Below-knee residual limb shape change measurement and visualization

Paul K. Commean, Barry S. Brunsden, Kirk E. Smith, Michael W. Vannier

Research output: Contribution to journalArticlepeer-review

24 Scopus citations


Objective: The ability to measure and visualize shape change (deformation) of a residual limb within and between prostheses is an important step toward improved prosthetic fit assessment. The objective was to develop measurement and visualization methods for below-knee residual limb soft tissue shape change after donning and loading a prosthesis to detect small shape changes (30mm or less). Design: Spiral X-ray computed tomography imaging was utilized to acquire 3D volumetric data of the below-knee residual limb and prosthesis in situ from poor- and a good-fitting prostheses without and with a load. A new sum projection depth-shaded cylindrical mapping technique to measure and visualize small changes in shape was developed. From the volumetric data, the relative displacement of small lead markers placed on the residual limb's skin surface were measured using multiplanar reconstruction images and cylindrical maps. Displacement measurements averaging 15mm or less were obtained. Results: The precision and accuracy was 1mm and 2mm, respectively, when measuring the shape change or deformation of the skin surface from the sum projection cylindrical maps. The skin surface deformation was at least 7mm or greater when comparing marker locations between scans with the prostheses in situ. Conclusion: These new image-based measurement and visualization methods provide a feasible means for measuring and displaying lower extremity residual limb shape change within and between different prostheses with and without loading.

Original languageEnglish
Pages (from-to)772-782
Number of pages11
JournalArchives of Physical Medicine and Rehabilitation
Issue number7
StatePublished - Jul 1998


Dive into the research topics of 'Below-knee residual limb shape change measurement and visualization'. Together they form a unique fingerprint.

Cite this