Aaron Fenster

Objective. Quantitative assessment of renal cell carcinoma (RCC) ablation margins requires reliable registration of pre- and post-ablation images. However, ablation-related tissue changes weaken local anatomical correspondence, making deformation estimates potentially non-unique. We developed an uncertainty-aware registration framework for more reliable margin estimation. Approach. An ensemble of plausible registrations was generated by stochastically sampling key registration settings, including the registration mask, similarity metric, B-spline grid spacing, and initial alignment, with Latin hypercube sampling used to improve parameter-space coverage. Registration uncertainty was characterized from this ensemble and propagated into probabilistic margin estimation. Evaluation used 25 biomechanically simulated cases with known deformation fields and 47 clinical RCC cases. Main results. In simulation, 98 ± 2% of kidney voxels had a mean displacement error below 1 mm, and uncertainty was strongly associated with registration error (median case-wise Spearman correlation, 0.99). In the clinical study, our method achieved anatomical alignment consistent with the deterministic baselines, with a median kidney Dice similarity coefficient of 0.944 and a median local mean absolute surface distance of 0.5 mm. In general, our method maintained a sensitivity of 1.00 while achieving higher positive predictive value (PPV) than the deterministic methods. Under the global criterion, PPV was 0.70 versus 0.54–0.58, although the pairwise differences between our method and each deterministic method were not statistically significant. Under the focal criterion, PPV was 0.78 versus 0.29–0.50. The pairwise differences were significant between our method and the standard and penalty-constrained B-spline methods (both p < 0.001), but not between our method and the diffeomorphic method ( p = 0.063). Significance. The framework extends registration uncertainty estimation to RCC ablation-margin assessment by converting variability in plausible correspondences into probabilistic margin maps. It highlights negative-margin regions consistently identified across registrations, reduces registration-sensitive false-positive local tumor progression predictions, and may support more reliable postoperative evaluation.

The 2D projective nature of X-ray radiography presents significant limitations in fluoroscopy-guided interventions, particularly the loss of depth perception and prolonged radiation exposure. Integrating magnetic trackers into these workflows is promising; however, it remains challenging and under-explored in current research and practice. To address this, we employed a radiolucent magnetic field generator (FG) prototype as a foundational step towards seamless magnetic tracking (MT) integration. A two-layer FG mounting frame was designed for compatibility with various C-arm X-ray systems, ensuring smooth installation and optimal tracking accuracy. To overcome technical challenges, including accurate C-arm pose estimation, robust fluoro–CT registration, and 3D navigation, we proposed the incorporation of external aluminum fiducials without disrupting conventional workflows. Experimental evaluation showed no clinically significant impact of the aluminum fiducials and the C-arm on MT accuracy. Our fluoro–CT registration demonstrated high accuracy (mean projection distance $\approx$ 0.7 mm), robustness (wide capture range), and generalizability across local and public datasets. We conducted a phantom insertion experiment using endoleak repair as a representative procedure. Needle insertion error was between 2 mm and 3 mm, with real-time guidance using enhanced 2D and 3D navigation. Overall, our results demonstrated the efficacy and clinical applicability of the MT-assisted approach. To the best of our knowledge, this is the first study to integrate a radiolucent FG into a fluoroscopy-guided workflow.

Hand-held three-dimensional (3D) ultrasound (US) imaging offers critical advantages over conventional two-dimensional (2D) systems by providing richer spatial context and more reliable volumetric measurements for diagnosis and interventions. However, existing 3D US solutions either rely on costly external tracking systems or suffer from cumulative drift errors in purely image-based methods. To address these limitations, we propose a novel, low-cost sensorless 3D US framework that integrates a custom coupling device with a coarse-to-fine optimization algorithm for robust pose refinement. The coupling device embeds geometrically structured patterns into the US images, enabling accurate probe pose estimation directly from image content and supporting reliable volumetric reconstruction. Validation using simulations, phantoms, and in vivo carotid artery studies confirms the method’s accuracy: 2.4% average drift rate in simulations, 2.4% volumetric error and linear measurement error of 0.07 mm, 0.11 mm, and 0.22 mm for lateral, axial and elevational directions in phantoms, and anatomical fidelity comparable to commercial 3D probes in vivo. By removing the need for expensive external tracking or specialized US probes and enhancing robustness over existing image-based methods, our approach improves the accessibility of 3D US imaging, particularly in resource-constrained clinical settings.

PurposeThermal ablation is a minimally invasive therapy used for the treatment of small renal cell carcinoma tumors. Treatment success is evaluated on postablation computed tomography (CT) to determine if the ablation zone covered the tumor with an adequate treatment margin (often 5 to 10 mm). Incorrect margin identification can lead to treatment misassessment, resulting in unnecessary additional ablation. Therefore, segmentation of the renal ablation zone (RAZ) is crucial for treatment evaluation. We aim to develop and assess an accurate deep learning workflow for delineating the RAZ from surrounding tissues in kidney CT images.ApproachWe present an advanced deep learning method using the attention-based U-Net architecture to segment the RAZ. The workflow leverages the strengths of U-Net, enhanced with attention mechanisms, to improve the network’s focus on the most relevant parts of the images, resulting in an accurate segmentation.ResultsOur model was trained and evaluated on a dataset comprising 76 patients’ annotated RAZs in CT images. Analysis demonstrated that the proposed workflow achieved an accuracy =0.97±0.02, precision =0.74±0.23, recall=0.73±0.25, DSC=0.70±0.22, Jaccard =0.58±0.22, specificity =0.99±0.01, Hausdorff distance =6.70±4.44 mm, and mean absolute boundary distance =2.67±2.22 mm.ConclusionsWe used 3D CT images with RAZs and, for the first time, addressed deep-learning-based RAZ segmentation using parallel CT images. Our framework can effectively segment RAZs, allowing clinicians to automatically determine the ablation margin, making our tool ready for clinical use. Prediction time is ∼1 s per patient, enabling clinicians to perform quick reviews, especially in time-constrained settings.

Objective: High dose rate brachytherapy is a standard treatment for cervical cancer; however, intraoperative visualization remains limited by modality-specific constraints and operator variability. We describe a novel integrated mechatronic system for automated registration and fusion of three-dimensional transrectal (3D TRUS) and transabdominal (3D TAUS) ultrasound images, designed to improve image-guided cervical brachytherapy. Methods: The system combines a tracked, four-degree-of-freedom mechatronic 3D TAUS arm with an existing 3D TRUS device and introduces a mechanical referencing tool to enable spatial alignment between the two imaging systems. Forward kinematic tracking of the 3D TAUS arm was implemented using Denavit–Hartenberg parameters, with tracking errors corrected using a linear regression-based transformation. Results: After correction, sub-millimeter tracking error was achieved across clinically relevant configurations. System performance, including geometric fidelity and volumetric accuracy, was evaluated using a custom multi-purpose imaging phantom. All observed geometric errors in the registered 3D US images were sub-millimeter, and volumetric measurements of agar-based inclusions deviated from nominal values by no more than 1.3 cm$^{3}$. The fused 3D TRUS and 3D TAUS images enhanced phantom visualization by mitigating shadow artifacts and expanding the field-of-view, enabling accurate segmentation. Conclusion: The proposed system enables reproducible, minimally operator-dependent ultrasound image fusion with clinically acceptable geometric and volumetric accuracy. Significance: This approach supports ultrasound-based image guidance and potential treatment planning in cervical brachytherapy, offering a cost-effective alternative to CT or MRI, particularly in settings with limited access to advanced imaging technologies.