Search Results (152)

Disappearing liver metastases (DLMs) remain a challenging problem in the management of colorectal liver metastases (CRLMs). Although modern systemic therapy may produce complete radiological responses, radiological disappearance does not reliably indicate complete pathological response. Conversely, mandatory treatment of every original metastatic site may cause unnecessary parenchymal sacrifice when a lesion is no longer identifiable on hepatobiliary contrast-enhanced magnetic resonance imaging or intraoperative ultrasonography. Current strategies—including treatment of the presumed site, selective pre-treatment fiducial marking where feasible and considered appropriate, and structured surveillance—therefore involve important trade-offs. We propose a hypothesis-generating, lesion-specific strategy for a narrowly selected subgroup: thermal ablation before systemic therapy of one or a few small CRLM, preferably no larger than 2 cm, considered at higher anticipated risk of post-treatment loss of localization. Selection should distinguish the probability of radiological disappearance from its clinical consequences. Candidate lesions should be clearly and unambiguously identifiable on high-quality baseline imaging, safely accessible for ablation with a margin greater than 5 mm and ideally approximately 10 mm, and located outside any planned resection field, particularly when their depth or anatomical position would make empirical treatment difficult after disappearance. Recent randomized evidence supports thermal ablation as definitive local therapy for appropriately selected small CRLM but does not validate the proposed pre-systemic treatment sequence. This approach should not delay urgently required systemic therapy, compromise biological selection, or be used indiscriminately in high-volume, rapidly progressive, uncontrolled extrahepatic, or technically unsuitable disease. It is also generally unnecessary when the lesion is encompassed by a planned resection or total hepatectomy within a liver transplantation pathway. Upfront ablation may secure local control while the target remains visible, preserve liver parenchyma, and avoid premature interruption of systemic therapy or acceleration of surgery solely because a responding lesion is becoming radiologically occult. However, it may cause overtreatment and procedural morbidity and remains unsupported by prospective comparative evidence. A multicentre feasibility study, followed by prospective comparison with contemporary standard sequencing and, where feasible and considered appropriate, fiducial-based localization, is required to evaluate this strategy. Full article

Precise end-effector positioning of a robotic arm is critical for accurate performance in robotics, which directly impacts the system’s performance in applications requiring high precision. This paper presents an optimized tag-based inverse kinematics approach for a 6 DoF robotic arm using a Modified Biogeography-Based Optimization (MBBO) algorithm, which is an enhanced version of the original Biogeography-Based Optimization (BBO), a population-based evolutionary algorithm inspired by the natural distribution of species across habitats. The target position is identified via AprilTag visual fiducial markers and integrated into the inverse kinematics solver. Performance was evaluated against Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and BBO through 3-dimensional simulations involving ten target points. Results show that MBBO achieves reduced positioning error compared to GA, PSO, and BBO, resulting in lower end-effector position errors. The findings highlight the effectiveness of combining visual tag detection with advanced optimization for precise robotic arm control. Full article

Reliable unmanned surface vehicle (USV) recovery requires near-field maritime remote sensing outputs that remain stable during the final approach. Planar fiducial geometry provides metric pose estimates, but its depth channel is sensitive to corner-localization noise, apparent marker shrinkage, glare, reflection, and vessel vibration. We present VARE (Visual-Adaptive Ranging and Estimation), a geometry-anchored perception pipeline that combines ArUco-based Perspective-n-Point pose recovery, dual-path bearing fusion, MiDaS-assisted range stabilization, depth-consistency confidence weighting, and innovation-adaptive temporal filtering. VARE is a system-level integration rather than a new neural architecture, PnP solver, or end-to-end docking controller. The pipeline explicitly separates image-centroid bearing, translation-vector bearing, marker-normal heading, camera-frame horizontal approach range, and lateral offset. Independent RTK-synchronized external references, with measured lever-arm corrections between the RTK antenna, camera optical center, and marker reference point, are used for pool and near-shore evaluation. In controlled land tests, VARE reduced independent-reference angular RMSE by 34.0–49.3% relative to the pixel-only baseline and by 22.0–36.4% relative to a static-filter PnP variant. Across 15 pool-based approach trials, the full vision-only configuration achieved a horizontal bearing RMSE of 0.46 degrees, a range MAE of 0.82 m, and a range RMSE of 0.90 m. Relative to the matched IPPE-square geometry baseline with One-Euro filtering, the corresponding descriptive reductions were 14.8%, 4.7%, and 5.3%; the modest range differences are not presented as universally significant. Trial-level summaries, confidence intervals, and data-availability provisions are added to support reproducibility. The results support VARE as a candidate perception module for RTK-referenced USV recovery guidance, while full six-degree-of-freedom validation, session-level dropout survival, and closed-loop capture success remain future work. Full article

