Employing body-fixed sensors and machine learning to predict physical activity in military personnel.

INTRODUCTION: This was a feasibility pilot study aiming to develop and validate an activity recognition system based on a custom-made body-fixed sensor and driven by an algorithm for recognising basic kinetic movements in military personnel. The findings of this study are deemed essential in informing our development process and contributing to our ultimate aim which is to develop a low-cost and easy-to-use body-fixed sensor for military applications. METHODS: Fifty military participants performed a series of trials involving walking, running and jumping under laboratory conditions in order to determine the optimal, among five machine learning (ML), classifiers. Thereafter, the accuracy of the classifier was tested towards the prediction of these movements (15 183 measurements) and in relation to participants' gender and fitness level. RESULTS: Random forest classifier showed the highest training and validation accuracy (98.5% and 92.9%, respectively) and classified participants with differences in type of activity, gender and fitness level with an accuracy level of 83.6%, 70.0% and 62.2%, respectively. CONCLUSIONS: The study showed that accurate prediction of various dynamic activities can be achieved with high sensitivity using a low-cost easy-to-use sensor and a specific ML model. While this technique is in a development stage, our findings demonstrate that our body-fixed sensor prototype alongside a fully trained validated algorithm can strategically support military operations and offer valuable information to commanders controlling operations remotely. Further stages of our developments include the validation of our refined technique on a larger range of military activities and groups by combining activity data with physiological variables to predict phenomena relating to the onset of fatigue and performance decline.

Differences in energy and nutrient-intake among Greek children between 1992/93 and 2006/07.

BACKGROUND: There is lack of data in Greece on the trends in energy and nutrient intake. The present study aimed to examine differences in these parameters among first-grade children of Crete, Greece, between 1992/93 and 2006/07. METHODS: Children (aged 5.7-7.6 years) from two-representative cross-sectional samples participated during 1992/93 (n = 245) and 2006/07 (n = 257). Nutrient and food intakes were assessed using estimated 3-day-food records. Nutrient intakes were estimated as amounts per 4185 kJ (1000 kcal) of energy intake (EI), whereas macronutrients were estimated as percentage of total EI. RESULTS: Between 1992/93 and 2006/07, respectively, significant increases (P < 0.05) were observed in the intake of energy (boys: 7996 versus 8646 kJ; girls: 7265 versus 7963 kJ), calcium (boys: 488 versus 587 mg; girls: 464 versus 594 mg), iron (boys: 5.8 versus 7.3 mg; girls: 5.7 versus 7.4 mg) and vitamins B2 (boys: 1.01 versus 1.17 mg; girls: 1.0 versus 1.16 mg) and B6 (boys: 0.80 versus 0.87 mg; girls: 0.75 versus 0.95 mg), for boys and girls, respectively. Among boys, the intake of protein (13.6 versus 14.6% of EI) and total trans fatty acids (TTFA) (0.84 versus 0.99 g) was also significantly lower in 1992/93 (P < 0.016). Between 1992/93 and 2006/07, total fat (boys: 42.9% versus 42.8%, girls: 42% versus 43.1%) and saturated fat (boys: 15.5% versus 15.4%, girls: 15% versus 15.3%) exceeded the 30-35% and <10% dietary recommendations, respectively, A significantly higher proportion of children in 2006/07, compared to 1992/93, had a higher intake of total energy (>120% of the recommended energy allowance). CONCLUSIONS: The intake of several macronutrients (protein, TTFA, total and saturated fat) in this sample of children did not meet the recommended intakes in either time period.

The use of various strength-power tests as predictors of sprint running performance.

