Publications

PURPOSE: To evaluate Frequency Alternation at Low duty cycle for Single Offset (FALSO) as a novel Magnetization Transfer (MT) preparation scheme to increase speed and/or spatial resolution of inhomogeneous MT (ihMT) MRI by reducing the number of volumes required.

METHODS: We compared FALSO MT to standard single frequency MT preparations using signal simulations and ihMT data acquired in a rat brain at 9.4 T and human brains at 3 T. Using FALSO MT preparations, combined with optimized Variable Flip Angle (VFA) MPRAGE readouts, we also acquired ihMT data at high resolution (down to 1.4 mm isotropic) to demonstrate the reduction in acquisitions required for ihMT images of sufficient SNR.

RESULTS: We found no statistically significant difference between ihMT ratios from data acquired using FALSO MT preparations versus averaging data following separate positive and negative frequency offset MT acquisitions (as in regular ihMT to account for MT asymmetry). Use of VFA (relative to constant flip angle) readouts allowed high-resolution ihMT images with a minimal number of volumes to be acquired and reduced variance in ihMT ratios within ROIs.

CONCLUSION: FALSO MT preparations allow for a reduction in the number of acquisitions required for ihMT while still controlling for MT asymmetry. IhMT data acquired with VFA readouts allow for higher resolution acquisitions in the same scan time. The combination of FALSO MT and VFA readouts can be used to increase the spatial and/or temporal resolution of ihMT experiments, allowing for easier clinical translation and improving the utility of ihMT in neurological studies of myelin.

PURPOSE: Inhomogeneous MT (ihMT) is an MRI modality sensitive to Myelin, the lipid-rich membrane surrounding the axons. A novel ihMT saturation strategy and its associated B 1 + correction at UHF are proposed. The aim is to generate a strong ihMT effect within a clinically compatible scan time while complying with SAR constraints at 7 T.

METHODS: The MT preparation module of the proposed approach was shortened, reducing power deposition and allowing for shorter scan time while maximizing SNR efficiency. Optimization of the sequence was validated on N = 4 (2 mm iso.) on a 7 T clinical scanner. Example high resolution images were acquired (1-1.4 mm iso.). Model-based B 1 + corrections were applied and their impact on ihMT quantification was studied. Numerical simulations and error quantifications were performed to complement experimental data.

RESULTS: Fewer bursts ( N B = 5 ) of cosine-modulated MT preparation pulses ( N P = 1 ) performed at maximal B 1 rms yielded strong ihMT signal with better SNR efficiency as compared to previously proposed long saturation trains ( N B = 10 ). B 1 + correction showed a reliability range ( B 1 , rel + ∈ [ 80 , 120 ] % ) which challenged applicability in strong hypo-intense B 1 + areas. High resolution acquisitions showed acceptable SNR ( ≥ 25 ) for resolutions down to 1 mm iso., enabling studies of fine structures (e.g., deep gray nuclei) with added specificity provided by ihMT.

CONCLUSION: ihMT applications at 7 T are enabled. While whole-brain imaging requires further work to compensate for B 1 + inhomogeneities, localized applications (e.g., Thalamus) are immediately feasible within clinically compatible scan time (<  8 min). Similar sequence improvements are expected at lower field strengths (e.g., B 0 = 3 T ).

OpenEvidence (OE) is a retrieval-augmented clinical AI tool. We evaluated the references OE returned to 150 standardized prompts across five specialties, analyzing all 4979 citations. No reference was confirmed as fabricated; three abstracts had attribution errors. Most references were recent and high-impact, though journal concentration varied by specialty. This study assessed reference existence and characteristics, not whether references support OE's clinical claims - a benchmark for evaluating clinical AI tools.

BACKGROUND: Large language model (LLM) chatbots are increasingly consulted for triage decisions. A structured benchmark reported that ChatGPT Health, a consumer-facing health assistant, under-triaged 51.6 % of true emergencies and was susceptible to social anchoring. Whether a physician-facing clinical decision support platform fails similarly is unknown.

OBJECTIVE: To characterize the triage safety profile of OpenEvidence, a retrieval-augmented, physician-facing platform, using the identical benchmark previously applied to ChatGPT Health.

METHODS: We evaluated 60 clinician-authored vignettes from 30 clinical scenarios across 21 domains. Each scenario was written with and without objective clinical data and crossed with demographic and contextual modifiers in a 2 × 2 × 2 × 2 factorial design, yielding 960 prompts (480 clear-case, 480 edge-case). Responses were classified against a clinician gold standard as correct triage, under-triage, over-triage, or evidence-seeking refusal. Analyses used cluster bootstrap resampling, mixed-effects logistic regression, and Holm-Bonferroni correction.