Unmanned Aerial Vehicles (UAVs) usually rely on visual fiducial markers for autonomous navigation and precision landing. However, standard markers suffer from limited operational ranges, becoming undetectable when the camera is either too far or too close. While recursive and fractal markers have been proposed to address this issue, existing approaches either require the marker’s center to remain visible, making them vulnerable to occlusion, or are limited in their recursion depth and placement. We propose a novel Recursive ArUco marker design. Our method allows any standard square fiducial marker to be transformed into a recursive marker with an arbitrary depth. By employing a modified bit-sampling strategy during detection, we embed complete markers within both the black and white bits of the parent marker. This approach guarantees unlimited recursion depth and robust detection even with partial occlusion, as it does not rely on the marker’s center being visible. Furthermore, by maintaining a single, unique identifier across all recursive scales, our proposal provides an extensive dictionary of multiple unique landing pads. Our markers allow fleets of UAVs to operate simultaneously, with each drone landing at its designated location—a feature not supported by existing approaches due to their structural and dictionary constraints. Full article

Background/Objectives: Cone-beam computed tomography (CBCT)-to-intraoral-scan registration in edentulous implant planning relies on radiopaque fiducial markers, yet how marker design governs marker-geometry reproduction is unclear. We quantified the effects of shape, size and configuration and where the error arose. Methods: Eighteen edentulous mandibular denture designs (three shapes × three sizes, 2–4 mm × two configurations; five markers each) were printed and CBCT-imaged three times at 0.2 mm voxel. Distances, angles, a leave-one-marker-out (LOMO) error and a markerless control were analyzed at the design level (n = 18). Results: Acquisition reproducibility was voxel-limited (median 49 μm). Marker size governed trueness and precision (partial ω²_p 0.25–0.65), with no shape effect detected. The LOMO error fell with size (541 to 359 μm; p = 0.004), whereas the markerless control did not (p = 0.653); the size effect differed between approaches. Twenty percent of 2 mm markers exceeded the 300 μm threshold—unstable localization, not failed detection. In the markerless control, 6 of 270 markers (2.2%) exceeded 1 mm despite acceptable global surface fits. Conclusions: A satisfactory global surface-fit residual may conceal substantial local misregistration, whereas fiducial landmarks permit independent verification of alignment. Marker size most directly governed reproduction fidelity; ≥3 mm diameters are recommended for the rotationally symmetric geometries tested. Full article

A camera-guided warehouse robot keeps its bearings by repeatedly estimating its pose against known visual references, and it must relocalize whenever that estimate is lost. What limits this process is often not identification but the availability of a usable reference along the route. The references used in practice are usually single-sided planar fiducial markers such as QR-like codes, ArUco markers, and AprilTags, which stay readable only within a limited cone about their surface normal; a spatial reference, by contrast, can in principle be recognized from any azimuth. We quantify what that difference is worth at the navigation level. The framework is built in Unity with NavMesh navigation and a purely geometric-availability model, rather than an image-based recognizer, whose single switchable property is the availability rule. In the idealized all-azimuth spatial-reference regime (the spatial regime), a reference is available from any direction; in the single-sided, angularly constrained planar-reference regime (the planar regime) it is available only within $\pm 20°$ of the surface normal. A full-factorial experiment with 54 configurations ($3\times 3\times 2\times 3$) and $n=100$ paired replications, 10,800 runs in all, was run in both regimes over four deployment factors: deployment scheme, camera field of view, recovery step, and detection range. Under this geometric model, the spatial regime reached $5.7\times$ higher reference coverage ($41.6\%$ vs. $7.3\%$) and a mission-completion rate 30 percentage points higher (86.2% vs. 55.9%). A paired Wilcoxon signed-rank test confirms the coverage difference ($p<0.001$, matched-pairs $d_z=1.84$), and McNemar’s test together with a logistic regression confirms the completion difference. In a factorial analysis of variance, the detection range dominates (partial ${\eta }^2=0.903$), and a strong deployment × range interaction concentrates the advantage in the rack aisles, where a planar reference is seen edge-on. Three further analyses point the same way: an angular-threshold sweep from $10°$ to $60°$, an equal-count deployment control, and route- and time-normalized visibility and relocalization metrics. The advantage also held across square, L-shaped, and U-shaped aisle layouts (32,400 runs in total), with a negligible regime × layout interaction. All these numbers are model-based estimates under an explicitly stated availability model: they measure the navigation-level value of azimuthal reference availability and do not validate any particular physical object, decoding algorithm, or AR device. Full article

