An interpretable RL framework for pre-deployment modeling in ICU hypotension management.

Computational methods from reinforcement learning have shown promise in inferring treatment strategies for hypotension management and other clinical decision-making challenges. Unfortunately, the resulting models are often difficult for clinicians to interpret, making clinical inspection and validation of these computationally derived strategies challenging in advance of deployment. In this work, we develop a general framework for identifying succinct sets of clinical contexts in which clinicians make very different treatment choices, tracing the effects of those choices, and inferring a set of recommendations for those specific contexts. By focusing on these few key decision points, our framework produces succinct, interpretable treatment strategies that can each be easily visualized and verified by clinical experts. This interrogation process allows clinicians to leverage the model's use of historical data in tandem with their own expertise to determine which recommendations are worth investigating further e.g. at the bedside. We demonstrate the value of this approach via application to hypotension management in the ICU, an area with critical implications for patient outcomes that lacks data-driven individualized treatment strategies; that said, our framework has broad implications on how to use computational methods to assist with decision-making challenges on a wide range of clinical domains.

A blueprint for success: Using an implementation framework to create a medication history technician pilot program.

BACKGROUND: Medication discrepancies at transitions of care may compromise patient safety. Trained pharmacy technicians can reduce harmful medication discrepancies at transitions of care by collecting medication histories. OBJECTIVE: We describe how to create a program integrating medication history technicians (MHTs) into the hospital discharge process using implementation science. PRACTICE DESCRIPTION: We created our MHT program at a Veterans Affairs (VA) hospital. PRACTICE INNOVATION: We used an evidence-based framework and implementation science to tailor our MHT program to meet local stakeholder needs. EVALUATION METHODS: We completed a literature review and review of current discharge practices. Then, we completed a workflow pilot, a needs assessment, and semistructured interviews with pharmacy technicians and pharmacists. We integrated these findings to identify barriers of MHT program implementation. Finally, we mapped these barriers to implementation strategies to create an MHT program implementation blueprint. RESULTS: The literature review and review of current discharge practices revealed opportunities for our program to reduce medication discrepancies. We applied these findings to our proof-of-concept workflow pilot, which reduced medication discrepancy rates at discharge. When we explored barriers in the needs assessment, we learned that 4 of 6 pharmacy technicians had some training conducting medication histories, but 5 of 6 requested additional training for the new MHT role. We explored these and additional barriers in semistructured interviews. Four themes emerged: elements of pharmacy technician training, challenges to implementation, program logistics and workflow, and pharmacy technician self-efficacy. We mapped barriers to implementation strategies to create an MHT program implementation blueprint, including developing pharmacy technician training materials, modifying our workflow, creating program evaluation materials, and strategizing how to overcome anticipated and current implementation barriers. CONCLUSIONS: We used implementation science to create a tailored MHT program. Others may adapt our implementation blueprint to fit local stakeholder needs.

Atomically Sharp Crack Tips in Monolayer MoS2 and Their Enhanced Toughness by Vacancy Defects.

We combine in situ transmission electron microscopy and large-scale molecular dynamics simulations to investigate brittle fracture in 2D monolayer MoS2, revealing that cracks propagate with a tip of atomic sharpness through the preferential direction with least energy release. We find that sparse vacancy defects cause crack deflections, while increasing defect density shifts the fracture mechanism from brittle to ductile by the migration of vacancies in the strain fields into networks. The fracture toughness of defective MoS2 is found to exceed that of graphene due to interactions between the atomically sharp crack tips and vacancy clusters during propagation. These results show that monolayer 2D materials are ideal for revealing fundamental aspects of fracture mechanics not previously possible with thicker materials, similar to studies of dislocation behavior in 2D materials.

Liver transplantation for metabolic liver diseases.

Liver transplantation has become an accepted treatment for several metabolic liver diseases. With advances in organ transplantation and immunosuppressive strategies, survival rates following liver transplantation are generally excellent. When the primary metabolic defect is hepatic in origin, liver transplantation not only replaces the dysfunctional organ but also cures the underlying metabolic defect. For conditions in which the primary metabolic defect is extrahepatic, liver transplantation is usually performed for hepatic complications, although disease recurrence may occur. This article reviews common metabolic liver diseases treated with liver transplantation in the adult population.