RESULTS: Among 449 clear-case responses that returned a recommendation, accuracy was 71.3 %. OpenEvidence under-triaged 12.5 % of emergency presentations versus 51.6 % in the previously reported ChatGPT Health benchmark, and over-triaged 68.0 % of nonurgent Home presentations (ChatGPT Health, 64.8 %). Anchoring statements did not alter recommendations (OR = 1.08, 95 % CI 0.62-1.88; Holm-adjusted p = 1.0). Objective clinical data eliminated emergency under-triage (25 % to 0 %; p = 0.005) and reduced nonurgent over-triage (78.7 % to 57.8 %; p = 0.014). In 65 of 960 responses (6.8 %), the platform declined to assign a triage level, exclusively in symptom-only Home or Routine prompts.

CONCLUSIONS: Under this benchmark, OpenEvidence produced fewer missed emergencies than the historical ChatGPT Health comparison, while errors concentrated in over-triage and evidence-seeking refusal. These findings support evaluating health AI within its deployment context and treating refusal as a distinct output category whose clinical implications require separate assessment.

BACKGROUND: [177Lu]Lu-PSMA-617 improves survival and quality of life in men with metastatic castration-resistant prostate cancer but requires substantial healthcare involvement. Our analysis evaluated the time toxicity associated with [177Lu]Lu-PSMA-617, representing the first dedicated radioligand time toxicity assessment.

MATERIALS AND METHODS: We conducted a multi-institutional retrospective study of patients initiating [177Lu]Lu-PSMA-617 between April 2022 and March 2023. Patients were followed from first [177Lu]Lu-PSMA-617 administration until 3 months after the final cycle, initiation of new systemic therapy, or death. Time toxicity was quantified as healthcare contact days during the at-risk interval. Associations with baseline characteristics were evaluated using negative binomial regression models. Univariate and multivariable models were performed.

RESULTS: Among 135 patients, median age was 69 years and median prior systemic therapies was 4. Patients completed a median of 5 [177Lu]Lu-PSMA-617 cycles over 7.6 months. Median healthcare contact days were 20 (IQR, 15-28), representing 9.5% (IQR, 6.4-13.9) of at-risk days. Fifteen patients (11%) experienced high time toxicity (>1 contact day per 5 days), while 42 (31%) had low time toxicity (≤1 contact day per 14 days). Among the 57 patients who received both cycle 1 and cycle 6, mean per cycle healthcare contact days were 4 and 3.6, respectively (paired Wilcoxon signed-rank test, p = 0.04). Routine oncology visits comprised 32.7% of contact days, compared with 8.8% from hospitalizations. Lower baseline hemoglobin was independently associated with greater time toxicity (RR = 0.87; 95% CI, 0.79-0.96; p = 0.004).

CONCLUSION: In heavily pretreated patients, [177Lu]Lu-PSMA-617 demonstrated relatively low time toxicity, supporting it as a time-efficient therapy.

OBJECTIVES: Two in three US physicians now use healthcare AI, and large language models (LLMs) are entering the triage workflows that determine which patients reach rheumatology and how quickly. We aimed to test whether nine prespecified cues in referral notes, patient descriptions and demographics shift AI-assisted triage decisions when the underlying clinical information is unchanged.

METHODS: We conducted a controlled, physician-validated experiment across 30 physician-authored rheumatology vignettes, each independently rephrased three times (90 case variants). We tested five LLMs from three providers - Anthropic, Google, and OpenAI - under nine dimensions spanning demographics, clinical context, and communication framing, with 57 controlled contextual modifications, four system-prompt personas, and five repetitions per cell, yielding more than 200,000 model queries. Sixteen clinical fields were graded against physician-validated ground truth, with excellent inter-rater agreement (Fleiss' kappa=0.92).

RESULTS: Baseline composite concordance with expert ground truth was high at 0.869. We had expected sociodemographic cues to be the strongest source of distortion. Instead, the largest shifts came from how the case was framed and described. When patients were described as anxious, models attributed symptoms to psychological rather than organic causes nearly three times as often as at baseline (13.1% vs 4.5%; odds ratio 3.2 versus stoic framing), a shift that risks relabeling organic disease as functional. Clinician anchoring in the referral note reduced concordance, consistently across models and rephrasings and significantly for acuity (dismissive anchor, vignette-level p = 0.01), mainly by downgrading urgency. In contrast, race or ethnicity, socioeconomic status, and language barrier produced no detectable effect, including in mixed-effects models that accounted for repeated vignette use.

CONCLUSION: Although baseline concordance with specialist ground truth was high, it was readily disrupted by how referral notes were worded and how patients described their symptoms, not by patient demographics. Before AI-assisted triage enters rheumatology referral pathways, systems should separate objective clinical evidence from interpretive framing, and urgency and psychological attribution should be treated as auditable safety signals.

Coiera and Fraile-Navarro question whether AI scribes are being evaluated on metrics that truly impact care. While current evaluations focus on the quality of the initial draft, signed clinical notes are dynamic, as their content can be copied, summarized, coded, and re-ingested by downstream AI tools. We argue that safety must be measured downstream, focusing on how small errors in initial documentation can compound across the patient's electronic health record.