INTRODUCTION: Existing clinical prediction algorithms mostly leverage small cohorts of structured data (e.g., medical imaging or laboratory data). However, large language models have demonstrated the ability to utilize unstructured data to outperform other machine learning approaches given sufficient data. Training large language models on unstructured clinical notes offers a possible alternative to structured data algorithm development in clinical tasks. METHODS: An unlabeled dataset of over seven million unstructured clinical notes (e.g., radiology reports and patient histories) was collected from four hospitals within the NYU Langone Health (NYULH) system and utilized to pre-train a bidirectional encoder representation with transformer (BERT) model. This model was further fine-tuned using a labeled dataset of discharge summaries to predict 30-day all-cause readmission. The resulting model, termed NYUTron, was assessed using a held-out retrospective cohort of patients from June to December 2021. RESULTS: Over the period of the retrospective study, there were a total 1,072 neurosurgery patients. The BERT model achieved an area under the receiver operating curve of 0.7883, a recall of 95.1% at a precision of 27.3%, and an accuracy of 82.1%. CONCLUSIONS: This study demonstrates how large language models and unstructured clinical notes can be used to provide information to physicians on clinical tasks with a flexible framework that is amenable to modification for other clinical tasks.
In-Vessel Retention (IVR) is one of the most important severe accident mitigation strategies of the third generation passive Nuclear Power Plants (NPP). It is intended to demonstrate that in the case of a core melt, the structural integrity of the Reactor Pressure Vessel (RPV) is assured such that there is no leakage of radioactive debris from the RPV. This paper studied the IVR issue using Finite Element Analyses (FEA). Firstly, the tension and creep testing for the SA-508 Gr.3 Cl.1 material in the temperature range of 25°C to 1000°C were performed. Secondly, a FEA model of the RPV lower head was built. Based on the assumption of ideally elastic-plastic material properties derived from the tension testing data, limit analyses were performed under both the thermal and the thermal plus pressure loading conditions where the load bearing capacity was investigated by tracking the propagation of plastic region as a function of pressure increment. Finally, the ideal elastic-plastic material properties incorporating the creep effect are developed from the 100hr isochronous stress-strain curves, limit analyses are carried out as the second step above. The allowable pressures at 0 hr and 100 hr are obtained. This research provides an alternative approach for the structural integrity evaluation for RPV under IVR condition.
Esketamine (ESK), an intravenous anesthetic, exerts antidepressant effects; however, the antidepression mechanism is not clear. The aim of this study was to explore whether microglial cannabinoid type 2 (CB2) receptor and protein kinase C (PKC) are involved in the antidepressant effects of ESK.
The tribological properties of metal-CuS/MoS2 composite coatings were improved by selecting the metal materials and optimizing the heat treatment process. The composite coatings were prepared by magnetron cosputtering of copper or aluminum metals, CuS, and MoS2. Then, the prepared Cu-CuS/MoS2 and Al-CuS/MoS2 were subjected to vacuum heat treatment at 220, 320, and 420 °C. The structure and tribological properties of the composite coatings were analyzed by scanning electron microscope, energy spectrum, x-ray diffraction, Raman spectrum, and multifunction friction tester. The results showed an island-like growth and a dense noncylindrical cross-sectional morphology of the metal-CuS/MoS2 composite coating. Metal- and CuS-doped MoS2 solid solution is a practical component in composite coatings. The wear scar of the composite coatings showed different color layers at different annealing temperatures due to composition deviation. After heat treatment at 320 °C, the friction coefficient of Cu-CuS/MoS2 coating was reduced to a minimum of 0.08. After heat treatment at 420 °C, the Al-CuS/MoS2 coating showed three different color layers of wear scar, and good tribological properties were obtained. The friction coefficient of the Al-CuS/MoS2 coating was as low as 0.06, and the adhesion of the Al-CuS/MoS2 coating was expressed in the HF1 level. The results show that the vacuum heat treatment can stimulate the action of CuS by forming a solid solution with MoS2, complementing the missing S atom during the sputtering deposition of MoS2 and improving the lubrication of the composite coating. In addition, although the action mechanisms are different, Cu and Al doping are all conducive to obtaining composite coatings with better tribological properties.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
