A compact system for intraoperative specimen imaging based on edge illumination x-ray phase contrast

Phys Med Biol. 2019 Nov 26;64(23):235005. doi: 10.1088/1361-6560/ab4912.

Abstract

A significant number of patients receiving breast-conserving surgery (BCS) for invasive carcinoma and ductal carcinoma in situ (DCIS) may need reoperation following tumor-positive margins from final histopathology tests. All current intraoperative margin assessment modalities have specific limitations. As a first step towards the development of a compact system for intraoperative specimen imaging based on edge illumination x-ray phase contrast, we prove that the system's dimensions can be reduced without affecting imaging performance. We analysed the variation in noise and contrast to noise ratio (CNR) with decreasing system length using the edge illumination x-ray phase contrast imaging setup. Two-(planar) and three-(computed tomography (CT)) dimensional imaging acquisitions of custom phantoms and a breast tissue specimen were made. Dedicated phase retrieval algorithms were used to separate refraction and absorption signals. A 'single-shot' retrieval method was also used, to retrieve thickness map images, due to its simple acquisition procedure and reduced acquisition times. Experimental results were compared to numerical simulations where appropriate. The relative contribution of dark noise signal in integrating detectors is significant for low photon count statistics acquisitions. Under constant exposure factors and magnification, a more compact system provides an increase in CNR. Superior CNR results were obtained for refraction and thickness map images when compared to absorption images. Results indicate that the 'single-shot' acquisition method is preferable for a compact CT intraoperative specimen scanner; it allows for shorter acquisition times and its combination of the absorption and refraction signals ultimately leads to a higher contrast. The first CT images of a breast specimen acquired with the compact system provided promising results when compared to those of the longer length system.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Breast Neoplasms / diagnostic imaging*
  • Breast Neoplasms / surgery
  • Female
  • Humans
  • Image Processing, Computer-Assisted / methods*
  • Intraoperative Period
  • Margins of Excision
  • Mastectomy, Segmental / methods
  • Phantoms, Imaging*
  • Radiography
  • Reoperation
  • Signal-To-Noise Ratio
  • Tomography, X-Ray Computed*
  • X-Rays