Background/Objectives: Magnetotactic bacteria therapy is an emerging active intratumoral drug delivery platform in which drug-loaded, magnetically responsive bacteria are injected into a tumor and navigated through its microenvironment toward defined targets using an external magnetic field, fundamentally distinct from passive intratumoral injection. No endoscopic ultrasound (EUS)- or ultrasound (US)-guided delivery workflow for MTBT has been described, and five critical technical prerequisites for clinical translation remain unaddressed: three-body image registration, minimum contrast concentration for cone-beam computed tomography (CBCT) bolus localization, needle repositioning accuracy, optimal fiducial strategy, and EUS/US procedural feasibility within the magnetic guidance apparatus. Methods: In three healthy female swine, we tested an image-guided intratumoral workflow at pancreatic and rectal sites. Gold rod fiducials were implanted under EUS in two animals and via transabdominal/endorectal US in one animal. A needle was navigated toward the implanted fiducial group using real-time gold-fiducial shine-through, and 2 mL of saline mixed with iodine contrast (Isovue-370) at 15%, 10%, or 5% v/v (one concentration per animal) was injected. Scan with CBCT was acquired before and after injection; fiducials, needle tip, and contrast bolus were segmented in 3D Slicer and their centroids compared to quantify targeting accuracy. Contrast visibility and artifact were scored by an expert radiologist on a five-point Likert scale. EUS feasibility within the magnetic apparatus was assessed using a full-scale cardboard replica of the CuraDrone PolarTrak. Results: Five percent v/v Isovue-370 was the minimum concentration for reliable CBCT bolus identification (280–360 HU; 1:1 injection-to-volume ratio). Needle repositioning accuracy was ≤2 cm in 5/6 sites using fiducial shine-through guidance. Gold rod fiducials served a dual role: CBCT tumor registration and real-time EUS/US navigation, not previously described for intratumoral injection. EUS was feasible within the PolarTrak with defined workflow adaptations. Conclusions: This pilot study demonstrates the technical feasibility of an EUS/US–CBCT workflow for MTBT, yielding quantitative, actionable parameters for first-in-human trial design. These feasibility findings, obtained in a small healthy animal cohort, are intended to inform the design of subsequent clinical studies. Full article

The accuracy of optical tracking tools is crucial for surgical navigation. While commercial tools are reliable, their proprietary design knowledge limits accessibility and adaptability for specialized clinical and research applications. This study introduces an open, reproducible optimization framework based on point-based rigid registration theory, defining a unified pose estimation deviation metric and deriving its analytical expression for both expectation and variance. The approach incorporates constraints for intra-group uniqueness and inter-group compatibility, using exhaustive configuration generation and geometric evaluation to rank designs by predicted accuracy. Numerical simulations confirmed the derived formula, with under 5% average prediction error for the expectation and strong agreement for the variance. Optimized four-fiducial tools were compared to commercial references via tip calibration, distance measurement, and registration tests. Most optimized tools (75%) achieved accuracy comparable to or modestly better than commercial tools (e.g., tip calibration 0.22 mm vs. 0.28 mm, distance measurement 0.18 mm vs. 0.20 mm, FRE 0.13 mm vs. 0.16 mm, TRE 0.48 mm vs. 0.51 mm). All four experiments showed a strong positive correlation between the theoretical metric and measured error (Pearson’s $r>0.98$, all $p<2.2\times {10}^{-7}$; exact Spearman $p\le 2.8\times {10}^{-6}$). Beyond these numerical results, the primary contribution is a systematic, open design methodology that formalizes knowledge historically proprietary to commercial vendors, enabling researchers and engineers to generate high-precision custom tracking tools for diverse surgical navigation scenarios. Full article

