OBJECTIVES: To quantify utilization and impact on documentation time of a large language model-powered ambient artificial intelligence (AI) scribe. MATERIALS AND METHODS: This prospective quality improvement study was conducted at a large academic medical center with 45 physicians from 8 ambulatory disciplines over 3 months. Utilization and documentation times were derived from electronic health record (EHR) use measures. RESULTS: The ambient AI scribe was utilized in 9629 of 17 428 encounters (55.25%) with significant interuser heterogeneity. Compared to baseline, median time per note reduced significantly by 0.57 minutes. Median daily documentation, afterhours, and total EHR time also decreased significantly by 6.89, 5.17, and 19.95 minutes/day, respectively. DISCUSSION: An early pilot of an ambient AI scribe demonstrated robust utilization and reduced time spent on documentation and in the EHR. There was notable individual-level heterogeneity. CONCLUSION: Large language model-powered ambient AI scribes may reduce documentation burden. Further studies are needed to identify which users benefit most from current technology and how future iterations can support a broader audience.

OBJECTIVE: This study evaluates the pilot implementation of ambient AI scribe technology to assess physician perspectives on usability and the impact on physician burden and burnout. MATERIALS AND METHODS: This prospective quality improvement study was conducted at Stanford Health Care with 48 physicians over a 3-month period. Outcome measures included burden, burnout, usability, and perceived time savings. RESULTS: Paired survey analysis (n = 38) revealed large statistically significant reductions in task load (-24.42, p <.001) and burnout (-1.94, p <.001), and moderate statistically significant improvements in usability scores (+10.9, p <.001). Post-survey responses (n = 46) indicated favorable utility with improved perceptions of efficiency, documentation quality, and ease of use. DISCUSSION: In one of the first pilot implementations of ambient AI scribe technology, improvements in physician task load, burnout, and usability were demonstrated. CONCLUSION: Ambient AI scribes like DAX Copilot may enhance clinical workflows. Further research is needed to optimize widespread implementation and evaluate long-term impacts.

Autonomous agents that must exhibit flexible and broad capabilities will need to be equipped with large repertoires of skills. Defining each skill with a manually-designed reward function limits this repertoire and imposes a manual engineering burden. Self-supervised agents that set their own goals can automate this process, but designing appropriate goal setting objectives can be difficult, and often involves heuristic design decisions. In this paper, we propose a formal exploration objective for goal-reaching policies that maximizes state coverage. We show that this objective is equivalent to maximizing goal reaching performance together with the entropy of the goal distribution, where goals correspond to full state observations. To instantiate this principle, we present an algorithm called Skew-Fit for learning a maximum-entropy goal distributions. We prove that, under regularity conditions, Skew-Fit converges to a uniform distribution over the set of valid states, even when we do not know this set beforehand. Our experiments show that combining Skew-Fit for learning goal distributions with existing goal-reaching methods outperforms a variety of prior methods on open-sourced visual goal-reaching tasks. Moreover, we demonstrate that Skew-Fit enables a real-world robot to learn to open a door, entirely from scratch, from pixels, and without any manually-designed reward function.

Planning methods can solve temporally extended sequential decision making problems by composing simple behaviors. However, planning requires suitable abstractions for the states and transitions, which typically need to be designed by hand. In contrast, reinforcement learning (RL) can acquire behaviors from low-level inputs directly, but struggles with temporally extended tasks. Can we utilize reinforcement learning to automatically form the abstractions needed for planning, thus obtaining the best of both approaches? We show that goal-conditioned policies learned with RL can be incorporated into planning, such that a planner can focus on which states to reach, rather than how those states are reached. However, with complex state observations such as images, not all inputs represent valid states. We therefore also propose using a latent variable model to compactly represent the set of valid states for the planner, such that the policies provide an abstraction of actions, and the latent variable model provides an abstraction of states. We compare our method with planning-based and model-free methods and find that our method significantly outperforms prior work when evaluated on image-based tasks that require non-greedy, multi-staged behavior.

