The clinical stakes could hardly be higher. Widespread low-dose CT screening has dramatically increased the detection of peripheral pulmonary lesions, yet confirming whether a nodule is malignant remains one of the most stubborn challenges in respiratory medicine. These lesions sit in the fourth-generation airways or beyond, deep in the lung periphery where a standard bronchoscope cannot simply reach and see. Transbronchial biopsy guided by navigation technology has emerged as a minimally invasive alternative to surgical biopsy or transthoracic needle puncture, but the field has been muddled by retrospective single-center studies, inconsistent definitions of diagnostic yield, and varying adjunctive tools, leaving clinicians without a reliable map of which technology suits which patient.

To cut through that confusion, the research team designed a parallel-controlled trial with a clever allocation scheme. Rather than randomizing in the traditional sense, patients were assigned in a predefined cyclic sequence of robotic, electromagnetic, and fluoroscopy bronchoscopy, repeated until each group reached 90 participants. The sequence was fixed before any procedure and could not be adjusted based on operator preference, lesion size, density, location, distance to the pleura, or suspected malignancy. All procedures were performed under general anesthesia by senior pulmonologists with more than five years of bronchoscopy experience, and robotic operators completed standardized simulation training plus at least 24 clinical procedures before enrolling patients, ensuring everyone operated past the initial learning curve.

One of the trial’s methodological strengths lies in its rigorous outcome definitions. The investigators distinguished between a standard overall diagnostic yield, which counts any diagnosis consistent with the final clinical assessment, and a stricter yield that only accepts malignant or specific benign pathology from the index procedure. Under the strict definition, the gap between technologies widened considerably: robotic bronchoscopy reached 81.1 percent, electromagnetic navigation 61.1 percent, and fluoroscopy just 47.8 percent. Sensitivity for malignancy followed the same gradient, at 0.89, 0.82, and 0.62 respectively, while specificity and positive predictive value were a perfect 1.0 across all three groups. Nondiagnostic cases were followed for six months and confirmed through surgery, reintervention, imaging changes, or multidisciplinary consultation, guarding against misclassification.

The subgroup analyses are where the trial becomes genuinely practice-changing. For lesions measuring between 15 and 30 millimeters, robotic bronchoscopy achieved a diagnostic yield of 90.6 percent, significantly higher than both competitors. For lesions in the peripheral third of the lung, the robotic system and electromagnetic navigation performed nearly identically at roughly 88 percent, while fluoroscopy lagged at 58.8 percent. Among lesions showing a concentric view on radial endobronchial ultrasound, the robotic system reached 90 percent versus 64.6 percent for fluoroscopy. Notably, for lesions smaller than 15 millimeters or larger than 30 millimeters, the three techniques performed comparably, suggesting that the robotic advantage is concentrated in the mid-sized, anatomically demanding nodules that clinicians find most vexing.

Why does the robot win in these challenging cases? The trial’s trajectory analysis offers a mechanical explanation. The robotic system demonstrated greater navigation and operational stability than electromagnetic navigation, with lower mean displacement from the planned pathway and a higher percentage of trajectories staying within five millimeters of the intended route, a difference that reached statistical significance. This matters because CT-to-body divergence, respiratory motion, airway deformation, and tool displacement can all sabotage sampling even when the virtual route looks perfect on the planning screen. By mechanically stabilizing the catheter and preserving alignment between the planned bronchial path and the actual trajectory, the robot minimizes the drift that accumulates as instruments traverse five or six generations of branching airways.

Perhaps the most forward-looking element of the study is its fusion of radiomics with artificial intelligence. The team extracted 111 quantitative imaging features from thin-slice chest CT scans, capturing texture heterogeneity, density dispersion, and three-dimensional morphology that the human eye cannot reliably grade. Features such as entropy, which quantifies the randomness of gray-level distribution, and regional variance, which reflects local heterogeneity in CT attenuation, turned out to be powerful predictors of which technology would succeed for a given lesion. Five machine learning algorithms were trained for each technique, and random forest models emerged as the champions, achieving an area under the curve of 0.984 for robotic bronchoscopy, 0.982 for electromagnetic navigation, and 0.974 for fluoroscopy, with accuracy approaching 98 percent for the two advanced platforms.

