The research, led by Zerui Sun and Yingdong Hu with senior author Hongxing Song, tackles a deceptively hard engineering problem. Cone-beam CT, acquired by rotating a mobile C-arm around the patient, delivers an excellent three-dimensional map of bony anatomy, but it is a snapshot that quickly goes stale as the body moves. Tracked ultrasound, by contrast, can be swept across the spine in real time and carries no ionizing radiation, yet its images are noisy, prone to artifacts, and often show only fragments of the bone surface. Stitching the two modalities together requires knowing exactly where the ultrasound probe is in space, exactly how the ultrasound image relates to the probe, and how to mathematically align a partial, artifact-laden surface to a complete CT-derived bone model. Each step introduces error, and the study’s central contribution is measuring, with unusual rigor, how much error each step actually contributes.

The team built an integrated system combining an ultrasound machine, a C-arm capable of cone-beam CT, an optical tracking system that follows instruments in space, a robot-controlled probe holder that sweeps the ultrasound transducer at a constant speed, and a navigation workstation that fuses everything together. Rather than testing on patients, they used a lumbar spine phantom covering the L1 to L4 vertebrae, scanned while submerged in water. Water eliminates the air-tissue interfaces that plague ultrasound in real bodies, giving the researchers a best-case environment in which to isolate the performance of the registration mathematics themselves. That choice matters for interpreting the results: this is a validation of a workflow, not yet a demonstration in living tissue.

The first technical hurdle was probe calibration, the process of determining the precise transformation between the ultrasound image pixels and the tracked probe coordinate frame. Getting this wrong by even a fraction of a millimeter poisons every downstream measurement. The researchers designed an optimized double-layer phantom containing four N-shaped wires, a classic calibration geometry in which wires cross the imaging plane at known orientations, and calibrated the probe using eighty image sets. The ultrasound image scale came out to 0.16 millimeters per pixel. The full-data calibration error, computed as a pooled three-dimensional vector root-mean-square over all residuals, was 0.627 millimeters, and the per-image-set RMS across the eighty sets averaged 0.621 plus or minus 0.086 millimeters. Critically, the team also performed a split-half held-out validation, calibrating on one subset of images and testing on a disjoint subset, and the error barely moved, to 0.639 plus or minus 0.052 millimeters. That held-out figure is the more honest number, because it shows the calibration generalizes rather than merely fitting the data it was tuned on.

With calibration established, the harder problem was registration: aligning the ultrasound-derived bone surface to the cone-beam CT surface. The CBCT volume was segmented using a 400-Hounsfield-unit threshold with manual refinement, a standard approach for isolating cortical bone. The researchers then applied a three-stage registration pipeline. A similarity transform provided a coarse initial alignment, a multiresolution point-to-point iterative closest point algorithm, or ICP, refined the rigid alignment across progressively finer scales, and finally a thin-plate spline, or TPS, deformation absorbed the residual nonrigid mismatch between the two surfaces. The thin-plate spline is a mathematically elegant interpolant, often described as the higher-dimensional analogue of a flexible metal sheet bent over fixed control points, and it allowed the ultrasound surface to warp slightly to accommodate soft-tissue artifacts and incomplete visibility without abandoning the rigid skeleton of the alignment.

The headline numbers are striking. Evaluated on ten independently acquired ultrasound scans at eight bilateral pedicle-center landmarks that had been deliberately withheld from the optimization process, the point-to-point ICP stage alone achieved a mean target registration error of 2.98 plus or minus 0.23 millimeters. After the thin-plate spline refinement, that error dropped to 1.26 plus or minus 0.08 millimeters, a statistically significant improvement with a p-value below 0.001. Whole-surface distance metrics told the same story: the mean, RMS, and 95th-percentile Hausdorff distances fell from 1.80, 2.05, and 3.61 millimeters to 0.77, 0.88, and 1.54 millimeters respectively. In practical terms, the fused ultrasound view of the spine matched the CT model to within roughly the thickness of a coin at the exact landmarks where a surgeon would place pedicle screws.

One of the study’s most valuable findings is almost a negative result. The team compared point-to-point ICP with point-to-plane ICP, a variant that aligns surface normals and often converges faster and more accurately on rigid surfaces. Point-to-plane did modestly improve the rigid registration stage, but when the full pipeline including the TPS refinement was run, the final target registration error changed by only 0.005 millimeters. The lesson is that the nonrigid refinement stage, not the choice of rigid ICP formulation, is what drives the final accuracy. Supplementary computational analyses probing operator effects, scan pitch, frame density, and dynamic inconsistency reinforced the picture that the pipeline is robust to reasonable variations in acquisition conditions.

Speed matters as much as accuracy in an operating room, and the timing figures are sobering but transparent. The effective ultrasound acquisition region spanned roughly 170 millimeters, which at a sweep speed of 5 millimeters per second takes about 34 seconds of robot-controlled probe motion. The cumulative machine-processing time after scanning was 143.7 seconds, though the thin-plate spline solve itself required only about 0.06 seconds. The authors are explicit that this human-in-the-loop workflow is not suitable for continuous real-time use; it is a registration procedure performed once per case, not a live-updating navigation display. For procedures like pedicle screw placement, where anatomy is fixed relative to the navigation frame once the patient is positioned, a one-time registration of this accuracy could still be clinically transformative, cutting radiation exposure while preserving three-dimensional guidance.

