Surgical process modeling of emerging technologies for CT-guided liver tumor ablation

<strong>Purpose</strong>. CT-guided thermal ablation is an increasingly prevalent means of treating liver tumors, with several emerging technologies for navigation and guidance intended to improve the geometric accuracy of targeting. To evaluate the potential advantages and tradeoffs in workflow and performance, this work developed a surgical process model (SPM) for CT-guided liver tumor ablation extended to systems for stereotactic navigation, laser guidance, and image registration. <strong>Method</strong>. Statistical SPMs were developed to evaluate workflow and performance in phases of treatment planning and adaptive treatment for standard-of-care CT guidance, stereotactic navigation with a tracking system on a mobile cart (Nav-on-Cart), stereotactic navigation with the tracking system integrated on the CT scanner gantry (Nav-on- Gantry), and Laser-Guidance using a system of lasers and actuated mirrors built into the CT scanner gantry. The effect of (rigid or deformable) image registration was similarly incorporated in the SPMs to evaluate the effect on interventional decision making and geometric coverage of the planned ablation volume in the Ablation Confirmation (AC) step. Model parameters were derived from expert feedback, technology specifications, prospective clinical observations, review of retrospective data, and published literature. Simulations (250 runs each) were performed for cases with up to 5 tumors, each &lt;50 mm diameter with models representing each guidance technology and with image registration at varying levels of target registration error (εreg). The median and interquartile range (IQR) in procedure cycle time, radiation dose, number of interventional scans, re-ablation rate, and geometric coverage were evaluated quantitatively for each approach and validated in comparison to literature review and/or retrospective clinical data. <strong>Results</strong>. The statistical SPM indicated that navigation and guidance technologies increased the time required for treatment planning: stereotactic Nav-on-Cart (median: 34.8 min), Nav-on-Gantry (median: 32.6 min), and Laser- Guidance (median: 31.4 min), compared to 28.5 min for standard CT-Guidance. The benefits of each technology were evident in the adaptive treatment phase, requiring 47.0 min, 47.9 min, and 53.4 min for Nav-On-Cart, Nav-on- Gantry, and Laser-Guidance, respectively, compared to 57.3 min for standard CT-Guidance. Each of the three guidance technologies resulted in a substantial reduction in the number of interventional scans acquired during the treatment phase: median values 9, 9, and 10 scans, respectively, versus 15 for standard CT-Guidance. In the AC step, accurate image registration improved the detection of insufficient margins and correspondingly increased the rate of re-ablation and the total adaptive treatment time. The decision to re-ablate was found to follow a logistic relationship governed by εreg and was central to achieving the minimum ablative margin (MAM, nominally 5 mm). For example, coarse registration with median εreg = 7.7 mm (comparable to previously reported rigid registration) achieved the target MAM ≥ 5 mm in just 43% of cases, accompanied by a re-ablation rate (R) of just 16%. By comparison, improved registration with εreg ranging 1.2-3.4 mm (comparable to previously reported deformable image registration methods) achieved MAM ≥ 5 mm in 71-88% of cases, with 40-66% re-ablation rate. The SPM provided invaluable insight on threshold levels of εreg required to achieve MAM – e.g., εreg &lt; 4.7 mm required for reliable AC. <strong>Conclusions</strong>. Statistical SPMs provide a valuable tool to understand performance gains in high-level outcome measures associated with emerging technologies and workflows in CT-guided liver tumor ablation. Simulations demonstrated agreement with clinical observations and previous literature and revealed performance gains in treatment time, reduced number of interventional scans, and increased margin coverage for various stereotactic and laser guidance technologies for improved targeting and image registration for improved ablation confirmation.