Artificial intelligence in sports biomechanics: New dawn or false hope?

Reference

R Bartlett. Journal of Sports Science and Medicine. Volume 5, 2006. Pages 474-479.

Source

http://www.jssm.org/

(NOTE: Due to copyright restraints, siliconcoach cannot give away this resource. You can search for the article from the link above, however, a fee or subscription may be charged by the supplier.)

Abstract

This article reviews developments in the use of Artificial Intelligence (AI) in sports biomechanics over the last decade. It outlines possible uses of Expert Systems as diagnostic tools for evaluating faults in sports movements (‘techniques’) and presents some example knowledge rules for such an expert system. It then compares the analysis of sports techniques, in which Expert Systems have found little place to date, with gait analysis, in which they are routinely used. Consideration is then given to the use of Artificial Neural Networks (ANNs) in sports biomechanics, focusing on Kohonen self-organizing maps, which have been the most widely used in technique analysis, and multi-layer networks, which have been far more widely used in biomechanics in general. Examples of the use of ANNs in sports biomechanics are presented for javelin and discus throwing, shot putting and football kicking. I also present an example of the use of Evolutionary Computation in movement optimization in the soccer throw in, which predicted an optimal technique close to that in the coaching literature. After briefly over-viewing the use of AI in both sports science and biomechanics in general, the article concludes with some speculations about future uses of AI in sports biomechanics.

Use of siliconcoach

Professor Bartlett talks about some of the possibilities in movement analysis and mentions that siliconcoach’s programmable ‘Wizards’ are the closest thing he has seen to an Expert System.

Assessing dynamic knee joint range of motion using siliconcoach

Reference

J Cronin, M Nash, C Whatman. Physical Therapy in Sport. Volume 7, Issue 4, November 2006, Pages 191-194.

Source

http://www.sciencedirect.com/

(NOTE: Due to copyright restraints, siliconcoach cannot give away this resource. You can search for the article from the link above, however, a fee or subscription may be charged by the supplier.)

Abstract

Objective: Compared to measuring static range of motion (ROM) the assessment of dynamic ROM has received very little research attention. The purpose of this study therefore was to determine the reliability of the siliconCOACH motion analysis system for assessing dynamic ROM of the knee joint.

Design: Test-retest reliability.

Setting: Laboratory.

Participants: Ten male subjects unable to fully extend their knee at 90° of hip flexion.

Main Outcome Measures: Static and dynamic ROM over four separate occasions using a video camera and siliconCOACH digitized footage.

Results: The variation between days for both static and dynamic measurements was minimal (CV<2.1%). With regards to test-retest reliability, the ICC values, were high (ICC ≥ 0.89) for both assessment techniques and the static and dynamic ROM measurements did not differ significantly (p<0.05) on any given testing occasion.

Conclusions: The high ICC and low CVs indicate a high degree of stability between testing days for the procedures used in this study to assess dynamic ROM. Software programmes such as siliconCOACH seem ideal for determining the end range of a movement for both static and dynamic ROM and would seem to offer a functional and cost effective assessment strategy for those practitioners and clinicians interested in the effects of various interventions on ROM.

Use of siliconcoach

These researchers were interested in the ability of siliconcoach software to reliably measure static and dynamic range of motion. The concluded that siliconcoach offers a reliable and functional (face validity) assessment strategy for those practitioners and clinicians interested in the effects of various interventions on dynamic range of motion

Cadence selection affects metabolic responses during cycling and subsequent running time to fatigue.

Reference

F Vercruyssen, R Suriano, D Bishop, C Hausswirth, J Brisswalter. British Journal of Sports Medicine. Volume 39, 2005, pages 267-272.

Source

http://bjsm.bmj.com

(NOTE: Due to copyright restraints, siliconcoach cannot give away this resource. You can search for the article from the link above, however, a fee or subscription may be charged by the supplier.)

Abstract

Objectives: To investigate the effect of cadence selection during the final minutes of cycling on metabolic responses, stride pattern, and subsequent running time to fatigue.

Methods: Eight triathletes performed, in a laboratory setting, two incremental tests (running and cycling) to determine peak oxygen uptake (VO₂PEAK) and the lactate threshold (LT), and three cycle-run combinations. During the cycle-run sessions, subjects completed a 30 minute cycling bout (90% of LT) at (a) the freely chosen cadence (FCC, 94 (5) rpm), (b) the FCC during the first 20 minutes and FCC-20% during the last 10 minutes (FCC-20%, 74 (3) rpm), or (c) the FCC during the first 20 minutes and FCC+20% during the last 10 minutes (FCC+20%, 109 (5) rpm). After each cycling bout, running time to fatigue (T_max) was determined at 85% of maximal velocity.

