Faculty Bibliography

Working memory (WM), the transient storage and manipulation of information, is a cognitive function that can improve with training. Accumulating evidence suggests that sleep, particularly slow oscillations (SOs) that occur during nonrapid eye movement sleep, supports this improvement. Yet, how sleep and SOs in particular relate to neural processes supporting WM remains poorly understood. In this study, we investigated how WM-related neural activity evolves across nocturnal sleep and how these changes relate to neural processing during SOs. Participants performed a WM task during fMRI before and after sleep, with simultaneous EEG-fMRI capturing neural activity during the first 2.5 hr of sleep. Our results showed significant overnight changes in WM-related activity, characterized by reduced recruitment of the dorsal precuneus during WM encoding/maintenance phases and frontal regions during the retrieval phase of WM. In addition, sleep increased item-specific reinstatement in a sensory processing region during WM. This pattern of results suggests a combined influence of sleep on enhancing sensory reinstatement while reducing potential top-down control (i.e., reduced parietal and frontal activity). Critically, SO-related activity was directly linked to overnight activity changes: Stronger SO activation in the premotor cortex and ventromedial pFC predicted greater overnight reductions in WM-related activity, and multivoxel analyses in the ventral attention network revealed a parallel relationship. These findings suggest that SOs play a critical role in WM function by facilitating the reorganization of WM-related processing.

Cancer management remains fragmented across its continuum, from late-stage diagnosis and salvage therapies to non-personalized surveillance. Here, we present Oncoformer, a unified multimodal transformer model trained on the China Oncology Multimodal Prediction and Surveillance Study (COMPASS) cohort (3.67 million individuals, 17.7 million clinical visits) and validated on independent external cohorts, including the UK Biobank. Oncoformer integrates longitudinal electronic health records with chest X-ray imaging to address multiple clinical tasks: pan-cancer diagnosis (area under the receiver operating characteristic curve [AUROC] = 0.956), future cancer prediction up to 1 year before diagnosis (AUROC = 0.869), tumor stage inference (mean AUROC > 0.90), patient-specific treatment-response forecasting, and recurrence-free survival stratification across ten cancer types (all p < 0.01). Staging predictions were independently validated against postoperative pathological endpoints and shown to converge on core cancer genomic pathways. By translating routine clinical data into a dynamic view of cancer evolution, Oncoformer provides a framework for risk-informed cancer prediction and treatment stratification using routine clinical data.

State sequence analysis of longitudinal categorical data seeks to synthesize pathways through different dimensions of the life course for descriptive, associative and predictive purposes. Given the number and variety of patterns in such data, measures of the dynamic features of sequences are used to characterize them. One, based on the information-theoretic notion of entropy, measures the uncertainty in the state that will be active at a given time. We customize its use to establish the extent to which we are ignorant, or unsure, of what happens next in a dynamic process, conditional on its past. Relying on different Markov chain models for nominal state sequences, we establish multiple measures of uncertainty that allow us to adjust expectations to reflect individual-specific differences and historical information. We establish complementary measures to assess the predictive power of the models in the context of this uncertainty. In so doing, we can summarize and contrast the change in uncertainty associated with different models. As is common in this field, we consider ways in which data can be stratified through demographics and clustering, and how this additional level of partitioning builds a more complete narrative of the social process.