For an autonomous agent to fulfill a wide range of user-specified goals at test time, it must be able to learn broadly applicable and general-purpose skill repertoires. Furthermore, to provide the requisite level of generality, these skills must handle raw sensory input such as images. In this paper, we propose an algorithm that acquires such general-purpose skills by combining unsupervised representation learning and reinforcement learning of goal-conditioned policies. Since the particular goals that might be required at test-time are not known in advance, the agent performs a self-supervised "practice" phase where it imagines goals and attempts to achieve them. We learn a visual representation with three distinct purposes: sampling goals for self-supervised practice, providing a structured transformation of raw sensory inputs, and computing a reward signal for goal reaching. We also propose a retroactive goal relabeling scheme to further improve the sample-efficiency of our method. Our off-policy algorithm is efficient enough to learn policies that operate on raw image observations and goals in a real-world physical system, and substantially outperforms prior techniques.

SUMMARY: Gene fusions are being discovered at an increasing rate using massively parallel sequencing technologies. Prioritization of cancer fusion drivers for validation cannot be performed using traditional single-gene based methods because fusions involve portions of two partner genes. To address this problem, we propose a novel network analysis method called fusion centrality that is specifically tailored for prioritizing gene fusions. We first propose a domain-based fusion model built on the theory of exon/domain shuffling. The model leads to a hypothesis that a fusion is more likely to be an oncogenic driver if its partner genes act like hubs in a network because the fusion mutation can deregulate normal functions of many other genes and their pathways. The hypothesis is supported by the observation that for most known cancer fusion genes, at least one of the fusion partners appears to be a hub in a network, and even for many fusions both partners appear to be hubs. Based on this model, we construct fusion centrality, a multi-gene-based network metric, and use it to score fusion drivers. We show that the fusion centrality outperforms other single gene-based methods. Specifically, the method successfully predicts most of 38 newly discovered fusions that had validated oncogenic importance. To our best knowledge, this is the first network-based approach for identifying fusion drivers. AVAILABILITY: Matlab code implementing the fusion centrality method is available upon request from the corresponding authors.

The increased bandwidth in the Internet puts great demands on network routers; for example, to route minimum sized Gigabit Ethernet packets, an IP router must process about 1.5/spl times/10/sup 6/ packets per second per port. Using the "rule-of-thumb" that it takes roughly 1000 packets per second for every 10/sup 6/ bits per second of line rate, an OC-192 line requires 10/spl times/10/sup 6/ routing lookups per second; well above current router capabilities. One limitation of router performance is the route lookup mechanism. IP routing requires that a router perform a longest-prefix-match address lookup for each incoming datagram in order to determine the datagram's next hop. We present a route lookup mechanism that when implemented in a pipelined fashion in hardware, can achieve one route lookup every memory access. With current 50 ns DRAM, this corresponds to approximately 20/spl times/10/sup 6/ packets per second; much faster than current commercially available routing lookup schemes. We also present novel schemes for performing quick updates to the forwarding table in hardware. We demonstrate using real routing update patterns that the routing tables can be updated with negligible overhead to the central processor.

Recently there has been much interest in combining the speed of layer-2 switching with the features of layer-3 routing. This has been prompted by numerous proposals, including: IP Switching [1], Tag Switching [2], ARIS [3], CSR [4], and IP over ATM [5]. In this paper, we study IP Switching and evaluate the performance claims made by Newman et al in [1] and [6]. In particular, using ten network traces, we study how well IP Switching performs with traffic found in campus, corporate, and Internet Service Provider (ISP) environments. Our main finding is that IP Switching will lead to a high proportion of datagrams that are switched; over 75% in all of the environments we studied. We also investigate the effects that different flow classifiers and various timer values have on performance, and note that some choices can result in a large VC space requirement. Finally, we present recommendations for the flow classifier and timer values, as a function of the VC space of the switch and the network environment being served.