Lung needle biopsies are among the most common and consequential procedures in interventional radiology. When a radiologist spots a suspicious nodule on a chest scan, the definitive diagnosis usually requires extracting a sliver of tissue through the chest wall, guided by computed tomography. The challenge is geometric as much as medical. The needle must reach the lesion while dodging ribs, avoiding the mediastinum and its great vessels, steering clear of pulmonary fissures, and traversing as little healthy lung parenchyma as possible. Every extra millimeter of lung crossed raises the risk of pneumothorax, the collapsed lung that occurs when air leaks through a puncture, as well as bleeding and hemoptysis. Yet despite decades of advances in imaging, the choice of trajectory still varies considerably from one radiologist to the next, a variability that AI and robotics have long been expected to tame.

The new study, led by Taylor Hoffman and Francois H. Cornelis with colleagues including Stephen B. Solomon and Debkumar Sarkar, set out to test whether a general-purpose generative AI model could shoulder part of that planning burden. The team conducted a retrospective, institutional review board-approved analysis of de-identified CT images from 30 consecutive lung biopsies performed by three interventional radiologists, each with at least a decade of experience. The cohort included 17 men and 13 women with a median age of 65 years, and the lesions ranged from 5 to 45 millimeters across, deliberately spanning technically difficult territory: five apical lesions at the top of the lung, seven subpleural nodules hugging the chest wall, three para-mediastinal lesions, and four near the lung base.

The technical setup was elegantly simple. Each axial CT image, acquired on a GE Healthcare scanner and exported in DICOM format, was cropped and magnified but not rescaled. Using the open-source image editor GIMP, the team overlaid a radial grid on the image and converted it to PNG. One author annotated the target lesion with a blue circle and outlined the sensitive structures in red, including the mediastinum and major vessels. The annotated image was then uploaded to ChatGPT-4 with a standardized prompt specifying five key parameters for an optimal entry angle: reaching the target, minimizing the distance traveled through lung tissue, and avoiding bones, fissures, and the mediastinum. The model was asked to read the radial grid and return the angle at which the needle should enter.

Before running the full analysis, the researchers validated the approach by comparing a pre-biopsy planning CT image with a procedural CT image captured during the actual biopsy, acquired at the same anatomical level and showing the needle, skin, and target in place. A separate validation dataset of 30 paired planning and procedural images was then used for data collection. Each case was processed independently in a fresh chat session, with iterations allowed whenever the model failed to incorporate all criteria. The median number of iterations needed to arrive at a complete answer was just two, with an interquartile range of one to three, suggesting the model could converge on a usable plan with minimal back-and-forth.

The results were striking. Independent reviewers, two experienced interventional radiologists who had not performed the original procedures and who assessed the AI-generated paths in blinded fashion, judged 28 of 30 paths, or 93.3 percent, safe and adequate for performing the biopsy. The two rejected paths were deemed theoretically feasible but not representative of best clinical practice: in one case the AI path crossed 20 millimeters of lung tissue, four times longer than the manual route, and in another it unnecessarily punctured lung parenchyma to reach a subpleural nodule, raising concerns about needle stability. More remarkably, the AI paths crossed a median of just 5 millimeters of lung tissue, compared with 23 millimeters for the paths the human radiologists actually chose, a difference that reached statistical significance with a p-value of 0.0063 on a Wilcoxon signed-rank test. Paths with zero lung traversal corresponded exclusively to subpleural nodules, where a direct pleural puncture enters the lesion immediately.

On the question of whether the AI simply rediscovered what the humans had already chosen, the answer was nuanced. The mean entry angle suggested by the AI was 179 degrees, against 174 degrees for the manual procedures, a difference that was not statistically significant. Using a definition of path match requiring entry angles within 10 degrees of each other and similar trajectories through tissue planes, 8 of 30 cases, or 26.7 percent, showed good concordance. The authors note that this match rate is lower than in earlier work, such as a study by Too and colleagues that used a three-dimensional convolutional neural network for lesion detection and path planning and reported 82 percent concordance with an average angular deviation of 2.30 degrees, alongside 93.5 percent sensitivity and 93.2 percent specificity in lesion detection. That earlier system, however, was purpose-built for the task rather than adapted from a general-purpose chatbot, and it defined path match more narrowly, within 5 degrees.

Why should a shorter path matter? The authors argue that reduced traversal through lung tissue could lessen tissue manipulation and lower the risk of pneumothorax and hemoptysis, potentially shortening procedure time, improving patient comfort, reducing anesthesia use, and increasing operating room efficiency. In the 30 procedures studied, no major complications occurred, and minor events were self-limiting: minimal pneumothorax in four cases, 13.3 percent, which resolved spontaneously, and mild hemoptysis in two cases, 6.7 percent. But the authors also flag a clinical caveat. With a coaxial technique, in which an outer introducer needle anchors the pathway for repeated sampling, a very short path through lung tissue might provide insufficient purchase, increasing the risk that the needle dislodges during the procedure. Shorter, in other words, is not automatically better in every scenario.