The authors are equally candid about the limits of what they have shown. Because the study used a water-submerged phantom, the notorious degradation of ultrasound in real tissue, with its shadowing from bone, scattering from fat, and air artifacts, remains untested. No cadaveric, animal, or clinical validation has yet been performed, and the study explicitly does not validate a complete robot-assisted surgical system because instrument execution, needle insertion, and robotic puncture were not assessed. The dynamic robustness analyses were computational rather than physical. These caveats are not footnotes; they are the difference between a promising engineering result and a proven clinical tool.

Still, the significance of the work lies in its methodological discipline. By withholding landmark points from optimization, reporting held-out calibration errors, using run-level statistics across ten independent scans, and quantifying every stage of the pipeline separately, the Beijing team has produced one of the more rigorous accuracy characterizations of ultrasound-to-CBCT fusion in the spine navigation literature. Millimeter-level fusion accuracy, achieved without any additional patient radiation beyond a single low-dose cone-beam acquisition, is exactly the performance envelope that next-generation navigation systems will need if ultrasound is to take on a larger role in the operating room. The road from phantom to patient runs through tissue, cadavers, and clinical trials, but this study has established a credible, precisely measured starting point for that journey.

Subject of Research: Phantom-based accuracy validation of ultrasound and C-arm cone-beam CT image fusion for spine navigation

Article Title: Phantom-based accuracy validation of multimodal ultrasound and C-arm cone-beam CT image fusion for spine navigation

Article References: Sun, Z., Hu, Y., Fan, Y., Wang, J., & Song, H. (2026). Phantom-based accuracy validation of multimodal ultrasound and C-arm cone-beam CT image fusion for spine navigation. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02889-w

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02889-w

Keywords: ultrasound, cone-beam CT, image fusion, spine navigation, target registration error, phantom study, probe calibration, iterative closest point, thin-plate spline, surgical navigation, radiation dose reduction, medical imaging

The solution, developed at Amsterdam University Medical Center and now demonstrated across three major European centres, is called contrast tagging. The concept is deceptively simple. During CT hepatic arteriography, or CTHA, contrast medium is injected directly into the hepatic artery through a small catheter threaded from the groin. At the precise moment when the tumour lights up with maximal enhancement, the radiologist switches on the ablation energy. The heat traps the iodinated contrast inside the tumour, effectively freezing a bright, hyperdense snapshot of the cancer’s exact boundaries into the post-ablation scan. When the unenhanced CT is performed immediately afterwards, the tumour appears as a clearly visible bright region sitting inside the darker ablation zone, allowing the physician to see, at a glance, whether the margins are adequate.

The technical choreography behind the technique is worth appreciating. Patients are brought to the angiography suite, where a 5 French sheath is placed in the right common femoral artery and a catheter is advanced to the common hepatic artery. The patient then moves to the CT suite, where a dual-phase CTHA is acquired 7 and 22 seconds after injecting 40 millilitres of a 1:1 iodinated contrast and saline solution at 4 millilitres per second. These two time points capture late arterial and early portal venous phases, revealing when the tumour reaches peak enhancement. The microwave antenna, in this series an Emprint HP system from Medtronic-Covidien, is positioned within the tumour under CT fluoroscopy, and energy delivery is triggered at the predetermined moment of maximal tumour perfusion. CT fluoroscopy confirms that contrast is being retained, that the lesion is shrinking as expected, and that gas formation around the zone is proceeding normally.

The pictorial essay, led by Domenico Santangelo and Susan van der Lei with senior author Martijn Meijerink, presents eleven clinical cases drawn from Amsterdam UMC, University Hospitals of Strasbourg, and the Royal Marsden Hospital in London. The cases span an impressive range of disease: hepatocellular carcinoma, colorectal liver metastases, a pancreatic neuroendocrine tumour metastasis in a 29-year-old woman with MEN-1 syndrome, and a breast cancer liver metastasis. Tumour sizes ranged from 12 to 27 millimetres, and despite variations in enhancement patterns, energy parameters, and antenna positioning, tagging consistently rendered the original tumour boundaries visible on the immediate post-ablation scan. In several cases, the tagged tumour could be seen nested perfectly within the ablation zone, giving radiologists direct visual confirmation of a complete kill with sufficient circumferential margin.

Why does the margin matter so much? Decades of outcome data show that the single most influential factor in preventing local tumour progression is the peri-ablation safety margin, defined as the distance between the tumour’s edge and the boundary of the ablation zone. Traditional teaching held that a circumferential margin of at least 5 millimetres was required, though more recent work using three-dimensional volumetric assessment suggests margins as small as 3 millimetres may suffice in specific settings. The problem is that measuring this margin has always depended on software that fuses pre-ablation and post-ablation images, a process that is notoriously fragile. Differences in breathing mode, patient positioning, tissue shrinkage, thermal injury, and perilesional oedema can all deform the liver and throw off the registration. Hydrodissection, in which fluid is injected to push vulnerable organs like the kidney or colon away from the ablation zone, distorts anatomy even further.