BACKGROUND:Magnetic resonance (MR) imaging-linear accelerator-based real-time adaptive planning for delivering ultra-hypofractionated stereotactic body radiotherapy (SBRT) has advanced clinical accuracy yet added complexity to the radiotherapy workflow. We retrospectively evaluated whether the addition of a Parallel Automated Contouring module with Enhancement of AI (PACE-AI) reduced contouring time and overall SBRT duration, and its impact on intra-fractional target motion. METHODS:This study included 250 fractions from 90 prostate cancer patients from which fraction time were tracked. All patients received definitive SBRT to the prostate on the 1.5-Tesla Unity (Elekta©) system, through the Adapt-To-Shape (ATS) workflow, which required real-time re-contouring of the normal tissues by the dosimetrists and target contouring by the physician for each of the fractions. We compared the fraction durations for 125 consecutive fractions treated with the incorporation of PACE-AI with 125 fractions previously treated without PACE-AI to determine whether PACE-AI improved efficiency and reduced treatment session duration. RESULTS:For the cohort treated without PACE-AI, the overall median duration was 67.6 min, including 23.9 min contouring time. With the incorporation of PACE-AI, the overall median duration was 51.2 min, representing a 24.3% reduction. The average contouring time was reduced by 55% to 10.8 min. In addition, the extent of positional shifts prior to beam delivery was significantly reduced from an average of 1.8 mm to 1.3 mm (p < 0.001) in both superior/inferior (range reduced from 0 to 6.3 mm to 0-4.2 mm) and anterior/posterior directions (range reduced from 0 to 6.6 mm to 0-5.0 mm). CONCLUSIONS:The incorporation of PACE-AI improved workflow efficiency during SBRT. The reduction in treatment duration also helped reduce organ motion during real-time adaptive planning.

INTRODUCTION/BACKGROUND:Patients with a suspected high-consequence infectious disease (HCID), such as Ebola virus disease, may require routine laboratory testing to guide management. We assessed the capabilities of frontline hospitals and the barriers they face performing laboratory testing for patients with a suspected HCID. METHODS:A one-time confidential REDCap survey querying capabilities to safely perform laboratory tests that the Centers for Disease Control and Prevention considers critical for patients with a suspected HCID was sent to 95 institutions in Baltimore, MD, Boston, MA, New York City, NY, and Washington, DC, from January to May 2025. RESULTS:Fifty (53%) institutions responded, mostly teaching hospitals (96%), with 24% reporting prior experience evaluating a patient with suspected Ebola virus disease. While many hospitals could perform on-site blood gas (70%), hemoglobin/hematocrit (68%), and lactate (68%) tests on a suspected HCID patient, fewer could safely perform a chemistry panel (64%), a urinalysis (60%), a complete blood count with differential (56%), and a malaria rapid diagnostic test (RDT) (48%) on a suspected HCID patient. The five tests respondents most often considered extremely or very important were hemoglobin/hematocrit (91%), chemistry panel (90%), CBC with differential and platelet count (89%), malaria RDT (88%), and blood gas (88%). Reported barriers to performing routine laboratory testing included issues related to patient and staff safety, infection control, lack of appropriate space, and funding. CONCLUSIONS:Our survey identified several barriers to implementing safe laboratory testing. The inability to conduct these laboratory tests may result in delays in care when an HCID is suspected.

