Towards robust 3D registration of non-invasive tactile elasticity images of breast tissue for cost-effective cancer screening

Research output: Working paperDiscussion paper

37 Downloads (Pure)

Abstract

This paper presents current progress on the development of Tactile Imaging, a developing technology for breast cancer screening finding traction in the marketplace, towards non-invasive fully 3D elasticity imaging of the breast. The paper identifies the necessary steps required, and subsequent progress, to develop the technology to image the whole breast robustly which is to be used as a safe screening tool in walk-in clinics. Tactile Imaging has been shown to be capable of binary lesion classification and has seen extensive development, to where benign biopsy rates could be reduced by 23%, but further work is required to make this a clinically practical system for widespread use. Using a hybrid system of Tactile, orientation, and camera sensors it has been demonstrated that robust composite tactile image mosaicking is feasible using the breast vein network as a base map. This paper further outlines the remaining steps needed to turn the current state-of-the-art system from a 2D demonstrator into a fully 3D imaging system that is competitive with other imaging methods, and associated challenges. These being chiefly preparing a phantom reference structure for use in pre-clinical validation, making more stable tactile sensors to reliably perform the new imaging techniques, and building bodies of evidence to build clinical trust in tactile imaging. This work describes that 3D tactile breast imaging is feasible, but that additional work is required to clinically demonstrate these new developments.
Original languageEnglish
Place of PublicationGlasgow
PublisherUniversity of Strathclyde
Pages1-4
Number of pages4
Publication statusPublished - 28 Aug 2023

Keywords

  • breast cancer
  • non-invasive imaging
  • tactile imaging

Fingerprint

Dive into the research topics of 'Towards robust 3D registration of non-invasive tactile elasticity images of breast tissue for cost-effective cancer screening'. Together they form a unique fingerprint.

Cite this