Systemic air embolism (SAE) is a rare but potentially catastrophic complication of computed tomography (CT)-guided percutaneous lung procedures, with reported incidence ranging from 0.02% to 0.4%. Despite its low frequency, SAE can result in severe neurological impairment or death, yet the factors that determine divergent clinical outcomes remain poorly characterized. We present two cases with contrasting neurological sequelae to elucidate the imaging spectrum and potential prognostic determinants of SAE. In Case 1, a 60-year-old man developed SAE after CT-guided fiducial marker placement for a 7-mm pure ground-glass nodule and achieved full clinical recovery. In Case 2, a 68-year-old man developed SAE following CT-guided percutaneous transthoracic needle biopsy of a 16-mm solid nodule and progressed to persistent semi-comatose status with cortical laminar necrosis. These cases illustrate the heterogeneous neurological outcomes of SAE and underscore the critical role of early post-procedure CT in detecting intravascular gas. Several factors may contribute to divergent outcomes, including procedural technique, nodule characteristics, and the extent and distribution of intracardiac and intracranial air. The absence of hyperbaric oxygen therapy in both patients further highlights the importance of prompt recognition and supportive management. Radiologists and interventionalists must maintain vigilant post-procedural surveillance to facilitate timely diagnosis and optimize patient outcomes. Full article

As thoracic surgeons more frequently address smaller lung lesions and perform lung-sparing resections, their objective is to resect an adequate specimen and margin without removing excess healthy lung tissue. Although perioperative lung nodule localization has been in practice for decades, the existing and emerging techniques used for the identification of targeted and occult lesions are more widely utilized today than they were in the past. In this review, we detail the logic behind this increase in use, classify the techniques into preoperative and intraoperative categories, and define the specific modalities available. Where applicable, we review the published data comparing techniques, detailing efficacy and safety. In the preoperative space, we describe standard computed tomography (CT)-guided localization, virtual-assisted lung mapping, electromagnetic navigation bronchoscopy, robotic-assisted bronchoscopy, and novel fiducial markers. In the intraoperative space, we describe classical localization techniques, novel applications of intraoperative cone-beam CT, and fluorescence-guided surgery and intraoperative molecular imaging (IMI). Lastly, we review emerging approaches for intraoperative molecular imaging including a report on agents in early-stage clinical trials and a brief survey of promising preclinical models. With each approach mentioned, we analyze the potential benefits and hazards, and appraise the evidence for (or against) the use of any specific modality. Full article

Split-type robots are increasingly deployed in unstructured confined environments such as underground coal mines, where autonomous navigation and cooperative tracking control remain critical challenges. This paper presents a visual target-assisted tracking control scheme for a split-type drilling robot, adopting an active leader–passive follower architecture. The leader robot performs autonomous mobility and obstacle avoidance using 3D LiDAR-based offline path generation and online optimal search. The follower robot uses AprilTag visual fiducial markers to estimate the six-degree-of-freedom relative pose via the Perspective-N-Point algorithm, and it tracks the leader using a two-dimensional fuzzy PID controller that adaptively tunes PID parameters. Extensive experiments are conducted in simulation, simulated tunnels, a large-scale robot platform, and a real drilling robot prototype. Results demonstrate that the leader achieves an average navigation error below 0.175 m, while the follower maintains an average relative tracking error within 0.06 m. The proposed method enables stable, comparable accuracy with smoother, less oscillatory response, and high-precision cooperative navigation for heavy-duty split-type robots, offering a practical solution for intelligent drilling operations in underground confined spaces. Full article

Accurate positioning is essential for inspection robots in caged chicken houses, where long straight corridors, sparse textures, and repetitive structures challenge conventional methods. This paper proposes CVIWM (Coupled Visual-Inertial-Wheel Odometry with Markers), a tightly coupled state estimation method that fuses visual, inertial measurement unit (IMU), wheel odometry (WO), and fiducial marker observations within a factor graph optimization framework. Wheel odometry preintegration suppresses IMU horizontal drift and provides absolute scale, while sparse AprilTag markers (10 m spacing) periodically reset accumulated errors. Experiments in an 80 m corridor of a commercial caged chicken house at 0.116 m/s and 0.232 m/s showed that CVIWM achieves average positioning errors of 2.402 cm and 3.253 cm. This high precision ensured reliable image acquisition (image shift $<83$ pixels), enabling 95.7% dead hen detection and 98.9% egg detection accuracy. CVIWM offers a low-cost, easy-to-deploy, high-accuracy solution for automated poultry house inspection, supporting smart livestock farming. Full article