IMPORTANCE/UNASSIGNED:The decision to transfuse a patient with myocardial infarction (MI) and anemia at a higher vs lower hemoglobin threshold must consider the potential benefit of reduced risk of 30-day death or MI and the potential risk of heart failure. OBJECTIVES/UNASSIGNED:To estimate bayesian posterior risk differences and posterior probabilities that a liberal vs restrictive transfusion strategy is associated with reduced risk of 30-day death or MI and whether the probabilities exceed predefined thresholds. DESIGN, SETTING, AND PARTICIPANTS/UNASSIGNED:The Myocardial Ischemia and Transfusion (MINT) trial recruited adults from April 26, 2017, to April 14, 2023, who were hospitalized with MI and anemia at 144 sites in 6 countries. Statistical analysis was performed from July 31, 2024, to February 18, 2026. INTERVENTION/UNASSIGNED:The MINT trial randomized participants to a restrictive (transfuse if hemoglobin is <7 to 8 g/dL) or liberal (maintain hemoglobin at >10 g/dL) transfusion strategy. MAIN OUTCOMES AND MEASURES/UNASSIGNED:Bayesian posterior risk differences were estimated for 30-day death or MI and for heart failure using 3 prior beliefs regarding the treatment strategies: noninformative, liberal strategy superiority, or restrictive strategy superiority. RESULTS/UNASSIGNED:The mean (SD) age of the 3504 participants was 72.1 (11.6) years and 1911 (54.5%) were men. Compared with the restrictive strategy, the risk of 30-day death or MI with the liberal strategy was 1.4% (95% credible interval, -0.8% to 3.5%) to 2.4% (95% credible interval, 0.3%-4.6%) lower, depending on prior beliefs. The probability that the liberal strategy was associated with a lower risk of 30-day death or MI ranged from 89.1% to 98.8%, and the probability that a liberal strategy was associated with at least 1 less death or MI per 100 treated was between 62.7% and 90.4%. Conversely, the risk of heart failure with the liberal strategy was 0.2% (95% credible interval, -1.6% to 1.2%) to 0.6% (95% credible interval, -2.0% to 0.8%) higher compared with the restrictive strategy, depending on prior beliefs. The probability that the liberal strategy was associated with a higher risk of heart failure ranged from 60.6% to 80.0%, and the probability that a liberal strategy was associated with at least 1 more heart failure event per 100 treated was between 13.3% and 29.3%. CONCLUSIONS AND RELEVANCE/UNASSIGNED:This post hoc analysis of a randomized clinical trial of patients with MI and anemia suggests that a liberal transfusion strategy was associated with a lower risk of 30-day death or MI, outweighing the increased risk of heart failure. Consistent with guideline recommendations and according to patients' values and clinician risk assessment, a liberal transfusion strategy may be reasonable. TRIAL REGISTRATION/UNASSIGNED:ClinicalTrials.gov Identifier: NCT02981407.

AIM/OBJECTIVE:The "2026 HRS Expert Consensus Statement Update on Cardiovascular Implantable Electronic Device Lead Management and Extraction" provides updated recommendations to guide clinicians in the management of cardiovascular implantable electronic device (CIED) leads. BACKGROUND:Since the publication of the "2017 HRS Expert Consensus Statement on Cardiovascular Implantable Electronic Device Lead Management and Extraction," the field has evolved quickly. New evidence on CIED lead management and the blooming development of new CIED technologies, including leadless pacing and implantable cardioverter-defibrillator leads implanted outside the vascular system and new lumenless pacing leads and lead extraction tools, have contributed to the field's rapid evolution. METHODS AND RESULTS/RESULTS:A comprehensive literature search was conducted in accordance with the Institute of Medicine standards. The writing committee reviewed evidence gathered through electronic literature searches encompassing clinical trials, original studies, and meta-analyses conducted on human subjects published in English from MEDLINE, PubMed, Embase, and the Cochrane Library up to December 2024. The comprehensive literature review supports each evidence-based recommendation and is compiled in the evidence tables. A predefined threshold of >70% approval for each recommendation was required, with a quorum of two-thirds of the writing committee. The final mean consensus of 108 recommendations was 93.61%. DISCUSSION/CONCLUSIONS:The recommendations from the "2017 Expert Consensus Statement on Cardiovascular Implantable Electronic Device Lead Management and Extraction" have been updated with new evidence to guide clinicians. The new recommendations address the latest CIED technologies with the advantages over transvenous leads; new evidence supporting diagnosis, treatment, and prevention for CIED infection; appropriate lead management in transcatheter tricuspid valve replacement for tricuspid regurgitation; and standardization of transvenous lead extraction approach, protocol, and facilities to improve the outcomes of CIED lead management and extraction.

