Collins, Jeremy D.

Dynamic gadopiclenol-enhanced nodal and pedal MR lymphangiography was safe, with no adverse events, and feasible for imaging peripheral and central conducting lymphatic anatomy.

The creation of an arteriovenous fistula in a uremic pig induces cardiac remodeling, which resembles contemporary clinical findings in patients with end-stage kidney disease (ESKD) and includes increased end-diastolic volumes and cardiomyocyte hypertrophy, potentially driven by fibrosis and senescence. This novel large animal model may now enable mechanistic and therapeutic studies aiming to ease the unacceptably high burden of cardiovascular disease on patients with ESKD.

Background Large language models (LLMs) are entering clinical workflows, yet their alignment with medical-ethical principles is unclear. Objective To evaluate how LLMs reconcile deontological and utilitarian reasoning in medically adapted “Trolley dilemma” scenarios. Methods We tested 14 LLMs, including GPT-o1-preview and DeepSeek-R1-Distill-Llama, using five medically adapted Trolley dilemmas sourced from philosophical and real-world scenarios. Each model generated 10 independent answers per case (overall 700 responses). Responses were forced to “yes” (utilitarian) or “no” (deontological). The primary outcome was the proportion of utilitarian choices. χ 2 tests compared models. Results LLM responses varied. Some models, such as GPT-4 and Gemma, selected the utilitarian option in 20% of cases, while others, like Qwen-2-7B, reached 80% ( p < .001). GPT-o1-preview and DeepSeek-R1-Distill-Llama selected utilitarian responses in 44% and 38% of cases, respectively. Some models selected the consequence-maximizing option in vignettes where doing so required overriding consent or endorsing intentional harm within the scenario framing. Seven of the fourteen models tested recommended impermissible medical actions, including nonconsensual limb amputation, killing a healthy person for organ harvesting, and transferring a patient to hospice against their wishes, in 44 of 700 outputs (6.3 %). Conclusions In this constrained forced-choice stress test, models occasionally endorsed boundary-violating actions within the vignette framing. These results do not generalize to routine clinical decision-making, but motivate further benchmarked evaluation of refusal behavior and boundary constraints in ethically sensitive medical contexts. We must carefully define how these models influence clinical decisions, establish robust ethical guidelines for their use, and ensure that fundamental ethical norms are not compromised.

A 32-fold accelerated real-time cine MRI sequence (radial sampling) demonstrated better temporal resolution and image quality than a 16-fold sequence (Cartesian sampling) in patients with cardiac implantable electronic devices and arrhythmias.

Background: The arteriovenous fistula (AVF) is a requirement for hemodialysis therapy in end-stage kidney disease (ESKD) patients. A consequence of AVF creation is compensatory cardiovascular hemodynamics, which are subsequently associated with cardiac remodeling. Relatedly, cardiovascular mortality and morbidity are elevated in ESKD patients, which worsens in dialysis patients. Currently, no suitable uremic large animal models exist to investigate the underlying mechanisms of AVF-induced cardiac remodeling. Research Question: This study aims to characterize cardiovascular changes secondary to AVF creation, supported by percutaneous transluminal angioplasty (PTA), in a uremic pig model. Methods: Chronic kidney disease (CKD) was induced via renal embolization, followed by AVF creation 28 days later. The AVF was created by anastomosis of the left common carotid artery to the left external jugular vein. AVF stenosis was alleviated 28 days thereafter via PTA, and cardiac MRI was performed at 14-, 28-, and 42-days post-PTA. Results: Increased end-diastolic volumes were observed in the left and right ventricles at day 42 vs 14 (200.8ml to 249.5ml, p=0.0457, and 204.6ml to 260.0ml, p=0.0268 respectively), while systolic function was preserved. Left ventricle (LV) stroke volume and blood flow through the aorta, pulmonary artery, and vena cava were also increased. In perivascular areas of the free wall of the LV at day 42 vs control, senescence markers showed increased p16 expression (fold change 10.21, p=0.0086) and decreased p21 expression (fold change 0.07, p=0.0357). The LV showed perivascular fibrosis (0.2% to 0.59%, p=0.021), via picrosirius red, reduced collagen-type IV expression (fold change 0.81, p=0.0089), and increased MMP2 levels (fold change 14.33, p=0.0213). Cardiomyocyte cross-sectional area was increased (from 800.9 µm 2 to 1250.0 µm 2 , p=0.0296), without CD4+ or CD68+ cell LV infiltration. Conclusion: In conclusion, AVF creation leads to modified left and right ventricular function and increased peripheral flow, potentially mediated by cellular senescence and fibrosis, resulting in progressive cardiac remodeling. This model of uremic kidney disease can now be used to evaluate mechanisms of AVF induced cardiac disease and test the efficacy of therapeutics like senolytics and anti-fibrotic agents.

