#  Hill-Type Muscle Models 

 



 ##  

  expand\_more  

 
  

 

   ![05_2013_fig1.png](/sites/g/files/omnuum6301/files/styles/hwp_1_1__720x720_scale/public/biewener/files/05_2013_fig1.png?itok=UgvQ_W9g) 

 

Many studies of human movement (and rehabilitation) rely on Hill-type muscle models to evaluate motor performance based on non-invasive electromyographic (EMG) recordings of muscle function. These studies use activation features derived from EMG recordings to drive the force length (F-L) and force-velocity (F-V) properties of the muscles involved in a particular movement task.

Together with James Wakeling's lab at Simon Fraser University, Research Associate Allison Arnold-Rife, postdoc Sabrina Lee (now at the Rehabilitation Institute of Chicago, Northwestern Univ.), we are conducting studies that seek to link *in vivo* and *in situ* validation of Hill-type muscle models based on animal studies to applications in human cycling performance.

By combining *in vivo* methods for recording muscle activation (from indwelling EMG electrodes), muscle length changes via sonomicrometry and muscle-tendon force from tendon buckles:

Sort   ![Sonomicrometry](/sites/g/files/omnuum6301/files/styles/hwp_1_1__360x360_scale/public/biewener/files/sonomicrometry.png?itok=yAF8WZxi) 

 



   ![Muscle-tendon force buckles](/sites/g/files/omnuum6301/files/styles/hwp_1_1__360x360_scale/public/biewener/files/tendonforcebuckles2.jpg?itok=ZAUF_IHF) 

 



   ![Goat hind limb](/sites/g/files/omnuum6301/files/styles/hwp_1_1__360x360_scale/public/biewener/files/goathindlimbanat.jpg?itok=vYqiigPM) 

 







   ![GoatLG muscle model](/sites/g/files/omnuum6301/files/styles/hwp_1_1__720x720_scale/public/biewener/files/goatlg_musclemodel.jpg?itok=tYQ3yIsx) 

 

We are developing novel muscle models that allow for differential recruitment of fast and slow motor units that can be validated against *in vivo* measures of muscle force and work output. Differential recruitment of the muscle is assessed by wavelet analysis. Our results show that a two-element muscle model gives improved predictive accuracy relative to standard models that assume uniform muscle activation, F-L and F-V properties.