[home] [Personal Program] [Help]

---

Session: Motor Control II
Session starts: Friday 25 January, 13:00
Presentation starts: 13:15
Room: Lecture room 558

---

Roberto Garcia van der Westen ()
Pieter Oomen ()
Lodewijk van Rhijn ()
Kenneth Meijer ()

Abstract:
Knee osteoarthritis (KNOA) is a condition characterized by the degeneration of the cartilage within the tibio-femoral joint. Increasing longevity of the population give arise to an increasing number of cases over time. The degeneration of articular cartilage is a complex process which involves interrelated biological, mechanical and structural pathways. Clinical and laboratory reports lead to conflicting conclusions on the influence of mechanical effects in the progression of KNOA. Therefore, the early diagnosis and treatment of KNOA could improve significantly from a better understanding of mechanical factors such as peak loads and moments. Unfortunately, it is impossible to measure the above described parameters in-vivo, at least not on a healthy population. Musculoskeletal modeling has therefore the potential to become a very powerful tool in the assessment of KNOA. Musculoskeletal modeling is a non-invasive method that allows the estimation of loads and moments of the knee joint. Before the implementation of such a tool in a clinical setting much research is needed in terms of understanding such models and how to validate them. The current study explores the influence of different recruitment criteria on the estimated knee loads. Motion capture data during activities of daily life were taken from 5 older healthy female participants and used in the AnyBody modeling software to create scaled patient specific models. Muscle activation, forces and moments on the knee were calculated based on an inverse dynamics approach, which uses optimization to solve the statical indeterminacy. Polynomial (p=2,3,4,and 5), Min/Max and composite criteria were compared. Obtained average peak loads are comparable to those shown in other studies (peak PD force = 3.1±0.7 BW), but increased compared to in-vivo studies with telemetrised prosthesis data (peak PD force = 2.5 BW). In terms of peak loads, all criteria except for the 2th order polynomial showed good agreement. Polynomial p=2 overestimated the proximo-distal forces by 13% for gait compared to other criteria. No significant differences were observed for stair ascent or descent. Muscle activation patterns showed good agreement with measured EMG data, only the Min/Max recruitment showed very fast activations, which might not be physiological feasible. Polynomial of the 3rd order showed smoother activations and similar peak forces. The assumption that the central nervous system tries to reduce the load on the muscles, might be good enough to solve the optimization problem for most activities. Moreover, from the available algorithms, the polynomial of 3rd order shows the most attractive features, both physiologically and computationally.