SHAP interpretability analysis revealed which features drove each model’s predictions, and the rankings differ revealingly by technique. For robotic bronchoscopy, regional variance, entropy, and clustering shadow topped the list, hinting that the robot excels when internal lesion complexity would otherwise destabilize sampling. For electromagnetic navigation, volume proportion, maximum diameter, and entropy led, while for fluoroscopy the dominant features were CT value variance, airway generation number, and lesion mass, reflecting that conventional guidance depends heavily on physical accessibility. The best models were then integrated into a free web-based platform that requires only nonidentifiable lesion-level and radiomic features, letting clinicians input preoperative imaging characteristics and receive technique-specific estimates of diagnostic success before ever entering the procedure room.

The authors are careful to frame their results not as a universal crowning of the robot but as an argument for lesion-stratified strategy. Electromagnetic navigation matched robotic performance in malignant, solid, and larger lesions, and fluoroscopy, despite its two-dimensional imaging and radiation exposure, remains a feasible option for selected simpler lesions in settings where advanced systems are unavailable. The researchers acknowledge limitations, including the nonrandomized allocation, the absence of external validation for the machine learning models, and a six-month follow-up that may be too short for indolent ground-glass lesions. They also note that virtual bronchoscopic navigation and cone-beam CT were not evaluated and could be integrated into future frameworks.

Still, the trial marks a turning point in how the field thinks about diagnosing lung cancer’s earliest peripheral signatures. Instead of asking which platform is best in the abstract, clinicians can now ask which platform is best for this nodule, in this location, with this texture, in this hospital. By pairing a rigorous multicenter comparison with an interpretable AI decision-support tool, the study sketches a future in which the diagnostic pathway for suspected peripheral lung cancer is personalized from the first CT scan, reducing repeat invasive procedures, accelerating treatment decisions, and ultimately giving patients with early-stage disease a faster route to curative care.

Subject of Research: Comparative diagnostic performance of robotic, electromagnetic navigation, and fluoroscopy bronchoscopy for peripheral pulmonary lesions

Article Title: Comparative diagnostic performance of robotic, electromagnetic navigation, and fluoroscopy bronchoscopy for lung cancer in peripheral pulmonary lesions: a prospective, multicenter, parallel-controlled trial

Article References: Zhong, C., Huang, J., Xu, L., You, Z., Wang, F., Sun, J., Jiang, J., Liu, D., Huang, J., Zhang, H., Li, H., Li, Z., He, W., Lin, Z., Zheng, X., He, L., Liu, J., Chen, D., Wang, G., & Li, S. (2026). Comparative diagnostic performance of robotic, electromagnetic navigation, and fluoroscopy bronchoscopy for lung cancer in peripheral pulmonary lesions: a prospective, multicenter, parallel-controlled trial. Journal of Advanced Research. https://doi.org/10.1016/j.jare.2026.09.008

Image Credits: AI Generated

DOI: 10.1016/j.jare.2026.09.008

Keywords: robotic bronchoscopy, electromagnetic navigation bronchoscopy, fluoroscopy, peripheral pulmonary lesions, lung cancer, diagnostic yield, radiomics, machine learning, bronchial biopsy, clinical trial, artificial intelligence, pulmonary nodules

The technology at the heart of the study, known as needle virtual reconstruction or NVR, addresses a fundamental bottleneck in image-guided interventions. Percutaneous procedures, in which needles are advanced through the skin to reach deep targets such as tumors or bone lesions, are commonly performed under cone-beam CT guidance. Cone-beam CT holds significant advantages over conventional CT in the interventional suite: it avoids the constraints of bore size and gantry orientation, and it integrates seamlessly with real-time fluoroscopy, allowing clinicians to visualize instruments continuously during an intervention. But there is a catch. Most commercially available guidance systems allow operators to plan a needle trajectory on an initial CBCT acquisition and monitor progress using fluoroscopy overlays, yet evaluating the needle’s latest position within the three-dimensional volume requires acquiring an entirely new CBCT scan. Each additional acquisition adds seconds to minutes of procedure time and delivers another dose of ionizing radiation to both patient and operator.

