Faculty Bibliography

OBJECTIVES/OBJECTIVE:Quantitative proton density fat fraction (PDFF) and R2* estimation at lower field strengths, such as 0.55 T, is challenging due to lower signal-to-noise ratio, reduced fat water chemical shift, and increased T2* relaxation times. In this study, we propose a 3D hybrid technique for abdominal imaging at 0.55 T that enables the simultaneous acquisition of T2-weighted and T1-weighted images and quantification of fat fraction and R2* parameters. MATERIALS AND METHODS/METHODS:Numerical simulations were performed to optimize a prototype radial hybrid turbo spin echo gradient echo (TSE-GRE) acquisition scheme for improved PDFF and R2* estimation accuracy. Phantom imaging experiments with and without motion were performed to evaluate the sensitivity of the estimation to external motion. Eleven volunteers were imaged on a prototype 0.55 T system. Data were acquired using the proposed technique under free-breathing conditions, and motion-compensated reconstruction was performed using the respiratory signal from a pilot-tone device. Image contrast and estimation performance were compared with conventional acquisition schemes in vitro and in vivo. RESULTS:Numerical simulations indicated R2* estimation accuracy was more sensitive to the choice of echo time compared with PDFF. Performing motion compensation reduced the mean error in R2* from 24 to 5 s-1 while the mean error in PDFF only reduced from 2.7% to 1.6%. The proposed technique generated T2-weighted images with comparable relative liver-spleen contrast as conventional imaging and there were no significant differences (P>0.05) in the PDFF and R2* values estimated from the hybrid technique compared with conventional multi-echo GRE. Further, the free-breathing acquisition allowed improved slice coverage while overcoming breath-hold limitations of conventional acquisition schemes. CONCLUSIONS:The use of a hybrid TSE-GRE acquisition technique can allow simultaneous morphological and quantitative PDFF and R2* estimation at 0.55 T under free-breathing conditions.

OBJECTIVE:To determine whether the timing of parathyroid surgery impacts the risk of renal stone retreatment and cardiovascular interventions. SUMMARY BACKGROUND DATA/BACKGROUND:Long-term, untreated primary hyperparathyroidism is associated with significant morbidity including nephrolithiasis and cardiovascular disease. METHODS:We conducted a Population-based Cohort study of New York and California state-wide data from 2000-2020. Adult patients who underwent renal stone treatment and subsequently diagnosed with primary hyperparathyroidism (pHPT) and underwent parathyroidectomy (PTX) were included. Patients were excluded if PTX was prior to index stone procedure, they underwent second stone treatment within 6 months, with stage V CKD, with secondary or tertiary hyperparathyroidism, with prior kidney transplant or hemodialysis, or with prior cancer diagnosis. Rate of renal stone retreatment and cardiovascular interventions after PTX in pHPT patients with nephrolithiasis who underwent parathyroid surgery at ≤ 2 years and >2 years after index stone procedure was measured. RESULTS:We identified 2,093 patients who underwent first-time stone treatment and subsequent PTX. The median time to PTX was 560 days (IQR 187-1477) and follow-up was 7.4 years (IQR 4.5-13.1). Delaying PTX for more than 2 years increased the risk of renal stone retreatment by 59% (HR 1.59; P<0.001), increased the risk of experiencing coronary disease or associated interventions by 118% (HR=2.18; P=0.01), and increased the risk of experiencing an overall cardiovascular event by 52% (HR 1.52; P<0.01). CONCLUSIONS AND RELEVANCE/CONCLUSIONS:In symptomatic pHPT, delaying PTX significantly increases the risk of requiring future stone retreatment and cardiac/vascular surgical interventions. This highlights the importance of early surgical referral and multidisciplinary approaches to optimize outcomes and resource utilization in pHPT.

Machine learning (ML) on structural MRI data shows high potential for classifying Alzheimer's disease (AD) progression, but the specific contribution of brain regions, demographics, and proteinopathy remains unclear. Using Alzheimer's Disease Neuroimaging Initiative (ADNI) data, we applied an extreme gradient-boosting algorithm and SHAP (SHapley Additive exPlanations) values to classify cognitively normal (CN) older adults, those with mild cognitive impairment (MCI) and AD dementia patients. Features included structural MRI, CSF status, demographics, and genetic data. Analyses comprised one cross-sectional multi-class classification (CN vs. MCI vs. AD dementia, n = 568) and two longitudinal binary-class classifications (CN-to-MCI converters vs. CN stable, n = 92; MCI-to-AD converters vs. MCI stable, n = 378). All classifications achieved 70-77% accuracy and 61-83% precision. Key features were CSF status, hippocampal volume, entorhinal thickness, and amygdala volume, with a clear dissociation: hippocampal properties contributed to the conversion to MCI, while the entorhinal cortex characterized the conversion to AD dementia. The findings highlight explainable, trajectory-specific insights into AD progression.