Thin-slice high-resolution computed tomography (CT) has improved the detection of small pulmonary nodules, increasing the demand for minimally invasive diagnostic and therapeutic resection. While lobectomy with lymph node dissection remains the standard surgical approach for many patients with resectable non-small cell lung cancer, accumulating evidence supports sublobar resection for selected small, peripheral, and ground-glass-dominant lesions when sufficient margins are achievable. In thoracoscopic and robotic surgery, localization of nodules ≤10 mm or lesions located >5 mm from the pleural surface can be challenging, and failure to identify the target may lead to conversion, larger resection than intended, or prolonged operative time. Several localization strategies have been developed, including CT-guided percutaneous wire/coil/dye marking, bronchoscopic dye mapping, and virtual-assisted lung mapping (VAL-MAP), robotic-assisted bronchoscopic dye or fiducial localization, radiofrequency identification microtag systems (Surgical Real-Time FInger Navigation and Detection) that provide real-time depth information, and single-stage intraoperative CT-guided marking and resection in hybrid operating rooms. This review synthesizes representative evidence and published outcome ranges, and compares workflows, marker-to-lesion precision metrics, complication profiles, operational burden, and cost structures. We emphasize the practical contrast between two-stage and single-stage workflows, the access-route differences between transthoracic and transbronchial techniques, and the need to report localization-to-incision “time at risk”. We also present an expert-consensus decision algorithm aimed at facilitating tailored selection of localization strategies for modern minimally invasive thoracic surgery. Full article

In recent years, data-driven approaches have become increasingly important in industrial computer vision applications, particularly for 6-Degrees-of-Freedom (6-DoF) object pose estimation. However, benchmark datasets may unintentionally introduce biases that affect the reliability of learned models. In this work, we investigate the shortcut bias induced by fiducial ArUco markers in the widely used Linemod dataset. Although such markers are typically absent in real industrial environments, they introduce unintended visual cues that neural networks tend to exploit. As a result, model selection based on state-of-the-art benchmarks can be biased, since the reported performance often reflects reliance on these shortcuts rather than on robust feature extraction. Using saliency map analysis, we show that often a large portion of the model’s attention is concentrated on these markers, revealing the presence of a shortcut that artificially boosts pose estimation performance. To mitigate this issue, we propose a data augmentation pipeline based on generative AI techniques that removes the markers and replaces the background with more realistic synthesized scenes. Experimental results indicate a noticeable drop in performance when models trained on the original Linemod dataset are evaluated in ArUco-free environments, confirming the presence of background-induced biases. Training with the proposed generative-swapped dataset leads to improved robustness and better generalization to unseen scenarios, although it does not fully eliminate the problem. Overall, the results highlight the impact of background-related biases in pose estimation benchmarks and suggest that the proposed augmentation strategy represents a practical and scalable step toward developing more reliable 6-DoF pose estimation systems for industrial applications, while leaving room for further improvements. Full article

The sharp dose gradients that underpin the dosimetric advantage of particle therapy over photon therapy can be undermined by the interplay effects due to intra-fraction motion in modern pencil beam scanning systems. Fluoroscopy-Guided Particle Therapy (FGPT) offers a promising path to improved motion management through real-time tracking of tumors or surrogate signals. The advent of flat-panel detector (FPD)-based technology has enabled tighter integration of fluoroscopy/fluorography into treatment units and accelerated clinical adoption and research, with commercial systems such as Hitachi’s Real-time Gated Particle Therapy (RGPT) now available. However, the need for implanted fiducial markers, with the associated invasiveness and risk of complications, limits the utility of RGPT to a few anatomic sites in selected patients. The full potential of FGPT, therefore, depends on reliable marker-less tumor tracking, which remains challenging because soft-tissue targets are obscured by overlapping anatomy along the X-ray path, leading to reduced reliability of traditional image-registration algorithms in the projection domain. Recent advances in deep learning and AI-driven image registration have renewed hope for overcoming these barriers, enabling real-time marker-less tracking for particle therapy. This review outlines the evolution of fluoroscopy technology from image intensifier (II) to FPD-based systems, summarizes historical and recent vendor-supported FGPT strategies, and surveys emerging AI-based algorithms in the literature. A general review of machine learning-based image registration is provided, challenges in generalizability and interpretability are highlighted, and potential paths toward reliable, clinically deployable FGPT are discussed. Full article