarsal head and opposing proximal surfaces were recorded. All patients’ data, irrespective of treatment, were aggregated and Spearman Rank Correlation coefficients used to assess for strength of correlation of active peak dorsiflexion and cartilage loss to HR grade. Outcomes data were then separated by treatment group and two-sided Fisher’s Exact test assessed these variables’ impact on clinical success (p<0.05). Results: In 202 patients, 59 (29%), 110 (55%), and 33 (16%) were classified as Coughlin1 Grades 2, 3, and 4, respectively. There was no correlation between grade and active peak dorsiflexion (-0.02, p=0.78). While rank correlations between grade and cartilage loss on the proximal phalanx and metatarsal head statistically significantly differed from zero, the magnitudes of the correlations were small, 0.176 (p=0.01) and 0.224 (p=0.001), respectively (Table 1). Among Grade 4 patients, 36.4% had no metatarsal cartilage remaining; but this was also found in 8.5% of Grade 2 patients. Similarly, 52.5% of Grade 2 patients had ≥50% metatarsal cartilage remaining; but this was also found in 21.2% of Grade 4 patients. None of the observed factors were significantly associated with likelihood of achieving composite success. Conclusion: This study examines the relationship of motion and intra-operative cartilage loss findings with a commonly used clinical and radiographic grading system for hallux rigidus. This study population included only candidates with HR considered a candidate for arthrodesis based on review of clinical symptoms however the Grade assigned maybe Coughlin Grade 2, 3 or 4. Irrespective of the Grade, positive outcomes were demonstrated within both treatment groups. The weak correlations of preoperative motion and intra-operative cartilage loss to grade suggests that clinical symptoms should be a significant determinant guiding the treatment option rather than radiographic or range of motion factors.

Ramp Angle, Not Plateau Height, Influences Transition Strategies

In a previous study, we found that participants modified how they transitioned onto and off of ramp configurations depending upon the incline. While the transition strategies were originally attributed to ramp angles, it is possible that the plateau influenced the strategies since the final surface height also differed. Ultimately, for the current study, we hypothesized that an individual’s transition strategies would have significant main effects for ramp angle, but not plateau height. Twelve healthy, young adults transitioned onto 3 distinct ramp configurations, a 2.4-m ramp angled at 12.5° ending at a plateau height of 53 cm, a 1.2-m ramp angled at 23.5° ending at a plateau height of 53 cm, and a 2.4-m ramp angled at 23.5° ending at a plateau height of 99.5 cm. Kinematics, kinetics, and muscle activity were measured during the stance phase before contacting the ramp. In support of our hypothesis, impact peak, active peak, and all of the muscle activity variables had a significant main effect for ramp angle, with greater vertical force peaks and muscle activity on steeper ramp transitions. These findings support our previous interpretation that individuals use estimations of ramp angle, not plateau height, to determine their transition strategies.

Altered Vertical Ground Reaction Forces in Participants With Chronic Ankle Instability While Running

Context: Altered gait kinetics may increase the risk of long-term injuries in participants with chronic ankle instability (CAI). Vertical ground reaction forces (vGRFs) can provide insight into how body loading is altered. Objective: To compare the components of vGRFs while running in participants with or without CAI. Design: Cohort study. Setting: University biomechanics laboratory. Patients or Other Participants: Twenty-four experienced, college-aged runners. Groups were categorized by the presence (CAI group) or absence (control group) of CAI through self-reported questionnaires. Intervention(s): After a warm-up period, all participants ran on an instrumented treadmill for 5 minutes at 3.3 m/s. Data were collected during the last 30 seconds. Five continuous trials of heel-to-toe running were identified per participant and averaged for statistical analysis. Main Outcome Measure(s): The dependent variables were impact peak force (N/body weight [BW]), active peak force (N/BW), time to impact peak force (milliseconds), time to active peak force (milliseconds), and average loading rate ([N/BW]/s). Results: A difference was found between groups (P = .002). The CAI group had higher impact peak forces (P = .001) and active peak forces (P = .002) compared with the control group. The CAI group also had an increased loading rate (P = .001) and a shorter time to reach the active peak force (P = .001) compared with the control group. No difference was seen between groups in the time to reach the impact peak force (P = .952). Conclusions: Participants with CAI produced altered vGRFs and loading rates while running. Altered loading rates could predispose individuals with CAI to stress-related injuries and repetitive sprains.