AIM: The aim of the present study was to assess the relationship between various strength-power tests and maximal running velocity parameters. METHODS: Nine trained males were tested on four separate occasions. On the first occasion the maximum running velocity (MRV), stride rate (SR) and stride length (SL) were measured over 35 m. On the second occasion maximal vertical jumps (squat jump [SJ], standing broad jump [SBJ], counter movement jump [CMJ] and drop jumps [DJ] from heights of 30, 50 and 80 cm) were performed on a force platform. On the third occasion the maximal bilateral isometric force (MBIF) of leg extensors and the force time characteristics (f-t 10-30%, f-t 10-60% and f-t 10-90%) were determined using a leg extension machine connected to a force plate. On the final fourth occasion peak anaerobic power was measured via repeated 6 sec maximum cycle sprints. Pearson product-moment correlation coefficients were calculated for all the aformentioned parameters. RESULTS: The correlation coefficients showed that MRV correlated significantly with f-t 10-60% and DJ 30 (r=-0.73 and r=0.73, P<0.05 respectively). In addition, SR and SL showed significant relationships with f-t 10-60% (r=-0.82, P<0.01 and r=0.75, P<0.05 respectively). CONCLUSION: The present findings suggest that the ability to produce force quickly, as measured by the time to achieve 60% of maximum voluntary contraction is related to sprinting performance, with the coefficient of determination accounting for 53% of the variance in the data. These data also show that sprinting ability is linked with DJ performance, especially the drop jump from a height of 30 cm. It is suggested that the above tests may prove useful in preparing and testing the sprinting ability and sprint specific strength levels.

Creatine supplementation: effects on urinary excretion and anaerobic performance.

AIM: The aim of the present study is to investigate Urinary creatine (URCR) and urinary creatinine (URCRN) response to CR supplementation in conjunction with exercise performance. METHODS: Twenty-one sprint trained males were randomly divided into 3 groups. Each group followed a different CR dosage (10 g, 25 g and 35 g x day(-1) for 4 days) and placebo (Pl) in the 1(st) and 2(nd) week, respectively. A double-blind design was used. Subjects' urine was collected every 24 hours during the entire period of supplementation (SP). All groups, at the end of each SP performed 3 times the Anaerobic Wingate Test (AWT) with 6 min active recovery (60 rpm) on a cycle ergometer. RESULTS: Significantly (p<0.01) higher peak and mean power values were produced during the CR compared to Pl condition. A significant correlation (r=0.7, p<0.05) was also observed between peak power improvement (PPI) and URCR for the 3 groups. No such relationship was found between URCRN. Across all groups, URCR and URCRN increased significantly following ingestion compared with Pl (p<0.001). URCR post- supplementation presented a 7.4 fold, 36 fold and 21 fold increase for 10 g, 25 g and 35 g dose respectively, whilst URCRN presented a mean 2.4 fold increase for all different doses, which clearly shows the magnitude of sensitivity of these indices to CR supplementation. A strong correlation (r=0.95, p<0.01) observed between dose of CR ingestion and mean URCR (MRUCR) with prediction formula: CR = -0.936 + (5.613 x MRUCR) (SEE=3.5). CONCLUSION: URCR was an effective measure of each CR dosage administered as well as of the excretion pattern that each group followed throughout the SP. Furthermore the strong relationship of URCR and PPI could be particularly useful for monitoring and optimising CR loading in athletic populations.

The effect of longer-term creatine supplementation on elite swimming performance after an acute creatine loading.

We investigated the effect of an acute creatine loading (25 g per day for 4 days) and longer-term creatine supplementation (5 g of creatine or 5 g of placebo per day for 2 months) on the performance of 22 elite swimmers during maximal interval sessions. After the acute creatine loading, the mean of the average interval swim times for all swimmers (n = 22) improved (44.3+/-16.5 s before vs. 43.7+/-16.3 s after supplementation; P<0.01). Three of the 22 swimmers did not respond positively to supplementation. After 2 months of longer-term creatine supplementation or placebo, neither group showed a significant change in swimming performance (38.7+/-13.5 s before vs. 38.7+/-14.1 s after for the creatine group; 48.7+/-18.0 s before vs. 48.7+/-18.1 s after for the placebo group). We conclude that, in elite swimmers, 4 days of acute creatine loading improves swimming performance significantly when assessed by maximal interval sessions. However, longer-term supplementation for 2 months (5 g of creatine per day) did not benefit significantly the creatine group compared with the placebo group.