The study is candid about its limits. With only 30 cases and a retrospective design, generalizability remains unproven. The analysis was performed in two dimensions on single axial slices rather than in three dimensions, a simplification that may underestimate the true spatial complexity of biopsy planning. The prompt did not require the AI to maintain a standard 5-millimeter safety margin from the subclavian and axillary neurovascular bundles, though those structures were kept out of the imaging field because patients were scanned with their arms raised. Iterative follow-up prompts were not scripted, which limits reproducibility, and the study measured technical feasibility only, not clinical outcomes such as diagnostic yield, complication rates, or procedure time. The authors also caution that GPT-4 itself may soon be superseded, arguing that the field should move toward specialized generative models trained specifically on interventional radiology rather than adapting general-purpose systems.

Even so, the trajectory of the technology is hard to ignore. The authors envision generative AI integrated with robotic biopsy platforms to automate trajectory planning, real-time procedural guidance in which the model continuously analyzes intraoperative imaging and suggests adjustments as the patient moves or breathes, pre-procedural virtual rehearsals that let clinicians simulate multiple approaches before committing to one, and eventual three-dimensional reconstruction that would eliminate manual image annotation altogether. Combined with augmented reality, AI-recommended paths could be projected directly into the clinician’s field of view. The same logic extends beyond the lung to tumor ablation, where precise positioning determines whether the entire tumor is covered, and to vascular interventions, where AI could recommend optimal access routes based on patient-specific anatomy. For now, the message of this small feasibility study is measured but provocative: a model built to converse with humans can, with a radial grid and a well-crafted prompt, plan a needle path through the lung that is measurably less invasive than the one a seasoned specialist picked. The needle, for the moment, remains in human hands, but the map may increasingly be drawn by machines.

Subject of Research: Feasibility of using a multimodal large language model to plan entry angles for CT-guided lung needle biopsies

Article Title: Feasibility study on using multimodal large language model for CT-guided lung biopsy trajectory planning

Article References: Hoffman, T., Worthington, M., Sarkar, D., Solomon, S. B., & Cornelis, F. H. (2025). Feasibility study on using multimodal large language model for CT-guided lung biopsy trajectory planning. CVIR Oncology, 1(1), Article 5. https://doi.org/10.1007/s44343-025-00005-3

Image Credits: AI Generated

DOI: 10.1007/s44343-025-00005-3

Keywords: generative AI, ChatGPT-4, lung biopsy, CT-guided biopsy, trajectory planning, interventional radiology, pneumothorax, needle path planning, multimodal large language model, Memorial Sloan Kettering, CVIR Oncology, medical imaging

The research, conducted at Fondazione IRCCS Istituto Nazionale dei Tumori in Milan, followed 1,059 consecutive patients, both adults and children, who were referred for biopsy between 2010 and 2022 with clinical suspicion of a lymphoproliferative disease, an umbrella term covering lymphomas and related disorders of the lymphoid tissues. Every one of these patients underwent imaging-guided core needle biopsy as their first diagnostic procedure, and the results, published in CVIR Oncology, are striking. In 96.6 percent of all referred patients, the needle biopsy alone delivered a conclusive histological diagnosis without any need for follow-up surgery. Among the 882 patients ultimately confirmed to have lymphoma, 95.9 percent received a complete, actionable diagnosis from the needle procedure alone.

Those numbers matter because of what they replace. Traditional excisional biopsy requires surgical pre-assessment, anesthesia, an incision, and recovery time, and carries risks of infection, scarring, and bleeding. It can also introduce delays at exactly the moment when speed matters most, since different lymphoma subtypes demand radically different treatment strategies, and treatment often cannot begin until the subtype is pinned down. Core needle biopsy, by contrast, is a minimally invasive procedure performed through a tiny puncture, typically under local anesthesia, in an outpatient or short-stay setting. For frail patients, for those needing urgent therapy, and for anyone with a deep-seated lesion that surgery would struggle to reach safely, the difference is not merely one of convenience but of clinical feasibility.

The technical backbone of the procedure is the semiautomated Tru-cut needle, a spring-loaded device available in gauges ranging from 14 to 18 Gauge in routine practice, with 16 Gauge being the workhorse in the Milan series. The operator pulls back a mandrel to load the spring, advances the needle into the target tissue while continuously visualizing the needle tip, and then fires the outer cutting cannula over the stylet, capturing a slender cylindrical core of tissue. At this institution, at least three or four passes through the lymph node or lesion are typically made, and the median number of samples in the recorded data was four. The cores are fixed in formalin, embedded in paraffin, stained with hematoxylin and eosin, and then classified according to the World Health Organization criteria using morphology, immunohistochemistry, and, when needed, FISH and molecular testing.

Equally important is how the target is chosen and how the needle finds its way there. Ultrasound guided the large majority of the procedures, 891 of the 1,059, and remains the technique of choice for superficial lymph nodes, soft tissues, salivary glands, liver, and kidney, thanks to its real-time visualization, low cost, and wide availability, often complemented by color Doppler mapping of the lesion’s vascular structure. Computed tomography took over for deep anatomical sites, accounting for 168 procedures, particularly retroperitoneal and mediastinal lesions nestled near vessels and organs. For these, the team used a coaxial technique, placing a larger introducer through which the biopsy needle could be advanced repeatedly, allowing multiple samples from a single puncture while reducing the risks of bleeding and tumor seeding. Contrast-enhanced CT scans helped identify viable tissue and avoid sampling necrotic or liquefied areas, and the technique proved its worth: CT-guided biopsies actually achieved a slightly higher diagnostic yield, 97.1 percent, than ultrasound-guided ones, at 95.8 percent.

