A solution for co-locating 2D histology images in 3D for histology-to-CT and MR image registration: closing the loop for bone sarcoma treatment planning

TL;DR

This paper introduces a method to accurately align 2D histology images within 3D CT volumes for better tissue characterization in bone sarcoma treatment planning, enhancing surgical precision.

Contribution

The work presents a novel, validated approach for co-locating histology slices in 3D CT, improving tissue analysis without disrupting clinical workflows.

Findings

Achieved a plane misalignment of 0.19 ± 1.8 mm.

User input caused 0.08 ± 0.2 mm translation and 0-1.6° deviation.

Method comparable to existing co-location accuracy in prostate studies.

Abstract

This work presents a proof-of-concept solution designed to improve the accuracy of radiographic feature characterisation in pre-surgical CT/MR volumes. The solution involves 3D co-location of 2D digital histology slides within ex-vivo, tumour tissue CT volumes. In the initial step, laboratory measurements obtained during histology dissection were used to seed the placement of the individual histology slices in corresponding tumour tissue CT volumes. The process was completed by aligning corresponding bone in histology images to bone in the CT using in-plane point-based registration. Six bisected canine humerus datasets of ex-vivo CT and corresponding measurements were used to validate dissection placements. Digital seeding exhibited a plane misalignment of 0.19 +- 1.8 mm. User input sensitivity caused 0.08 +- 0.2 mm in plane translation and between 0 and 1.6 degrees deviation. These are of similar magnitude to reported misalignment of 0.9-1.3 mm and 1.1-1.9 degrees in related prostate histology co-location [1]. Although this work only reported on animal femur sarcoma CT images and histology slide images, the solution can be generalised to various sarcoma geometries and presentation sites. Furthermore, the solution co-locates high-fidelity data to advance tissue characterisation with minimal disruption to existing clinical workflows. Improved tissue characterisation accuracy, supported by the resolution of histology images, can enhance surgical planning accuracy and patient outcomes by bringing the insights offered by histology closer to the start of the presentation-diagnosis-planning-surgery-recovery loop.