Almanac - Retrieval-Augmented Language Models for Clinical Medicine.

BACKGROUND: Large language models (LLMs) have recently shown impressive zero-shot capabilities, whereby they can use auxiliary data, without the availability of task-specific training examples, to complete a variety of natural language tasks, such as summarization, dialogue generation, and question answering. However, despite many promising applications of LLMs in clinical medicine, adoption of these models has been limited by their tendency to generate incorrect and sometimes even harmful statements. METHODS: We tasked a panel of eight board-certified clinicians and two health care practitioners with evaluating Almanac, an LLM framework augmented with retrieval capabilities from curated medical resources for medical guideline and treatment recommendations. The panel compared responses from Almanac and standard LLMs (ChatGPT-4, Bing, and Bard) versus a novel data set of 314 clinical questions spanning nine medical specialties. RESULTS: Almanac showed a significant improvement in performance compared with the standard LLMs across axes of factuality, completeness, user preference, and adversarial safety. CONCLUSIONS: Our results show the potential for LLMs with access to domain-specific corpora to be effective in clinical decision-making. The findings also underscore the importance of carefully testing LLMs before deployment to mitigate their shortcomings. (Funded by the National Institutes of Health, National Heart, Lung, and Blood Institute.).

Almanac: Retrieval-Augmented Language Models for Clinical Medicine.

Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In this study, we develop Almanac, a large language model framework augmented with retrieval capabilities for medical guideline and treatment recommendations. Performance on a novel dataset of clinical scenarios (n= 130) evaluated by a panel of 5 board-certified and resident physicians demonstrates significant increases in factuality (mean of 18% at p-value < 0.05) across all specialties, with improvements in completeness and safety. Our results demonstrate the potential for large language models to be effective tools in the clinical decision-making process, while also emphasizing the importance of careful testing and deployment to mitigate their shortcomings.

Viability of preloaded Descemet membrane endothelial keratoplasty grafts with 96-hour shipment.

OBJECTIVE: To assess feasibility and compare the effects of 96-hour shipment of Descemet membrane endothelial keratoplasty (DMEK) grafts as a scroll or a tri-fold on cell viability. METHODS AND ANALYSIS: DMEK grafts were prepared at the Rocky Mountain Lions Eye Bank. Twenty pre-stripped DMEK grafts, paired from 10 donors, were either tri-folded in an endothelium-in configuration using microforceps and loaded into a plastic Treyetech cartridge, or suctioned in a scrolled endothelium-out configuration into a modified Jones Tube. Grafts were shipped via FedEx to a secondary location and back for 48 hours each way, resulting in a total shipping time of 96 hours. After shipping, grafts were removed from inserters onto glass slides and unfolded using viscoelastic with endothelium facing upwards. Calcein-AM stained grafts were imaged with a fluorescent microscope and endothelial cell loss (ECL) was measured using trainable segmentation in Fiji by a masked grader. RESULTS: A total of 20 grafts were shipped for 96 hours, split between preloaded tri-folded (n=10) and preloaded scrolled (n=10) tissues. No significant difference in ECL was observed across groups after prolonged shipping (14.8% vs 13.7% ECL respectively, p=0.68). CONCLUSION: For preloaded DMEK after 96 hours, both scrolled and tri-folded tissue demonstrated clinically acceptable levels of ECL. The data suggest a wider window of time for endothelial cell viability and is promising for the prospect of international shipment of preloaded grafts.

Portable, non-invasive video imaging of retinal blood flow dynamics.

Retinal blood flow (RBF) information has the potential to offer insight into ophthalmic health and disease that is complementary to traditional anatomical biomarkers as well as to retinal perfusion information provided by fluorescence or optical coherence tomography angiography (OCT-A). The present study was performed to test the functional attributes and performance of the XyCAM RI, a non-invasive imager that obtains and assesses RBF information. The XyCAM RI was installed and used in two different settings to obtain video recordings of the blood flow in the optic nerve head region in eyes of healthy subjects. The mean blood flow velocity index (BFVi) in the optic disc and in each of multiple arterial and venous segments was obtained and shown to reveal a temporal waveform with a peak and trough that correlates with a cardiac cycle as revealed by a reference pulse oximeter (correlation between respective peak-to-peak distances was 0.977). The intra-session repeatability of the XyCAM RI was high with a coefficient of variation (CV) of 1.84 ± 1.13% across both sites. Artery-vein comparisons were made by estimating, in a pair of adjacent arterial and venous segments, various temporal waveform metrics such as pulsatility index, percent time in systole and diastole, and change in vascular blood volume over a cardiac cycle. All arterial metrics were shown to have significant differences with venous metrics (p < 0.001). The XyCAM RI, therefore, by obtaining repeatable blood flow measurements with high temporal resolution, permits the differential assessment of arterial and venous blood flow patterns in the retina that may facilitate research into disease pathophysiology and biomarker development for diagnostics.

Preloading Trifolded Grafts for Descemet Membrane Endothelial Keratoplasty Affects Scroll Formation.

