In Vivo Passive Forces of the Index Finger Muscles: Validating Current Models

Details

• Personal Author:
  Dennerlein J ; Lee D ; Qin J
• Description:
  INTRODUCTION: The passive properties of the muscle have been found to be an important element in the biomechanical modeling of hand movements (Dennerlein et al., 1998; Keir et al., 1996; Sancho-Bru et al., 2001). However, due to the unique physiological characteristics of the intrinsic hand muscles and the complicated structure of the extensor hood mechanism, it is difficult to model the passive forces of the hand. Although few studies have measured the extrinsic muscle passive length-force property from cadaver fingers (Keir et al., 1996; Lee et al., 1990), the biomechanical parameters that predict the passive forces from these cadaver studies have limited in vivo validation. Therefore, we utilized a standard biomechanical model that incorporates typical exponential force-length relationship and in vivo fingertip forces. Through comparing the two, we implemented a particle swarm optimization (PSO) to determine the parameters of the exponential relationship. PSO is a population based stochastic computer algorithm developed by Eberhart and Kennedy in 1995 and it has been successfully applied in many research and application areas. It has been demonstrated that PSO gets better results in a faster, cheaper way compared with other methods (Schutte et al., 2005). METHODS: We conducted a laboratory experiment measuring the passive fingertip forces of index finger in 18 postures in ten subjects. EMG confirmed no activity in the forearm muscles. Fingertip forces were averaged across the ten subjects for each of the 18 postures. For each posture we computed change of muscle length of six muscles -- flexor digitorum profundus (FDP), flexor digitorum superficialis (FDS), radial interosseous (RI), ulnar interosseous (UI), lumbrical (LU) and extensor digitorum communis (EDC) -- based on the excursions of the tendons of the finger adopted from the method used by Lee et al. (1990). Based on these length changes, we calculated muscle passive forces via the exponential relationship with its length excursion (e.g. Lee et al., 1990). Fp = Beta1 x e^(Beta2xDeltal) where, Fp is the passive muscle force, Beta1 and Beta2 are biomechanical coefficients, and Deltal is the muscle length excursion from reference position. From these passive forces we calculated the fingertip force via Ftip = (J^t)-1RFp where Ftip is the fingertip force, J is the Jacobian matrix for the one of the 18 postures, and R is the moment arm matrix for the posture. Ftip Particle swarm optimization searched the parameter space for the biomechanical coefficients Beta1 and + Beta2 which yield the passive fingertip forces that are closest to the measured experimental data. RESULTS AND DISCUSSION: The average values from ten sets of the global best solutions of Beta1 and Beta2 after 10000 iterations were different than values presented by Lee and Rim (Table 1), which are based on fitting the length-force curves from three cadaver fingers. The simulated betas of FDP, FDS and LU from the model were higher than their reported values, and betas of RI, UI and EDC were lower. This may due to the reason that the 18 postures measured in the experiment did not cover the whole range of muscle excursions. The excursions of FDP, FDS and LU calculated from the postures are on the lower side of the exponential curve and the excursions of RI, UI and EDC are on the higher side. We also only explored flexion forces and no extension due to the experimental protocol. This may explain the railing of the EDC passive force parameters. Another reason that may cause the difference is subject variability. The subjects in this study were young and healthy (average age = 29.9 +/- 6.4 years), while the average age was 72 years for the cadavers. The average best global fitness value of the ten runs was 0.0966 N / posture, and the average passive fingertip force across all postures was 0.4957 N. The optimization time for 10000 iterations is about 30 minutes on an IBM computer (Duo processor 6600 at 2.40 GHz). In addition, we have assumed a specific configurations for the tendons of the finger; however, the extensor mechanism of the finger is quite complex and hence the model may not be completely representative. CONCLUSIONS: A biomechanical model of passive muscle forces of the index finger using PSO algorithm was proposed. This model produced promising results to estimate and predict the passive forces and moments for different postures. Future work needs to be done on collecting data on a wider range of excursion distribution to improve the robustness of the model. [Description provided by NIOSH]
• Subjects:
  Biomechanical Phenomena Musculoskeletal System Occupational Health Safety
• Keywords:
  Biomechanical Modeling Biomechanics In Vivo Study Muscle Function Physiological Testing Upper Extremities
• Publisher:
  International Society of Electrophysiology and Kinesiology (ISEK)
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Abstract or Summary
• Place as Subject:
  Massachusetts ; OSHA Region 1
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  229-230
• NIOSHTIC Number:
  nn:20060534
• Citation:
  XVIIth Congress of the International Society of Electrophysiology and Kinesiology, Muscles in Motion: Moving Research into Clinical Practice, June 18-21, 2008, Niagara Falls, Ontario, Canada. McLean L, Bonato P eds. Victoria, British Columbia, Canada: International Society of Electrophysiology and Kinesiology (ISEK), 2008 Jun; :229-230
• Email:
  jax@hsph.harvard.edu
• Editor(s):
  McLean L; Bonato P
• Federal Fiscal Year:
  2008
• NORA Priority Area:
  Healthcare and Social Assistance ; Services
• Performing Organization:
  Harvard University School of Public Health
• Peer Reviewed:
  False
• Start Date:
  20070301
• Source Full Name:
  XVIIth Congress of the International Society of Electrophysiology and Kinesiology, Muscles in Motion: Moving Research into Clinical Practice, June 18-21, 2008, Niagara Falls, Ontario, Canada
• End Date:
  20120228
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:7f7010dbe603bb66fcfe8b5a40395e4478646d53d203bad28eed17e64e62260806cc44b85a86a3e48ddf93c6180b80d0ca086e8fa79bc7e1e103c62cdf9fbb82
• Download URL:
  https://stacks.cdc.gov/view/cdc/224402/cdc_224402_DS1.pdf
• File Type:
  [PDF - 761.42 KB ]