NVR software, marketed as Needle ASSIST with Stereo 3D by GE HealthCare, sidesteps this problem by mathematical triangulation. Instead of a full volumetric rotation, the system needs only two fluoroscopic projections captured with the needle in place. Because the needle’s silhouette appears from two different angles, and because the geometry of the fluoroscopic gantry is precisely known, the software can reconstruct the needle’s three-dimensional position and map it into the coordinate space of the initial planning CBCT. The virtual needle then appears alongside the planned trajectory, giving the operator an immediate, radiation-light assessment of whether the instrument is on course, short of target, or straying toward danger. The workflow relies on two automatically generated views: a so-called bull’s eye view that looks directly down the barrel of the needle to guide entry point and orientation, and a progress view that displays the full needle shaft to track insertion depth.

To validate this approach rigorously, the research team conducted a retrospective single-center cohort study at a large cancer hospital, with institutional review board approval under protocol 16-402. They enrolled consecutive patients who underwent percutaneous image-guided interventions using NVR software between January 2023 and November 2024. The final cohort comprised 40 procedures in 40 cancer patients: 37 biopsies and 3 kyphoplasties, the latter being vertebral augmentation procedures in which cement is delivered into collapsed vertebrae. All spinal interventions followed a transpedicular approach, threading the needle through the bony pedicle of the vertebra, while other procedures used anatomically safe paths selected to avoid critical structures. Every case followed societal guidelines for anticoagulation management and bleeding risk, and procedures were performed under monitored anesthesia care or general anesthesia by an interventional radiologist with more than two decades of experience.

The imaging protocol followed a standardized sequence. Before any needle was inserted, an initial cone-beam CT was acquired using a 200-degree rotational scan at 40 degrees per second over roughly five seconds, capturing between 147 and 244 frames depending on the suite. Variable kVp and mAs settings dynamically compensated for differences in patient anatomy, and the reconstructed three-dimensional volume spanned a 24-centimeter-diameter cylinder in a 512-cubed matrix. The trajectory was planned tableside by defining target and entry points, and the planned line was overlaid on live fluoroscopy. Bone anatomy derived from the initial CBCT was superimposed to confirm accurate registration and flag any patient movement, with tableside adjustments made as needed. After needle insertion, the virtual reconstruction was generated automatically from two fluoroscopic projections, and critically, a final CBCT was acquired as ground truth to validate needle position before the intervention proceeded. It is this final scan that allowed the team to measure, in hindsight, how well the virtual needle matched reality.

The validation methodology was deliberately conservative. The researchers registered the pre-insertion planning CBCT with the final ground-truth CBCT based on bony landmarks, without displaying the virtual reconstruction. They then measured the maximal distance between the virtual needle tip and the actual needle tip on the bull’s eye view using a dedicated imaging workstation. Two authors performed the measurements independently and reached consensus, with a third author arbitrating disagreements. In parallel, they assessed safety agreement: reviewers first confirmed on the final CBCT that all real needle trajectories avoided critical structures, then checked whether the software’s virtual predictions had correctly indicated those same safe paths. The results were unambiguous. The median discrepancy between virtual and actual needle tip positions was 1.3 millimeters, with an interquartile range of 0.9 to 2.2 millimeters, and the virtual reconstructions correctly predicted the safety profile in every case, with no erroneous depiction of a needle touching a critical structure.