PURPOSE/OBJECTIVE:Optical coherence tomography angiography (OCTA) has enabled detailed in vivo imaging of macular neovascularization (MNV) in neovascular age-related macular degeneration (nAMD), leading to a proliferation of morphological descriptive terms. This scoping review aimed to systematically map the existing literature on OCTA-based MNV morphology and evaluate its correlation with disease activity. METHODS:A systematic literature search covering publications through 2025 was performed. After screening 2,445 titles and obtaining 60 full texts, 43 studies were included. Data on morphological terminology, study design, and correlation with disease activity were extracted and synthesized. RESULTS:The included studies, encompassing a total of 2,712 eyes, identified a vast and heterogeneous lexicon of qualitative descriptors, including terms such as "medusa," "sea-fan," and "glomerulus", characterized by significant terminological overlap and inconsistent definitions across studies. Inconsistent, low-certainty associations between specific OCTA morphologies and MNV activity were identified across studies, with findings limited by significant methodological heterogeneity. The evidence for these patterns as reliable biomarkers of disease activity is weak and often contradictory, and several studies were found to use OCTA features to define activity, creating circular arguments that undermine their conclusions. CONCLUSION/CONCLUSIONS:The current terminology for OCTA morphology in nAMD is fragmented and inconsistently applied. The link between these patterns and disease activity is poorly established, severely limiting their clinical utility. These findings highlight a need for a standardized, consensus-based framework for describing and interpreting OCTA findings in nAMD.

PURPOSE/OBJECTIVE:The goal of this work is to introduce an automated method to assess the quality of under-sampled MRI reconstructions. THEORY AND METHODS/METHODS:We propose a general framework for pixel-wise uncertainty quantification in accelerated MRI reconstructions, enabling automatic identification of unreliable regions without using ground-truth fully-sampled reference images. Our method integrates conformal quantile regression with learning-based image reconstruction methods to estimate statistically rigorous pixel-wise uncertainty intervals. We trained and evaluated our model on Cartesian undersampled brain and knee data obtained from the fastMRI dataset using acceleration factors ranging from 2 to 10. An end-to-end Variational Network was used for image reconstruction. RESULTS:Quantitative experiments demonstrate strong agreement between predicted uncertainty maps and true reconstruction error. Using our method, the corresponding Pearson correlation coefficient was higher than 90% at acceleration levels at and above four-fold; whereas it dropped to less than 70% when the uncertainty was computed using a simpler heuristic notion (magnitude of the residual). Qualitative examples further show the uncertainty maps based on quantile regression capture the magnitude and spatial distribution of reconstruction errors across acceleration factors, with regions of elevated uncertainty aligning with pathologies and artifacts. CONCLUSION/CONCLUSIONS:The proposed framework enables evaluation of reconstruction quality without access to fully-sampled ground-truth reference images. It represents a step toward adaptive MRI acquisition protocols that may be able to dynamically balance scan time and diagnostic reliability.

PURPOSE/UNASSIGNED:To determine whether exposure to antibiotics is associated with new-onset International Classification of Diseases (ICD) coding of geographic atrophy (GA). METHODS/UNASSIGNED:This case-control study of patients ages sixty and older used health insurance claims data from the Merative MarketScan Research Databases. A total of 3254 cases with new-onset ICD coding of GA between 2019 and 2021 were matched by year to 3249 controls without GA using propensity scores estimated by age, hypertension, U.S. Census Bureau region, and Charlson Comorbidity Index. For cases, we analyzed prescription drug claims of antibiotics in the two years before GA diagnosis. For controls, we analyzed claims in the two years before a randomly selected eye examination. We calculated the odds of new-onset ICD coding of GA controlling for age-related macular degeneration risk factors. RESULTS/UNASSIGNED:Exposure to any antibiotics (OR = 1.25; 95% confidence interval [CI], 1.11-1.41; P < 0.001), tetracyclines (OR = 1.18; 95% CI, 1.03-1.36; P = 0.021), quinolones (OR = 1.16; 95% CI, 1.03-1.31; P = 0.017), or broad-spectrum antibiotics (OR = 1.20; 95% CI, 1.07-1.34; P = 0.003) was associated with increased odds of new-onset ICD coding of GA. Greater cumulative day supply was associated with increasing odds, suggesting a dose-dependent relationship. CONCLUSIONS/UNASSIGNED:Exposure to antibiotics is associated with increased odds of new-onset ICD coding of GA. This association may be dose dependent. These findings suggest that antibiotic exposure could be a novel modifiable risk factor for GA, although this conclusion is tempered by potential confounding by health status or misclassification of GA and by multiple comparisons. Further investigation will be needed to validate these findings.