Background: The prevalence of cardiac implantable electronic devices (CIED) is increasing, with over 300,000 new implants annually in the U.S. Stress cardiac magnetic resonance (CMR) has proven as a powerful modality for myocardial characterization, enabling accurate diagnosis of different cardiac conditions. In patients without devices, CMR findings have also demonstrated prognostic value for major adverse cardiovascular events (MACE) and guide clinical decision-making. However, its use in CIED patients has historically been limited due to safety and image quality concerns. Hypothesis: We aimed to evaluate the prognostic value of CMR imaging using wideband sequences in CIED patients. Methods: CIED patients referred for clinical CMR were included prospectively and retrospectively between January 2020 and February 2025. All patients underwent a stress CMR protocol, which included cine, wideband late gadolinium enhancement (LGE), wideband T1 mapping, and wideband stress/rest perfusion imaging. Clinical follow-up is ongoing. The primary endpoint was the occurrence of MACE, defined as the composite of myocardial infarction, heart failure hospitalization, implantable cardioverter defibrillator (ICD) therapy, cardiac death, revascularization, or heart transplant. Results: A total of 89 patients were included (median age 69.0 [58.0–75.5] years; 62.9% male). Device types included pacemakers (33.7%), ICDs (48.3%), and cardiac resynchronization therapy devices (16.9%). Over a median of follow-up of 613 days [355-991], 21 patients experienced a MACE. In unadjusted models, higher left ventricular ejection fraction (LVEF) (per 5%) was significantly associated with a lower risk of MACE (odds ratio [OR] 0.80, 95% confidence interval [CI] 0.65–0.95; p=0.016), as was higher stress myocardial blood flow (MBF) (per 0.25 mL/g/min increase: OR 0.78, 95% CI 0.60–0.98; p=0.05). In contrast, greater LV myocardial mass (per 10 g) was significantly associated with a higher risk of MACE (OR 1.27, 95% CI 1.08–1.52; p = 0.006), figure 1. Conclusion: The use of wideband sequences enabled the assessment of prognostic imaging value in CIED patients. LVEF remains a strong and consistent predictor of MACE in this population and LV mass was significantly associated with higher rates of MACE reflecting the influence of underlying cardiomyopathy. While stress perfusion CMR imaging showed promising trends, further studies with longer follow-up are needed to clarify its prognostic value in this population.

Abstract Objectives Large language models (LLMs) offer potential in natural language processing tasks in healthcare. Due to the need for high accuracy, understanding their limitations is essential. The purpose of this study was to evaluate the performance of LLMs in classifying radiology reports for the presence of pulmonary embolism (PE) under various conditions, including different prompt designs and data perturbations. Materials and Methods In this retrospective, institutional review board approved study, we evaluated 3 Google’s LLMs including Gemini-1.5-Pro, Gemini-1.5-Flash-001, and Gemini-1.5-Flash-002, in classifying 11 999 pulmonary CT angiography radiology reports for PE. Ground truth labels were determined by concordance between a computer vision-based PE detection (CVPED) algorithm and multiple LLM runs under various configurations. Discrepancies between algorithms’ classifications were aggregated and manually reviewed. We evaluated the effects of prompt design, data perturbations, and repeated analyses across geographic cloud regions. Performance metrics were calculated. Results Of 11 999 reports, 1296 (10.8%) were PE-positive. Accuracy across LLMs ranged between 0.953 and 0.996. The highest recall rate for a prompt modified after a review of the misclassified cases (up to 0.997). Few-shot prompting improved recall (up to 0.99), while chain-of-thought generally degraded performance. Gemini-1.5-Flash-002 demonstrated the highest robustness against data perturbations. Geographic cloud region variability was minimal for Gemini-1.5+-Pro, while the Flash models showed stable performance. Discussion and Conclusion LLMs demonstrated high performance in classifying radiology reports, though results varied with prompt design and data quality. These findings underscore the need for systematic evaluation and validation of LLMs for clinical applications, particularly in high-stakes scenarios.