Procedural metrics from the study paint a picture of efficient, low-dose practice. The median procedure duration was 57.5 minutes, median fluoroscopy time was 128 seconds, and the median total dose area product, combining fluoroscopy and CBCT, was 28.9 Gy·cm². Technical success, defined as placement of the needle tip within the boundaries of the targeted lesion or bone on the final CBCT, was achieved in 100 percent of cases. For the applicable tumor cases, the median target lesion size was 24 millimeters. The cohort itself skewed older and heavier, with a median age of 71 years and a median body mass index of 26.1, a detail that matters because larger patients typically require higher radiation doses for adequate image quality. Despite this, the reported doses remained within published reference levels for comparable procedures, suggesting that the guidance software did not inflate radiation burden even in technically challenging anatomy.

The broader implications reach beyond the operating suite. Previous research has suggested that virtual needle guidance technologies can meaningfully cut radiation exposure by eliminating the multiple verification CBCT scans traditionally interspersed through needle positioning. One earlier study cited by the authors reported reductions in air kerma and dose area product of 27 percent and 35 percent respectively when such technology replaced repeated volumetric checks. The accuracy figures now reported go further, raising the possibility that even the final confirmation CBCT, acquired after every needle placement in this study, might eventually be dispensable for select cases, with two quick fluoroscopic views providing sufficient reassurance more simply and faster. The authors are careful on this point: eliminating the final CBCT would require further evaluation, including a randomized or controlled comparison, before it could become standard practice. Guidance software of this kind has already found roles in musculoskeletal intervention, endoleak treatment, and thermal ablation, but the new study is distinctive in specifically quantifying reconstruction accuracy against a ground-truth scan.

The study is not without caveats, and the authors lay them out candidly. It was a single-center retrospective analysis spanning a variety of anatomical targets, so multi-institutional validation would strengthen the findings. The absence of a control group means the team cannot definitively attribute reductions in radiation dose or procedure time to the software itself. Accuracy was measured in a two-dimensional plane on the bull’s eye view rather than in full three dimensions, which would offer a more complete picture of reconstruction error. And the operating physician’s substantial experience, more than twenty years of practice, may have influenced fluoroscopy times and radiation doses in ways that might not generalize to less seasoned operators. There is also a commercial relationship to note: two of the authors serve as consultants for GE HealthCare, the manufacturer of the software and imaging systems used in the study, and the research received support in part through a National Institutes of Health cancer center support grant.

Even with those limitations, the study marks a persuasive step toward smarter, leaner image-guided surgery. The vision it sketches is an interventional suite in which a single volumetric scan at the start of a procedure anchors all subsequent navigation, and every subsequent question about needle position is answered with two milliseconds-fast fluoroscopic frames rather than a full rotation of the C-arm. For cancer patients, who often undergo repeated biopsies and ablative procedures over the course of their illness, the cumulative savings in radiation, anesthesia time, and waiting could be substantial. For the clinicians, a trustworthy virtual needle means fewer interruptions, faster workflows, and continuous three-dimensional awareness of where their instrument truly sits relative to vessels, nerves, and bone. If larger controlled studies confirm these results, the humble biopsy needle may soon travel through the body shadowed by a digital twin that is accurate to little more than a millimeter, and the era of confirming positions with repeated scans may give way to one of calculating them.

Subject of Research: Accuracy and safety of virtual needle reconstruction software during cone-beam CT-guided percutaneous procedures in cancer patients

Article Title: Virtual reconstruction to assess needle position during percutaneous procedures performed under cone-beam computed tomography: safety and accuracy

Article References: Geevarghese, R., Kiely, L., Petre, E. N., Solomon, S. B., & Cornelis, F. H. (2026). Virtual reconstruction to assess needle position during percutaneous procedures performed under cone-beam computed tomography: safety and accuracy. CVIR Oncology, 2(1), Article 9. https://doi.org/10.1007/s44343-026-00042-6

