Killekar, Aditya

Abstract Background Previous studies evaluated large language model (LLM) performance on the American Society of Nuclear Cardiology (ASNC) Board Preparation Exam. Without domain-specific context, the best model (GPT-4o) achieved 63.1%, below the estimated 65% passing threshold and the 78% mean score of human fellows-in-training (FITs). Providing textbook context improved GPT-4o to 73.8% on text-only questions, but still fell short of human trainees. Whether next-generation LLMs with retrieval-augmented generation (RAG) can exceed this gap is unknown. Methods Claude Opus 4.7 and GPT-5.5 were administered all 168 questions (141 text-only, 27 image-based) from the 2023 ASNC Board Preparation Exam across 5 iterations each, using RAG with a nuclear cardiology textbook, companion atlas, and ASNC clinical guidelines. Claude used local FAISS-based semantic retrieval; GPT-5.5 used Azure’s cloud-hosted vector store. Performance was compared to prior LLM results and 13 human FITs. Results Across 5 iterations, Claude Opus 4.7 achieved a mean accuracy of 86.3% ± 1.4% (text 88.8%, image 73.3%). GPT-5.5 achieved 86.7% ± 2.2% (text 88.5%, image 77.0%) but refused a mean of 12.2 questions (7.3%) per iteration due to safety filters. Both models surpassed the human FIT mean (78.0%) and the estimated passing threshold. Compared to GPT-4o without context (63.1%), this represents a 23-percentage-point improvement in 18 months. Conclusion Next-generation LLMs with RAG now surpass average human trainee performance on nuclear cardiology board preparation questions, suggesting significant potential as educational tools and knowledge-reference aids in cardiovascular imaging. Condensed Abstract Across 5 iterations each, Claude Opus 4.7 and GPT-5.5 with retrieval-augmented generation achieved mean accuracies of 86.3% and 86.7% on the 2023 ASNC Board Preparation Exam (168 questions), both surpassing the mean human fellow-in-training score of 78%. GPT-5.5 refused a mean of 12.2 questions (7.3%) per iteration due to safety filters. These results represent a 23-percentage-point improvement over the best prior LLM without context (63.1%), demonstrating that RAG-enhanced LLMs have reached human-level proficiency in nuclear cardiology knowledge. Graphical Abstract Overview of the three-study research arc evaluating LLM performance on the 2023 ASNC Board Preparation Exam. Study 1 (2024) tested four LLMs without context (best: GPT-4o, 63.1%). Study 2 (2025) added textbook context to GPT-4o (73.8%). Study 3 (2026, current) evaluated Claude Opus 4.7 and GPT-5.5 with retrieval-augmented generation across 5 iterations each (mean 86.3% and 86.7%, respectively), both surpassing the human fellow-in-training mean of 78%. Right panel shows the performance scale with key thresholds.

Abstract Purpose Spatial distribution of coronary artery calcium (CAC) may provide additional prognostic value in patients undergoing SPECT and PET myocardial perfusion imaging (MPI). We aimed to automatically identify CAC in proximal segments from attenuation correction CT (CTAC) scans using artificial intelligence (AI) and to evaluate prognostic significance in two large international multicenter registries. Methods From hybrid MPI/CT imaging (N=43,099) across 15 sites, we included 4,552 most relevant patients with 1) no prior coronary artery disease; 2) AI-derived mild CAC scores (1-99); and 3) normal perfusion (stress total perfusion deficit <5%). The independent associations between AI-identified proximal CAC and major adverse cardiovascular events (MACE) and all-cause mortality (ACM) were evaluated using multivariable Cox regression, likelihood ratio test (LRT), and continuous net reclassification index (NRI). Results Among the patients with mild CAC and normal perfusion (mean age 65±12 years, 51% male), 1,730 (38%) had proximal CAC. Over 3.6 (inter-quartile interval 2.1, 5.2) years follow-up, 599 (13%) and 444 (10%) patients had MACE or ACM, respectively. Proximal CAC was associated with an increased risk of MACE (adjusted hazard ratio [HR] 1.24, 95% CI 1.03–1.48, P=0.02) and ACM (adjusted HR 1.25, 95% CI 1.01–1.53, P=0.04) after the adjustment of CAC score and density, clinical risk factors, and perfusion deficit. Proximal CAC improved the risk stratification of MACE (LRT P=0.02; NRI 12%) and ACM (LRT P=0.04; NRI 12%). Conclusion In patients with mild CAC and normal perfusion, AI detection of proximal CAC identified a higher-risk group for adverse outcomes, highlighting its prognostic utility. Graphical Abstract From patients who underwent hybrid myocardial perfusion imaging (MPI) from 15 sites, we analyzed those without prior coronary artery disease (CAD), mild coronary artery calcium (CAC) scores (1-99), and normal perfusion (stress total perfusion deficit <5%). A previously developed AI model was used to identify CAC lesions in proximal coronary segments on CT attenuation correction maps (CTAC). We evaluated associations with major adverse cardiovascular events (MACE) and all-cause mortality (ACM), showing risk stratification of proximal CAC and improvement by net reclassification index (NRI). CAC lesion color: green, left anterior descending artery (LAD) with left main artery; red, left circumflex artery (LCX); yellow, right coronary artery (RCA). Adjusted hazard ratios (HRs) are shown with 95% confidence intervals.

