Evaluating Wheelchair Cushion Designs


Pressure ulcers are a common risk for wheelchair users after spinal cord injuries (SCI), and can cause deep tissue injuries at the interface between the ischial tuberosity and overlying soft tissues. Research into wheelchair cushions explores how they can be optimised to better distribute loads between buttocks and a support surface, when SCI pathoanatomies and pathophysiologies such as cortical bone loss, muscular atrophy and spasms are present. In this study, the biomechanical performance of an air cell-based cushion was compared to a standard, flat foam cushion using Simpleware software and Finite Element (FE) simulation. A range of different stresses were analysed using different anatomical model variations, with the goal of evaluating which cushion provide the best protection for wheelchair use.


  • Single coronal MRI slice of a male subject acquired from previous work
  • Segmentation of data and generation of model variants in Simpleware ScanIP
  • Generation of 3D thin slice FE model in Simpleware +FE for analysis in Pardiso solver of FEBio (v1.5.1)
  • ACB cushion found to be more effective at distributing internal loads

Thanks to

Faculty of Engineering, Tel Aviv University: A. Levy • A. Gefen 
Efficacy Research, Standards and Public Policy, ROHO Inc: K. Kopplin


Levy, A., Kopplin, K., Gefen, A., 2014. An air-cell-based cushion for pressure ulcer protection remarkably reduces tissue stresses in the seated buttocks with respect to foams: Finite element studies. Journal of Tissue Viability, 23(1), 13-23.

Image Processing

A single coronal MRI slice was acquired from a male subject to generate a 4 mm thick anatomically realistic geometrical model of the left buttock. The subject was scanned using a non-weight bearing configuration and then a full weight-bearing on a semi-rigid support. The data was imported to ScanIP to segment the MRI image (864 x 608 pixels) into tissue components representing the IT bone, the gluteus maximus skeletal muscle, fat tissue and skin, and either a flat foam cushion or an ACB cushion. The slice was also uniformly extruded to a 4-mm depth, representing the MRI resolution in the Z axis. Fifteen model variants were created incorporating a range of anatomical adaptations, and were tested using two flat foam cushion and an ACB cushion.

FE Mesh Generation

Simpleware ScanIP+FE was used to generate a 3D mesh for FEA. The model included 50,000 4-node linear tetrahedrons assigned to the different tissues, as well as 10,400 8-node linear hexahedrons relating to the flat foam cushion, and 245,235 4-node linear tetrahedrons assigned to the ACB cushion. The dense mesh of the cushion was needed to allow convergence. Simulations were set up using PreView (v1.8) and analysed using the Pardiso linear solver of FEBio (v1.5.1), before being post-processed using PostView (v1.4). The runtime of each model variant took between 7 and 9 hours using a 64-bit Windows 7 workstation (Intel Core i7 920 2.57 GHz CPU and 12 GB of RAM).

FE Simulation and Result

Properties for the slice were assigned, including elastic moduli and boundary conditions for simulating weight bearing. Air cells collapse patterns in the ACB cushion were verified by photographing a transparent phantom being placed over a cushion. The FE testing compared peak effective, compressive, tensile and shear Cauchy stresses for muscle, fat and skin tissues. Results demonstrated greater immersion and lower peak stress components for the ACB cushion compared to a standard foam. FE modelling of the ACB cushion was able to demonstrate its effectiveness for reducing the risk of pressure ulcers, while suggesting the potential benefits of future full 3D FE modelling of buttocks-cushion interactions.