Image Credits: AI Generated

DOI: 10.1007/s44343-026-00042-6

Keywords: cone-beam CT, needle virtual reconstruction, interventional radiology, percutaneous biopsy, image-guided procedures, fluoroscopy, radiation dose, cancer, technical success, kyphoplasty, needle guidance, medical imaging

The review team searched three major biomedical databases, SCOPUS, PubMed, and Web of Science, and identified 129 records for consideration. After screening, 41 studies met the criteria for inclusion and provided usable data on disease type, imaging modalities, and reported radiation dose metrics. The populations covered were diverse: children with congenital heart disease, scoliosis, cystic fibrosis, inflammatory bowel disease, esophageal atresia, osteogenesis imperfecta, spina bifida, hydrocephalus, urological conditions, bleeding disorders, craniosynostosis, cleft palate, asthma, pulmonary hypertension, and organ transplant recipients, among others. This breadth is important, because the authors found that radiation exposure was not uniform across conditions but was strongly influenced by the specific diagnosis, the age at which imaging began, the severity of the disease, and the imaging modalities that each disease trajectory demands.

One of the clearest technical findings of the review is that the modality mix matters enormously. Plain radiography, the conventional X-ray, was by far the most frequently performed examination across nearly all chronic disease groups. However, radiographs deliver relatively small doses per examination, and the review concluded that they were not the dominant contributors to cumulative burden. Instead, computed tomography and fluoroscopy, both of which involve substantially higher dose outputs and, in the case of fluoroscopy, prolonged real-time exposure, accounted for the majority of the cumulative effective dose reported in the included studies. In some patient groups, cumulative doses from these higher-yield modalities exceeded 20 to 50 millisieverts, thresholds that radiation protection specialists regard as significant when accumulated during childhood.

The biological rationale for concern lies in the interaction between ionizing radiation and growing tissue. Effective dose, measured in millisieverts, is a calculated quantity that weights absorbed dose by the radiation sensitivity of the organs exposed, allowing comparison across different types of examinations. Pediatric patients are more sensitive to radiation-induced carcinogenesis than adults for several reasons: their tissues are actively dividing, their longer life expectancy leaves more time for radiation-induced cancers to manifest, and stochastic effects, meaning probabilistic DNA damage that may lead to malignancy decades later, do not have a known safe threshold in the linear no-threshold framework commonly used for protection purposes. A child with a chronic disease diagnosed in infancy may therefore face decades of potential risk following exposures delivered in the first years of life.

Several disease-specific patterns emerged from the included literature. Children with congenital heart disease, particularly those requiring staged surgical palliation or interventional cardiac catheterization, consistently appeared among the most heavily exposed groups, because cardiac fluoroscopy and CT angiography are central to both diagnosis and treatment. Studies cited in the review estimated cumulative doses during staged single-ventricle palliation and documented measurable chromosomal DNA damage in exposed children. In scoliosis management, repeated spinal radiographs required for curve monitoring, combined with intraoperative imaging, produced substantial cumulative exposure, prompting the development of low-dose slot-scanning systems that reduce dose compared with standard radiographs. Children with inflammatory bowel disease frequently underwent CT during acute flare-ups before magnetic resonance enterography became the preferred alternative, and retrospective cohorts documented cumulative doses high enough to raise malignancy concerns.

Other chronic conditions illustrated subtler but still meaningful exposure pathways. Infants with esophageal atresia undergo repeated contrast studies and fluoroscopic procedures in the first months of life, and one French study cited in the review explicitly asked how low these cumulative doses could realistically be pushed. Children with spina bifida and shunt-treated hydrocephalus accumulated exposure through serial imaging of the brain and spine, while pediatric stone disease generated exposure through fluoroscopy-guided procedures such as percutaneous nephrolithotomy and shockwave lithotripsy. Even conditions considered lower risk, such as developmental dysplasia of the hip, appeared in the literature, with one study reassuringly concluding that repeated pelvic radiographs during harness treatment carry very low radiation risk. Meanwhile, pediatric cleft palate patients showed a three- to five-fold increase in cumulative radiation exposure from dental radiology compared with age- and gender-matched peers.