PURPOSE: The trifolded, endothelium-in approach to Descemet membrane endothelial keratoplasty (DMEK) facilitates tissue insertion into the anterior chamber. We hypothesized that preloading the trifolded donor grafts in a cartridge for 48 hours before insertion would induce biomechanical changes that decrease their scrolling tendency compared with those loaded immediately before insertion. METHODS: Ten Descemet membrane donor grafts, peeled and cut to 8.0 mm, were prepared by a single eye bank technician. Each graft was trifolded and pulled into a DMEK cartridge and stored for 48 hours. They were then pulled with microforceps into a petri dish filled with balanced salt solution. A video was recorded of the graft becoming a scroll over a 2-minute period. Each graft, serving as its own control, was then trifolded, pulled into the cartridge, and the process repeated. Images from 1, 5, 10, 60, and 120 seconds were extracted from video recording of the procedures. Scroll width was analyzed by graders masked to group assignment. A paired t test was used to determine differences in scroll width at each time point between the 48-hour and instant trifolding conditions. RESULTS: All grafts scrolled after removal from the cartridge into balanced salt solution. We measured a significant difference at all time points 1 through 120 seconds (4.02 preloaded vs. 2.91-mm instant trifold, P = 0.035). CONCLUSIONS: Preloading DMEK grafts in a trifolded configuration for 48 hours reduces the scrolling tendency of Descemet membrane for at least 2 minutes.

An Evaluation of Sensing Technologies to Measure Intraoperative Leg Length for Total Hip Arthroplasty.

Total hip arthroplasty (THA) procedures have been identified as high-volume procedures with growing prevalence. During the procedure, orthopedic surgeons largely rely solely on qualitative assessment to ensure an excessive limb length discrepancy (LLD) is not introduced from the implant selection. LLD can result in back pain and gait complications, with some cases of LLD requiring a revision procedure to mitigate. To address this issue, we evaluated several methods of sensing distance intraoperatively to determine the best approach to measure leg length during the THA procedure. A testing setup using a sawbones model of hip anatomy in the decubitus position was used as a simulation of the THA procedure to test the accuracy of each of the sensing modalities.

Comparison of Tri-folded and Scroll-based Graft Viability in Preloaded Descemet Membrane Endothelial Keratoplasty.

PURPOSE: To compare corneal endothelial damage associated with 2 techniques for preloaded Descemet membrane endothelial keratoplasty (DMEK): a tri-folded graft stored in a plastic cartridge designed for DMEK and a scrolled graft stored in a modified Jones Tube, at the time of preparation and after shipping. METHODS: DMEK grafts were prepared at the Rocky Mountain Lions Eye Bank. The grafts were either tri-folded and loaded in a plastic cartridge or scrolled and loaded into a modified Jones Tube. In each group, the grafts were then either immediately removed from the cartridges or shipped for 48 hours. The grafts were then stained with Calcein AM and imaged using a fluorescent microscope. Endothelial cell loss (ECL) was determined using trainable segmentation in Fiji by 2 graders. At each time point, rates of ECL loss were compared across the 2 groups. To explore the role of donor characteristics, a multivariable regression model was produced to account for method (tri-folding vs. scroll), donor age, donor gender, death-to-preservation time, death-to-preparation time, and shipping. RESULTS: A total of 40 grafts were prepared, processed, imaged, and analyzed. No significant difference in cell loss was seen between groups at either time point alone. In the multivariate model, no significant increase in cell loss was associated with either tri-folding (3.7% less ECL; P = 0.051) or shipping (4.3% less ECL; P = 0.049). CONCLUSIONS: All techniques used resulted in clinically acceptable levels of ECL. Tri-folded tissue in a plastic cartridge did not result in ECL inferior to a scroll when prepared either immediately or preloaded for 48 hours.

Viability of Descemet Membrane Endothelial Keratoplasty Grafts Folded in the Eye Bank.

PURPOSE: Preloaded, trifolded grafts in Descemet membrane endothelial keratoplasty require transfer of the trifolding process from the corneal transplant surgeon to the eye bank technician. We sought to assess whether trifolding may be safely conducted by an eye bank technician with cell loss comparable to standard peeling and lifting. METHODS: A total of 10 grafts were stained, peeled, and transferred directly onto a bed of Calcein-AM and Amvisc Plus by an eye bank technician. Five grafts were removed and stained as a scroll, and 5 grafts were trifolded with the endothelium in before transfer. Photographs were acquired with an inverted fluorescence microscope, and image segmentation was performed. A t test was conducted to compare differences in endothelial cell loss across groups. RESULTS: Mean cell loss in the scroll group was 18.5% [95% confidence interval (CI): 15.2%-21.9%] compared with 7.6% of the trifolded group (95% CI: 1.7%-13.5%). A 2-tailed t test indicated decreased cell loss in the trifolded group (P = 0.013). CONCLUSIONS: Despite additional manipulation of the graft, trifolding of Descemet membrane and endothelium may be performed by an eye bank technician without significantly increased cell loss relative to graft preparation as a scroll.