Background Disease‐causative variants in LMNA ‐encoded lamin A/C cause a genetic cardiomyopathy characterized by atrioventricular block, atrial fibrillation, ventricular arrhythmias, and systolic dysfunction. The influence of LMNA variant type/localization on late gadolinium enhancement (LGE) patterns and clinical outcomes remains unclear. Methods Retrospective analysis of 822 genotype‐positive patients with arrhythmogenic/dilated cardiomyopathy was used to identify those with disease‐causative variants in LMNA . Data on LGE distribution and prevalence of advanced heart failure, thromboembolic and sudden cardiac death/major ventricular arrhythmia events were extracted from the electronic record and analyzed by variant type/localization. Results Among the 72/116 (62%) LMNA variant‐positive patients with cardiac magnetic resonance imaging data, LGE was observed in 40/72 (56%) cases. Most exhibited a nonischemic, midmyocardial or subepicardial pattern (73%), and 6/40 (15%) showed a unique “pseudo‐infarct” transmural pattern, predominantly affecting the apical segments. All 6 patients with this distinct LGE pattern harbored C‐terminal IgD (immunoglobulin‐like domain) variants (p.Arg471His or p.Arg541His). In patients with clinically manifest disease (76/116), those with IgD‐localizing variants had a lower prevalence of atrioventricular block (25% versus 72%, P =0.002) and atrial fibrillation (50% versus 81%, P =0.019) but higher rates of thromboembolic events (42% versus 16%, P =0.038). During a median follow‐up of 37 months, IgD variant presence independently predicted sudden cardiac death/major ventricular arrhythmia (hazard ratio, 2.391 [95% CI, 1.046–5.464]; P =0.039). Conclusions LMNA missense variants localizing to IgD present a distinct apical pseudo‐infarct LGE pattern associated with increased risk of ventricular arrhythmias and thromboembolic events but reduced atrioventricular block and atrial fibrillation. Multicenter studies are warranted to develop variant‐specific risk‐stratification strategies in cardiac laminopathy.

Abstract Background: Prostate artery embolization (PAE) requires a careful understanding of pelvic arterial anatomy and identifying prostatic artery variants. Pre-procedure CTA and intra-procedural cone beam CT are traditional means of planning and performing PAE, with the latter providing guidance for embolization. Photon counting detector (PCD) CT is a novel technology enabling ultra-high spatial resolution (UHR) whole-body imaging, with reconstructed slices as thin as 200 um. For PAE, we obtain a single UHR PCD CT acquisition, which provides both detailed pre-procedure pelvic arterial anatomy information and a registration reference for intra-procedural embolization guidance. Case Presentation: In five patients who underwent technically successful PAE via a left transradial approach, the pre-procedure diagnostic UHR PCD prostate CTA delineated bilateral prostatic artery origins and course in all cases, as confirmed with conventional angiograms. Further, registration of the UHR PCD CT for embolization guidance was successful in all cases, augmenting vessel selection. No complication occurred. Conclusion: UHR PCD CT is a novel acquisition option that is well suited to PAE due to its ability to delineate small vessels. Further, our experience has shown that a single pre-procedure scan can also be utilized for intra-procedural embolization guidance, obviating the need for a traditional cone beam CT angiogram. Further research is needed to assess reductions in radiation dose, contrast use, and intra-procedural time with these techniques.

AbstractPurposeOverall there is a lack of evidence on the accuracy and precision of phase‐contrast (PC) MRI in patients with cardiac implantable electronic devices (CIEDs). The purpose of this study is to determine whether aortic velocity measurements are influenced by a CIED.MethodsWe scanned 21 adult patients and 8 pediatric volunteers using clinical standard PC and real‐time PC (rt‐PC) sequences with and without a CIED generator taped on human subjects (below the left clavicle for adults and children; also on the abdomen for children) to mimic image artifacts. Peak and mean velocities above the aortic valve were calculated.ResultsThe Bland–Altman analyses on peak velocity measurements in pediatric subjects showed that both the accuracy and precision worsen as the distance between the CIED and aortic valve decreases (i.e., from abdomen to below the left clavicle). Specifically, both the bias and the coefficient of variation (CV) for both clinical PC and rt‐PC increased from the abdominal position (clinical: bias = −1.1%, CV = 4.3%; rt‐PC: bias = −0.3%, CV = 3.4%) to the clavicle position (clinical: bias = −4.0%, CV = 8.1%; rt‐PC: bias = 8.2%, CV = 7.3%). A similar trend was observed for mean velocity. The mean difference in peak and mean velocity measurements between rt‐PC with CIED (either position) and clinical standard PC with no CIED was within 7.5%. In adult patients, the mean difference between rt‐PC with CIED and clinical standard PC with CIED in peak velocity was 6.9%, and the CV was 7.9%.ConclusionThis study demonstrates that aortic velocity measurements are influenced by CIED in both adult and pediatric subjects.