Organ transplant recipients represent another group highlighted by the review. Children receiving heart transplants accumulated considerable exposure within the first post-transplant year through echocardiography-adjacent imaging, catheterization, and CT surveillance for complications such as rejection, infection, and vascular stenosis. A cohort study of pediatric transplant recipients more broadly documented diagnostic imaging exposure that was markedly elevated compared with healthy children. Similarly, children with osteogenesis imperfecta, the brittle bone disorder, required serial skeletal radiographs throughout childhood to monitor fractures and surgical interventions, with one cited study estimating associated lifetime cancer risk from these cumulative exposures.

What emerges from the synthesis is not a reason for alarm or for avoiding medically necessary imaging, the authors emphasize, but rather a roadmap for safer practice. The review calls for evidence-based referral guidelines specific to pediatric chronic disease populations, so that clinicians weigh the diagnostic yield of each examination against its dose contribution within the context of a child’s total imaging history. It also highlights the importance of standardized imaging protocols optimized for children, including weight- and age-based parameter adjustment, substitution of ultrasound or magnetic resonance imaging where diagnostically equivalent, and the use of dose modulation technologies in CT. Equally critical is dose reporting: recording cumulative effective dose in the patient record so that ordering physicians can see the full picture rather than evaluating each request in isolation. Principles such as ALARA, keeping exposure as low as reasonably achievable, and its extensions emphasizing appropriate use and avoiding unnecessary procedures, are framed as essential operational standards rather than abstract ideals.

The authors also point toward the practical infrastructure needed to make dose stewardship routine. Electronic health record integration of dose-tracking systems, standardized dose metrics across institutions, and education of referring clinicians about the relative doses of different modalities all feature in the review’s recommendations. International collaborative efforts, such as the HARMONIC project cohort studies on radiation exposure in children with congenital heart disease cited within the review, exemplify the kind of multinational, disease-stratified data collection the field needs to quantify risk precisely and track the impact of protection measures over time. The review itself was a literature-based analysis, so no individual patient data were collected, and no ethics approval was required.

For families, the message is one of partnership rather than fear. Parents of children with chronic diseases can and should ask whether each proposed imaging examination is necessary, whether a lower-dose alternative is available, and whether the child’s cumulative imaging history has been considered. For the medical community, the review consolidates more than a decade of evidence into a single argument: the child with a chronic disease is not a series of isolated imaging encounters but a single, longitudinally exposed patient whose total radiation burden deserves active management. As imaging technology continues to advance and dose reduction becomes increasingly feasible, the findings serve as both a benchmark of current exposure levels and a challenge to ensure that the children who depend most on medical imaging are also the best protected from its long-term consequences.

Subject of Research: Cumulative ionizing radiation exposure from medical imaging in pediatric patients with chronic diseases

Article Title: Cumulative ionizing radiation exposure in pediatric patients with chronic diseases: a narrative review

Article References: Eimane, A., Francavilla, M., Grigorjevs, A., Granata, C., Limantoro, I., Olteanu, B.-S., Sofia, C., Nievelstein, R. A., Kardos, M., Kasznia-Brown, J., Salerno, S., & Apine, I. (2026). Cumulative ionizing radiation exposure in pediatric patients with chronic diseases: a narrative review. Pediatric Radiology. https://doi.org/10.1007/s00247-026-06785-x

Image Credits: AI Generated

DOI: 10.1007/s00247-026-06785-x

Keywords: pediatric radiology, cumulative radiation dose, ionizing radiation, CT, fluoroscopy, radiation protection, chronic disease, effective dose, congenital heart disease, scoliosis, medical imaging, ALARA