PURPOSE/OBJECTIVE:To assess whether slit-scanning specular microscopy (CellChek C; Konan Medical) can repeatedly image the same corneal location using anatomic landmarks (posterior corneal rings and corneal undulations) and unique cells identified during imaging. METHODS:A total of 203 eyes (113 patients) with and without corneal pathology were imaged to assess the prevalence of anatomic landmarks. A subcohort of 20 healthy eyes was used to identify unique cells adjacent to anatomic landmarks. Landmarks were then used to locate the same cells on repeat imaging approximately 1 week later. Endothelial cell density (ECD), coefficient of variation, and percent hexagonality were calculated. Intraclass correlation coefficient and 95% limits of agreement were used to measure variability and reproducibility of imaging. RESULTS:Approximately 91% of eyes had either posterior corneal rings or undulations present. Undulations were more common than posterior corneal rings in both healthy and diseased corneas. Among subcohort eyes, unique cells were found adjacent to anatomic landmarks in 100% of eyes. Landmarks were used to reimage the same cells in 75% of eyes. There was minimal variation in ECD, coefficient of variation, and hexagonality; intraclass correlation coefficient and 95% confidence intervals were 0.891 [0.715-0.962], 0.612 [0.179-0.849], and 0.793 [0.499-0.925], respectively. The 95% limits of agreement for ECD was -359.9-260.98. CONCLUSIONS:Landmarks identified with slit-scanning specular microscopy allowed reliable reimaging of the same endothelial location, providing a powerful tool to better understand the role of the peripheral endothelium in health and disease.

Recent advances in Artificial Intelligence (AI) methodologies and their application to medical imaging has led to an explosion of related research programs utilizing AI to produce state-of-the-art classification performance. Ideally, research culminates in dissemination of the findings in peer-reviewed journals. To date, acceptance or rejection criteria are often subjective; however, reproducible science requires reproducible review. The Machine Learning Education Sub-Committee of the Society for Imaging Informatics in Medicine (SIIM) has identified a knowledge gap and need to establish guidelines for reviewing these studies. This present work, written from the machine learning practitioner standpoint, follows a similar approach to our previous paper related to segmentation. In this series, the committee will address best practices to follow in AI-based studies and present the required sections with examples and discussion of requirements to make the studies cohesive, reproducible, accurate, and self-contained. This entry in the series focuses on image classification. Elements like dataset curation, data pre-processing steps, reference standard identification, data partitioning, model architecture, and training are discussed. Sections are presented as in a typical manuscript. The content describes the information necessary to ensure the study is of sufficient quality for publication consideration and, compared with other checklists, provides a focused approach with application to image classification tasks. The goal of this series is to provide resources to not only help improve the review process for AI-based medical imaging papers, but to facilitate a standard for the information that should be presented within all components of the research study.

BACKGROUND:Heart utilization from donation after circulatory death (DCD) donors remains highly variable across the United States, potentially resulting in missed transplantation opportunities. This study aimed to quantify the frequency of clinically viable, non-utilized DCD hearts and identify usage barriers. METHODS:We conducted a retrospective, national registry analysis of donors ≤55 years old who donated ≥1 organ, focusing primarily on DCDs. Donor characteristics, particularly age, warm ischemic time (WIT), and EF, as well as reasons for non-recovery and offer refusal, were analyzed. SRTR's heart yield model was employed to identify non-utilized DCD hearts clinically comparable to transplanted DCD hearts. RESULTS:In 2023, 613 DCD hearts were transplanted, accounting for 13.5% of all heart transplants. Only 15.5% of DCD hearts from donors ≤55 years old were utilized. Marked variation in risk-adjusted DCD heart yield was observed between states, OPOs, and Regions. Donors of transplanted DCD hearts had a median age of 32, WIT 24 minutes, and EF 63%. The yield model identified between 701-1,243 non-utilized DCD hearts with characteristics comparable to transplanted cases. Concerns about delayed progression to circulatory arrest after life support withdrawal was a key reason for non-utilization. CONCLUSIONS:Despite wider acceptance of DCD heart transplantation, an increasing proportion of DCD hearts remain unused despite favorable characteristics. Concerns related to delayed progression to circulatory arrest are a significant barrier to heart utilization. Addressing geographic variability and improving predictive models for donor viability could double DCD heart utilization and expand heart transplantation volume by approximately 700-1,200 (15-27%) annually.