This is where tagging shows its most compelling advantage: it is immune to anatomical distortion. Because the contrast signature marks the tumour itself on the same scan that shows the ablation zone, the relationship between the two is preserved regardless of how much the surrounding liver has shifted or deformed. In one illustrative case, an 80-year-old man with a hepatocellular carcinoma in segment VI underwent hydrodissection with a 5:1 saline-contrast mixture to displace his right kidney. Despite the expected anatomical alteration, the tagged tumour remained clearly visible, and the minimal ablation margin could be assessed reliably without depending on co-registration integrity. Notably, the tagging in this case appeared less pronounced, which the authors attribute to the lower iodine concentration of the contrast agent used, 270 milligrams per millilitre compared with 300 or 350 in other cases.

The essay is refreshingly honest about the technique’s pitfalls. In one case, a bilobed hepatocellular carcinoma was only partially ablated in a first pass; because no second contrast bolus was injected before treating the anterior component, that portion retained no contrast and remained invisible on the post-ablation scan. Tagging, the authors stress, depends on real-time contrast uptake immediately before energy delivery, so any tumour component not enhanced at the moment of ablation will not be tagged. Another case showed that small satellite nodules several millimetres from the main lesion failed to tag, likely because they sat in the passive conduction heating zone, farther from the antenna tip. The authors also demonstrate that tagging is transient: in one patient, the hyperdense rim marking the original tumour faded progressively on scans performed three days and three weeks later, meaning margin assessment must happen immediately after the procedure.

Perhaps most striking is the technique’s success with tumours that are hard to see in the first place. Two cases involved hypovascular colorectal metastases in patients heavily pretreated with chemotherapy, lesions so faint on early-phase CTHA that they were barely perceptible. Even so, subtle peripheral enhancement captured at the 22-second mark was enough to tag the tumour borders, and the tagged region on the post-ablation scan matched the lesion’s morphology on the pre-procedural MRI almost perfectly. In another case, tagging succeeded in a tumour adjacent to a region previously treated with stereotactic body radiotherapy, a scenario where altered perfusion typically complicates image interpretation. A research bolus-tracking experiment, acquiring single-slice scans every 2 seconds after injection, identified peak enhancement at 10 seconds in one patient, supporting the hypothesis that individualized timing maximizes contrast retention and tagging quality.

The technique also compares favourably with the established alternative of tagging tumours with ethiodized oil, or Lipiodol. While Lipiodol accumulates durably in tumours and persists after ablation, it reduces tumour perfusion in ways that can make ablation zones less predictable, and it carries risks of systemic oil embolism, hepatotoxicity, and diffuse non-tumour retention when injected lobarly. The contrast tagging approach requires no additional time or cost when integrated into an existing CTHA-guided ablation workflow, and the authors report it demanded a near-zero learning curve at centres already performing CTHA guidance. No tagging-specific complications were recorded, and no catheter dislocations or groin hematomas occurred in the cohort. The estimated radiation exposure, a CTDI of 77 milligray across the typical 6 to 8 CT acquisitions, remains within the expected range for these complex procedures.

Looking forward, the authors argue that tagging could dovetail naturally with artificial intelligence and deep learning platforms that automate lesion segmentation, registration, and margin quantification. A persistent hyperdense landmark corresponding to the original tumour gives fusion algorithms a reliable anchor in the immediate post-ablation scan, potentially transforming margin assessment from a subjective art into an objective, reproducible measurement. The authors are careful to note that confirmation software remains essential, since it is still unclear whether tagging is perfectly specific to tumour tissue or may also involve adjacent enhancing parenchyma near the antenna tip. Prospective studies are now needed to validate the technique’s impact on local tumour control and to clarify the biological mechanism of contrast retention. If those trials succeed, contrast tagging during CTHA-guided ablation could become a standard component of image-guided liver cancer treatment, giving physicians something they have never had before: a clear, immediate, and honest picture of whether the tumour is truly gone.

Subject of Research: Contrast tagging during CT hepatic arteriography-guided thermal ablation of liver tumours

Article Title: The role of CT hepatic angiography with contrast tagging in liver tumour ablation – an international pictorial essay

Article References: Santangelo, D., van der Lei, S., Johnston, E. W., Cazzato, R. L., De Cobelli, F., & Meijerink, M. R. (2025). The role of CT hepatic angiography with contrast tagging in liver tumour ablation – an international pictorial essay. CVIR Oncology, 1(1), Article 12. https://doi.org/10.1007/s44343-025-00012-4

Image Credits: AI Generated

DOI: 10.1007/s44343-025-00012-4

Keywords: liver tumour ablation, CT hepatic arteriography, contrast tagging, microwave ablation, safety margins, minimal ablation margin, interventional radiology, hepatocellular carcinoma, colorectal liver metastases, image fusion, local tumour progression, CVIR Oncology