Abstract Background Quantitative myocardial blood flow (MBF) and myocardial flow reserve (MFR) provide incremental diagnostic and prognostic value in cardiac PET, but their widespread use is limited by the technical demands of dynamic imaging protocols. We evaluated the feasibility of using artificial intelligence (AI) to predict MBF and MFR from static and gated PET images, without the need for dynamic acquisition. Methods A machine learning (XGBoost) model was trained on 82Rb PET multi-center dataset using static perfusion imaging, injected dose, hemodynamic measures, clinical data and CT-derived features (including body composition) from the hybrid CT attenuation scan. Model performance was evaluated externally in an independent cohort. Results In total, 10,566 (derivation-cohort) and 7,842 (external-cohort) patients were included in this multi-center study. On the external-cohort, AI approach achieved an Area under the curve (AUC) of 0.92 (0.92–0.93) for abnormal stress MBF and 0.91 (0.90–0.92) for abnormal MFR; Intra-class correlation (ICC) 0.80 (0.78–0.82) and 0.78 (0.76–0.79), respectively. AI MFR closely mirrored the prognostic performance of measured MFR, showing nearly identical Kaplan–Meier risk stratification (both p<0.0001) and maintaining strong, and independently significant associations with all-cause mortality (HR 3.4 [2.8–4.2] vs. 4.6 [3.6–5.8]; both p<0.001), and demonstrated similar added value to perfusion for mortality prediction. Conclusion AI-predicted virtual stress MBF and MFR assessment using static and gated PET data is feasible and generalizable across cohorts. By removing the dependency on dynamic acquisitions, this approach has the potential to broaden the clinical adoption of flow quantification. STRUCTURED GRAPHICAL ABSTRACT PET: Positron Emission Tomography, CT: Computed Tomography, MFR: Myocardial Flow Reserve Key Question: Can machine learning models trained on dynamic PET datasets accurately predict regional stress myocardial blood flow (MBF) and myocardial flow reserve (MFR) from static image features, physiological parameters, and CT-based anatomical measures? Key Finding: Artificial intelligence can accurately estimate MBF and MFR from non-dynamic PET data, with strong agreement to reference standards. Take-home Message: By eliminating reliance on dynamic PET acquisitions, machine-learning has the potential to broaden clinical adoption of quantitative flow assessment.

BACKGROUND: Computed tomography (CT) attenuation correction scans are an intrinsic part of positron emission tomography (PET) myocardial perfusion imaging using PET/CT, but anatomic information is rarely derived from these ultralow-dose CT scans. We aimed to assess the association between deep learning-derived cardiac chamber volumes (right atrial, right ventricular, left ventricular, and left atrial) and mass (left ventricular) from these scans with myocardial flow reserve and heart failure hospitalization. METHODS: We included 18 079 patients with cardiac PET/CT from 6 sites. A deep learning model estimated cardiac chamber volumes and left ventricular mass from CT attenuation correction imaging. Associations between deep learning-derived CT mass and volumes with heart failure hospitalization and reduced myocardial flow reserve were assessed in a multivariable analysis. RESULTS: During a median follow-up of 4.3 years, 1721 (9.5%) patients experienced heart failure hospitalization. Patients with 3 or 4 abnormal chamber volumes were 7× more likely to be hospitalized for heart failure compared with patients with normal volumes. In adjusted analyses, left atrial volume (hazard ratio [HR], 1.25 [95% CI, 1.19–1.30]), right atrial volume (HR, 1.29 [95% CI, 1.23–1.35]), right ventricular volume (HR, 1.25 [95% CI, 1.20–1.31]), left ventricular volume (HR, 1.27 [95% CI, 1.23–1.35]), and left ventricular mass (HR, 1.25 [95% CI, 1.18–1.32]) were independently associated with heart failure hospitalization. In multivariable analyses, left atrial volume (odds ratio, 1.14 [95% CI, 1.0–1.19]) and ventricular mass (odds ratio, 1.12 [95% CI, 1.6–1.17]) were independent predictors of reduced myocardial flow reserve. CONCLUSIONS: Deep learning–derived chamber volumes and left ventricular mass from CT attenuation correction were predictive of heart failure hospitalization and reduced myocardial flow reserve in patients undergoing cardiac PET perfusion imaging. This anatomic data can be routinely reported along with other PET/CT parameters to improve risk prediction.