Why do users override alerts? Utilizing large language model to summarize comments and optimize clinical decision support.

OBJECTIVES: To evaluate the capability of using generative artificial intelligence (AI) in summarizing alert comments and to determine if the AI-generated summary could be used to improve clinical decision support (CDS) alerts. MATERIALS AND METHODS: We extracted user comments to alerts generated from September 1, 2022 to September 1, 2023 at Vanderbilt University Medical Center. For a subset of 8 alerts, comment summaries were generated independently by 2 physicians and then separately by GPT-4. We surveyed 5 CDS experts to rate the human-generated and AI-generated summaries on a scale from 1 (strongly disagree) to 5 (strongly agree) for the 4 metrics: clarity, completeness, accuracy, and usefulness. RESULTS: Five CDS experts participated in the survey. A total of 16 human-generated summaries and 8 AI-generated summaries were assessed. Among the top 8 rated summaries, five were generated by GPT-4. AI-generated summaries demonstrated high levels of clarity, accuracy, and usefulness, similar to the human-generated summaries. Moreover, AI-generated summaries exhibited significantly higher completeness and usefulness compared to the human-generated summaries (AI: 3.4 ± 1.2, human: 2.7 ± 1.2, P = .001). CONCLUSION: End-user comments provide clinicians' immediate feedback to CDS alerts and can serve as a direct and valuable data resource for improving CDS delivery. Traditionally, these comments may not be considered in the CDS review process due to their unstructured nature, large volume, and the presence of redundant or irrelevant content. Our study demonstrates that GPT-4 is capable of distilling these comments into summaries characterized by high clarity, accuracy, and completeness. AI-generated summaries are equivalent and potentially better than human-generated summaries. These AI-generated summaries could provide CDS experts with a novel means of reviewing user comments to rapidly optimize CDS alerts both online and offline.

Ketogenic-Mimicking Diet as a Therapeutic Modality for Bipolar Disorder: Biomechanistic Rationale and Protocol for a Pilot Clinical Trial.

There is growing interest in the investigation of ketogenic diets as a potential therapy for bipolar disorder. The overlapping pharmacotherapies utilized for both bipolar disorder and seizures suggest that a mechanistic overlap may exist between these conditions, with fasting and the ketogenic diet representing the most time-proven therapies for seizure control. Recently, preliminary evidence has begun to emerge supporting a potential role for ketogenic diets in treating bipolar disorder. Notably, some patients may struggle to initiate a strict diet in the midst of a mood episode or significant life stressors. The key question addressed by this pilot clinical trial protocol is if benefits can be achieved with a less restrictive diet, as this would allow such an intervention to be accessible for more patients. Recent development of so-called ketone esters, that once ingested is converted to natural ketone bodies, combined with low glycemic index dietary changes has the potential to mimic two foundational components of therapeutic ketosis: high levels of ketones and minimal spiking of glucose/insulin. This pilot clinical trial protocol thus aims to investigate the effect of a 'ketogenic-mimicking diet' (combining supplementation of ketone esters with a low glycemic index dietary intervention) on neural network stability, mood, and biomarker outcomes in the setting of bipolar disorder. Positive findings obtained via this pilot clinical trial protocol may support future target engagement studies of ketogenic-mimicking diets or related ketogenic interventions. A lack of positive findings, in contrast, may justify a focus on more strict dietary interventions for future research.

Multiphoton Bleaching of Red Fluorescent Proteins and the Ways to Reduce It.

Red fluorescent proteins and biosensors built upon them are potentially beneficial for two-photon laser microscopy (TPLM) because they can image deeper layers of tissue, compared to green fluorescent proteins. However, some publications report on their very fast photobleaching, especially upon excitation at 750-800 nm. Here we study the multiphoton bleaching properties of mCherry, mPlum, tdTomato, and jREX-GECO1, measuring power dependences of photobleaching rates K at different excitation wavelengths across the whole two-photon absorption spectrum. Although all these proteins contain the chromophore with the same chemical structure, the mechanisms of their multiphoton bleaching are different. The number of photons required to initiate a photochemical reaction varies, depending on wavelength and power, from 2 (all four proteins) to 3 (jREX-GECO1) to 4 (mCherry, mPlum, tdTomato), and even up to 8 (tdTomato). We found that at sufficiently low excitation power P, the rate K often follows a quadratic power dependence, that turns into higher order dependence (K~Pα with α > 2) when the power surpasses a particular threshold P*. An optimum intensity for TPLM is close to the P*, because it provides the highest signal-to-background ratio and any further reduction of laser intensity would not improve the fluorescence/bleaching rate ratio. Additionally, one should avoid using wavelengths shorter than a particular threshold to avoid fast bleaching due to multiphoton ionization.

Colossal Magnetoresistance without Mixed Valence in a Layered Phosphide Crystal.

Materials with strong magnetoresistive responses are the backbone of spintronic technology, magnetic sensors, and hard drives. Among them, manganese oxides with a mixed valence and a cubic perovskite structure stand out due to their colossal magnetoresistance (CMR). A double exchange interaction underlies the CMR in manganates, whereby charge transport is enhanced when the spins on neighboring Mn3+ and Mn4+ ions are parallel. Prior efforts to find different materials or mechanisms for CMR resulted in a much smaller effect. Here an enormous CMR at low temperatures in EuCd2 P2 without manganese, oxygen, mixed valence, or cubic perovskite structure is shown. EuCd2 P2 has a layered trigonal lattice and exhibits antiferromagnetic ordering at 11 K. The magnitude of CMR (104 %) in as-grown crystals of EuCd2 P2 rivals the magnitude in optimized thin films of manganates. The magnetization, transport, and synchrotron X-ray data suggest that strong magnetic fluctuations are responsible for this phenomenon. The realization of CMR at low temperatures without heterovalency leads to a new regime for materials and technologies related to antiferromagnetic spintronics.

High throughput instrument to screen fluorescent proteins under two-photon excitation.

Two-photon microscopy together with fluorescent proteins and fluorescent protein-based biosensors are commonly used tools in neuroscience. To enhance their experimental scope, it is important to optimize fluorescent proteins for two-photon excitation. Directed evolution of fluorescent proteins under one-photon excitation is common, but many one-photon properties do not correlate with two-photon properties. A simple system for expressing fluorescent protein mutants is E. coli colonies on an agar plate. The small focal volume of two-photon excitation makes creating a high throughput screen in this system a challenge for a conventional point-scanning approach. We present an instrument and accompanying software that solves this challenge by selectively scanning each colony based on a colony map captured under one-photon excitation. This instrument, called the GIZMO, can measure the two-photon excited fluorescence of 10,000 E. coli colonies in 7 hours. We show that the GIZMO can be used to evolve a fluorescent protein under two-photon excitation.