BACKGROUND:Identifying nearfield and farfield signals is critical to mapping and ablating cardiac arrhythmias. This assessment is qualitative, depending on the "sharpness" of pulmonary vein (PV) electrograms. Electrogram peak frequency (PF) analysis is hypothesized to be a quantitative measure of signal proximity. OBJECTIVE:To confirm if PF defines nearfield versus farfield electrical signals and if it can be used during ablation to establish pulmonary vein isolation (PVI). METHODS:We created a cohort of 30 patients with AF undergoing PVI. Left atrial and PV maps of PF were generated before and after PVI. In the first 10 patients with paroxysmal AF (cohort 1), a cutoff value was selected to predict nearfield versus farfield signals. This cutoff was validated in a cohort of 10 patients with paroxysmal AF (cohort 2) and a cohort of 10 patients with persistent AF (cohort 3). RESULTS:PF was lower in farfield electrograms than nearfield electrograms. A PF cutoff of 300 Hz had a sensitivity of 93.2% (95% CI 81.3% - 98.6%) in cohort 1, 90.0% (95% CI 76.3 - 97.2) in cohort 2, and 98.6% (95% CI 90.1 - 99.7%) in cohort 3 for differentiating farfield from nearfield electrograms. The specificity was 100.0% (95% CI 98.2% - 100.0%) and the AUC was 0.99 (95% CI 0.97 - 1.00) in all patients. CONCLUSIONS:We confirmed the hypothesis that PF distinguishes nearfield from farfield electrograms. PF analysis improves the recognition of PV isolation. Mapping and ablation strategies utilizing PF should be pursued to improve ablation outcomes.

The past decade has witnessed remarkable advancements in musculoskeletal radiology, driven by increasing demand for medical imaging and rapid technological innovations. Contrary to early concerns about artificial intelligence (AI) replacing radiologists, AI has instead enhanced imaging capabilities, aiding in automated abnormality detection and workflow efficiency. MRI has benefited from acceleration techniques that significantly reduce scan times while maintaining high-quality imaging. In addition, novel MRI methodologies now support precise anatomic and quantitative imaging across a broad spectrum of field strengths. In CT, dual-energy and photon-counting technologies have expanded diagnostic possibilities for musculoskeletal applications. This review explores these key developments, examining their impact on clinical practice and the future trajectory of musculoskeletal radiology.

Generalized anxiety disorder (GAD) features excessive worry, paired with physiological arousal that causes distress and interferes with daily activities. Individuals with GAD symptoms often experience persistent sleep problems, which may worsen clinical outcomes. The Contrast Avoidance Model (CAM; Newman & Llera, 2011) posits that worry is used in GAD to avoid emotional shifts by perpetuating a negative state. Research has demonstrated the CAM uniquely explains GAD pathology and may similarly explain co-occurring sleep difficulties. To better understand the high rates of sleep difficulties in GAD symptomology, this study examined the role of contrast avoidance in this relationship in a sample of individuals with anxiety symptoms. Greater contrast avoidance was expected to be positively correlated with sleep difficulties, and GAD severity was expected to be indirectly associated with sleep difficulties via greater engagement in contrast avoidance. A sample of 255 anxious adults was recruited online, and participants completed self-report measures of GAD severity, contrast avoidance, and sleep difficulties. Significant correlations were observed between study variables. The mediation model accounted for 28% of the variance in sleep difficulties and revealed a significant indirect effect of contrast avoidance in the association between GAD severity and sleep difficulties. These results indicate that contrast avoidance may be important when examining sleep difficulties among those with GAD symptoms.

BACKGROUND & AIMS/OBJECTIVE:Perianal fistulising Crohn's disease (pfCD) affects one-fifth of patients with Crohn's disease (CD), significantly affecting their quality of life. Magnetic resonance imaging is the gold standard for evaluating fistula healing in pfCD, but variability in radiological definitions hampers meaningful clinical interpretation and consistent trial design. This study aimed to establish an international consensus on the definition of radiological healing in pfCD. METHODS:The study was conducted in 2 phases. Phase 1 involved a systematic review to identify magnetic resonance imaging-based variables and indices used to define healing in pfCD, assessing methodological quality using the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) framework. Phase 2 utilized a 2-round online Delphi consensus process with 84 international experts, followed by a stakeholder meeting to achieve consensus (agreement threshold >80%). Results were reported as per Accurate Consensus Reporting Document (ACCORD) guidelines. RESULTS:A radiologically healed fistula can be defined by the absence of T2-weighted hyperintensity, a completely fibrotic fistula tract and, when contrast is used, the absence of contrast enhancement on post-contrast T1-weighted images (95% consensus). Radiological improvement of a fistula can be defined (80% consensus) by the presence of a least one essential criterion: an increasingly fibrotic fistula tract, an unequivocal reduction in one or more of the following: T2-weighted hyperintensity, fistula diameter, fistula length, abscess size, or contrast enhancement of the fistula tract. CONCLUSION/CONCLUSIONS:This international Delphi consensus standardizes radiological endpoints in pfCD, improving consistency in clinical and research settings. Future studies will validate this definition and assess how radiological changes predict long-term clinical outcomes and quality of life improvements.