A particularly sophisticated element of the workflow was the use of positron emission tomography, or PET-CT, for target selection. In 42.5 percent of lymphoma cases, a PET-CT scan performed within 30 days before the biopsy was available to guide the choice of where to sample. Because fluorodeoxyglucose uptake correlates with the histological grade and proliferative activity of lymphoma cells, hot spots on a PET image mark the metabolically active, viable regions of a tumor, allowing operators to aim the needle at the most informative tissue and steer clear of dead or fibrotic zones. The study found that PET-CT availability did not measurably raise the overall binary diagnostic yield, but the authors argue that its true value is qualitative: distinguishing active disease from post-treatment fibrosis, and selecting the clinically most informative target in cases of suspected relapse or transformation. In one illustrative case, a retroperitoneal mass showed poor enhancement on contrast CT, yet PET imaging revealed intense uptake in the same region, and the needle was placed precisely where the PET signal was strongest, yielding a conclusive diagnosis that conventional imaging alone might have missed.

Safety was another standout finding. Not a single patient in the entire cohort suffered a biopsy-related complication of grade 2 or higher on the standard CTCAE v5 scale, meaning no severe bleeding, infection, or event requiring unplanned intervention. Minor local bleeding occurred in roughly 5 percent of cases, and no patient required hospitalization beyond the planned 24 hours after deep-tissue biopsies. Standard precautions, including blood counts and coagulation testing, stopping anticoagulation at least 24 hours before the procedure, and platelet transfusion for patients with counts below 50,000 per microliter, kept the risk profile firmly in check. Even the 22 pediatric patients in the cohort were managed successfully, underscoring that minimally invasive image-guided biopsy can be appropriate even in children when multidisciplinary teams weigh anesthetic risk, lesion location, and protocol requirements.

The study is honest about where the needle falls short. Thirty-six lymphoma patients, 4.1 percent of the cohort, still required a subsequent excisional biopsy, either because the needle sample was inadequate or because it was suggestive but not conclusive. Notably, certain subtypes were overrepresented in this surgical group: half of the patients with nodular lymphocyte-predominant Hodgkin lymphoma and 15 percent of those with peripheral T-cell lymphoma needed surgery to reach a final diagnosis, a pattern consistent with pathological guidance suggesting that these entities are best assessed when an entire lymph node’s architecture can be examined. Among the 36 surgical cases, the excisional diagnosis differed from the needle-based impression in 9 patients, confirmed it in 12, and provided a new conclusive answer in 15, all of whom ended up with a definitive histological result. The authors stress that these figures argue not against needle biopsy, but for prompt surgical follow-up whenever the needle result does not inspire confidence for treatment decisions, ideally after early multidisciplinary discussion among hematologists, radiologists, and pathologists.

The broader picture emerging from more than a decade of data is one of a diagnostic landscape in transition. The Lugano classification, the standard framework for lymphoma management, still reserves core needle biopsy for situations where surgery is not possible or for relapse. Yet this Milanese experience, one of the largest single-center intent-to-diagnose cohorts ever assembled, shows that with careful patient selection, meticulous imaging guidance, judicious needle gauge choice tailored to lesion location, multiple sampling, and PET-informed targeting, the needle succeeds where the scalpel was once thought indispensable. Follicular lymphoma, diffuse large B-cell lymphoma, and classical Hodgkin lymphoma, the three most common diagnoses in the cohort, were reliably captured by needle sampling, and 177 patients with suspected lymphoma actually turned out to have entirely different conditions, including metastatic carcinoma, sarcoma, neurogenic tumors, and retroperitoneal fibrosis, all of which the needle correctly identified. In an era when medicine increasingly prizes precision with minimal invasiveness, this study makes a compelling case that for most patients with suspected lymphoma, the first and often only diagnostic procedure needed is one that fits through a needle’s eye.

Subject of Research: Diagnostic yield and procedural techniques of imaging-guided core needle biopsy in patients with suspected lymphoproliferative disease

Article Title: Core needle biopsy in patients with suspected lymphoproliferative disease: diagnostic yield in a large single-center cohort and procedural techniques

Article References: Core needle biopsy in patients with suspected lymphoproliferative disease: diagnostic yield in a large single-center cohort and procedural techniques. (n.d.). https://doi.org/10.1007/s44343-026-00051-5

Image Credits: AI Generated

DOI: 10.1007/s44343-026-00051-5

Keywords: core needle biopsy, lymphoma, lymphoproliferative disease, excisional biopsy, ultrasound-guided biopsy, CT-guided biopsy, PET-CT, interventional radiology, diagnostic yield, hematologic malignancies, Tru-cut needle, minimally invasive diagnosis