Ablation zone visualization enhancement by periodic contrast-enhancement computed tomography during microwave ablation

Abstract

Purpose Intra-procedural contrast-enhanced computed tomography (CECT) has been proposed to monitor the growth of ablation zone in order to evaluate the efficacy of microwave ablation more accurately than post-ablation image. However, the dilemma of keeping image quality or following dose regulation becomes a huge challenge to apply CECT in real-time monitoring. The purpose of this study is to evaluate the feasibility of applying local highly constrained backprojection reconstruction (HYPR-LR) on periodic CECT image to enhance visualization of ablation zone by statistical and pathological- radiological analysis. Methods Low-dose ( CTDIvol≤1.49 italicmGy), temporal CECT volumes were acquired during microwave ablation on normal porcine liver. HYPR processing was performed on each volume after image registration. Ablation signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were collected to evaluate the degree of enhancement of image quality and ablation zone visualization. Ablation zone was manually segmented on HYPR and non-HYPR images and compared the similarity to gross pathology by measuring Dice coefficient. The dimension of ablation zone was also compared to gross pathology by correlation and measurement difference. Results The SNR and CNR of ablation zone were increased after involving certain slices into HYPR processing. The manually segmented ablation zone was highly similar to gross pathology with DICE coefficient 0.81±0.03, while the low-dose CECT had smaller DICE coefficient with 0.72±0.05. Both HYPR image and low-dose CECT has high correlation coefficient to gross pathology with 0.99 and 0.94, but the variance of measurements were less on HYPR image than those on unprocessed images. The relative difference in area, length of long axis and short axis for HYPR image were 13.1±5.6%, 9.7±4.2% and 15.2±2.8 %, which were less than those for low-dose CECT with 37.5±6.0%, 17.7±2.8% and 28.9±5.4%. Conclusion HYPR processing applied to periodic CECT images can enhance ablation zone visualization. HYPR processing may potentially enable CECT in real-time ablation monitoring under strict regulation of radiation dose.

Comparison of different 4D CT-Perfusion algorithms to visualize lesions after microwave ablation in an in vivo porcine model

“…Shape similarity was assessed through calculating the absolute and relative difference between lesion areas, with the relative difference being defined as the absolute difference in area between the histology and CTP source images divided by the area in histology. We also calculated the relative differences in lesion size, defined as the difference between measured area in histology and CT source images divided through the area in histology [4,23].…”

Two‐dimensional ultrasound‐computed tomography image registration for monitoring percutaneous hepatic intervention

“…The research presented in this paper describes utilization of the “CT virtual slice” concept previously described by Wu et al . The CT virtual slice is obtained from a 3D CT volume by colocating the imaging planes corresponding to a 2D clinical US B‐mode image used to guide placement of the ablation antenna into the liver tumor to be ablated.…”

“…The research presented in this paper describes utilization of the "CT virtual slice" concept previously described by Wu et al 60 The CT virtual slice is obtained from a 3D CT volume by colocating the imaging planes corresponding to a 2D clinical US B-mode image used to guide placement of the ablation antenna into the liver tumor to be ablated. Following computation of the colocated virtual slice from the 3D CT volume comprised of cross-sectional images or slices through the body, we account for local deformations present in these due to differences in the respiratory and cardiac phases and local deformation present with the US transducer.…”

“…In this section, we describe computation of the 2D CT virtual slice that would correspond to the clinical B‐mode image used for antenna placement. A flow‐chart of the registration processing sequence is shown in Fig.…”

“…In this paper, we derive an algorithm for extracting, matching and colocating a CT virtual slice to clinical US B‐mode images from 20 patients during ablation procedures under general anesthesia . We utilize the non‐anatomical microwave antenna to initiate and colocate possible planes and finalize the virtual slice with anatomical landmarks common between the CT and US B‐mode image. Finally, an affine transformation is utilized to register the CT virtual slice to the US B‐mode imaging plane.…”