Results: A significant increase in T_max was found after FCC-20% (894 (199) seconds) compared with FCC and FCC+20% (651 (212) and 624 (214) seconds respectively). VO₂, ventilation, heart rate, and blood lactate concentrations were significantly reduced after 30 minutes of cycling at FCC-20% compared with FCC+20%. A significant increase in VO₂ was reported between the 3rd and 10th minute of all T_max sessions, without any significant differences between sessions. Stride pattern and metabolic variables were not significantly different between T_max sessions.

Conclusions: The increase in T_max after FCC-20% may be associated with the lower metabolic load during the final minutes of cycling compared with the other sessions. However, the lack of significant differences in metabolic responses and stride pattern between the run sessions suggests that other mechanisms, such as changes in muscular activity, probably contribute to the effects of cadence variation on T_max.

Use of siliconcoach

These researchers used the timer and frame by frame capabilities of siliconcoach software to determine the kinematic variables of running under the different conditions

Characteristics of anterior cruciate ligament injuries in Australian Football.

Reference

JL Cochrane, DG Lloyd, A Buttfield, H Seward. Journal of Science and Medicine in Sport. Volume 10, Issue 2, April 2007, Pages 96-104.

Source

http://www.sciencedirect.com/

(NOTE: Due to copyright restraints, siliconcoach cannot give away this resource. You can search for the article from the link above, however, a fee or subscription may be charged by the supplier.)

Abstract

Anterior cruciate ligament (ACL) injuries are the most costly injuries in football at both professional and amateur levels (Orchard J, Seward H, McGivern J, Hood S. Intrinsic and extrinsic risk factors for anterior cruciate ligament injury in Australian footballers. Am J Sports Med 2001;29:196-200.). In this study video analysis of 34 ACL injuries in Australian football was performed to investigate the causes of these injuries. Factors that may have contributed to the cause of the injury were analysed, rated and reported. The factors analysed were: type of manoeuvre, direction the knee ‘gave way’, running speed, knee angle, cutting angle and if the player was accelerating or decelerating. The majority of the injuries analysed occurred in non-contact situations (56%). Of these 37% occurred during sidestepping manoeuvres, 32% in landing, 16% land and step, 10% stopping/slowing and 5% crossover cut manoeuvres. Ninety-two percent of the non-contact injuries occurred at extended knee angles of 30° or less, which is also commonly known to place stress on the ACL and reduce the protective role of hamstrings. Over half (54%) of non-contact injuries occurred whilst decelerating. It would be expected that greater speed and angle cut too would increase the frequency of ACL injury. The results could not confirm this with most injuries occurring at running speeds of slow jogging to running and equal number of injuries occurred at cutting to angles of the ranges 15-45° and 45-75°. These results give greater understanding into potential causes or contributors of ACL injury and information to assist in the development of knee injury prevention programs.

Use of siliconcoach

These researchers used the slow-motion, frame by frame and measurement features of the siliconcoach software to analyse the mechanics of injuries in Australian rules football players.

Disk degeneration and fast bowling in cricket: an intervention study.

Reference

B Elliott, M Khangure. Medicine and Science in Sports and Exercise. Volume 34, Number 11, 2002, Pages 1714-1718.

Source

http://www.acsm-msse.org

(NOTE: Due to copyright restraints, siliconcoach cannot give away this resource. You can search for the article from the link above, however, a fee or subscription may be charged by the supplier.)

Abstract

Purpose: The purpose of this study was to identify the relationship between the incidence of lumber disk degeneration and bowling technique after 3 yr of educational intervention.

Methods: Two groups of cricketers from the Western Australian fast-bowling development squads acted as subjects in this longitudinal study. Group 1 comprised 24 fast bowlers, of mean age 13.4 yr at the commencement of the study. They attended at least three of the four yearly testing sessions between 1997 and 2000. A further 17 of mean age (in 1998) of 13.2 yr attended a minimum of two of three yearly testing sessions between 1998 and 2000, and comprised group 2. Players were filmed laterally and from above by two video cameras during each testing session. Specific technique variables that previously had been linked with an increased incidence of lumbar disk abnormalities were measured from the videos. Magnetic resonance imaging (MRI) scans of the lumbar disks of each player were also recorded at approximately the same time. A yearly half-day clinic and six small group coaching sessions spread over the season were held to assist the bowlers develop techniques that had been linked with a reduction in back injuries.

Results: Data showed that small group coaching significantly reduced the level of shoulder alignment counter-rotation in young fast bowlers. The incidence and progression of lumbar disk degeneration were also significantly reduced in parallel with this decreased shoulder counter-rotation.

Conclusion: Technique assessment and modifications through an educational process aimed at reducing mechanical features that have been linked to back injury decreased the incidence and/or progression of lumbar spine disk degeneration.

Use of siliconcoach

These researchers used siliconcoach software from two camera views to look at trunk and shoulder biomechanics in young cricket bowlers. They also looked at spinal X-rays to gauge spinal degeneration and attempted to find a link between abnormal trunk and shoulder motion when bowling and disk degeneration.