Clinical pre-consultation, where patients share health information prior to an appointment, can offer better patient-centered care by freeing time for meaningful patient–physician conversations. Conversational agents powered by large language models (LLMs) can automate this process to make it more scalable and consistent, but doing so can cause information overload that exacerbates physicians’ workload as they spend time parsing through data. This paper examines the opportunities and challenges of using pre-consultation agents to mediate information transfer between patients and physicians, with the aim of producing clinically useful yet patient-driven summaries that capture their histories and concerns. Through sessions with both patients and physicians, we show that automatically generated pre-consultation summaries can increase patients’ confidence and sense of control over their health information while fostering a more collaborative dynamic. We conclude with design recommendations for integrating pre-consultation agents into clinical workflows.

Pre-consultation serves as a critical information exchange between healthcare providers and patients, streamlining visits and supporting patient-centered care. Human-led pre-consultations offer many benefits, yet they require significant time and energy from clinical staff. In this work, we identify design goals for pre-consultation chatbots given their potential to carry out human-like conversations and autonomously adapt their line of questioning. We conducted a study with 33 walk-in clinic patients to elicit design considerations for pre-consultation chatbots. Participants were exposed to one of two study conditions: an LLM-powered AI agent and a Wizard-of-Oz agent simulated by medical professionals. Our study found that both conditions were equally well-received and demonstrated comparable conversational capabilities. However, the extent of the follow-up questions and the amount of empathy impacted the chatbot’s perceived thoroughness and sincerity. Patients also highlighted the importance of setting expectations for the chatbot before and after the pre-consultation experience.

Menstrual trackers currently lack the affordances required to help individuals achieve their goals beyond menstrual event predictions and symptom logging. Taking an initial step towards this aspiration, we propose, validate, and refine five functional design requirements for future interface designs that facilitate menstrual data exploration. We interviewed 30 individuals who menstruate and collected their feedback on the practical application of these requirements. To elicit ideas and impressions, we designed two proof-of-concept interfaces to use as design probes with similar core functionalities but different presentations of phase timing predictions and signal arrangement. Our analysis revealed participants’ feedback regarding the presentation of predictions for menstrual-related events, the visualization of future signal patterns, personalization abilities for viewing signals relevant to their menstrual experience, the availability of resources to understand the underlying biological connections between signals, and the ability to compare multiple cycles side-by-side with context.

Medical consultations over synchronous text-based platforms are becoming increasingly popular for virtual care, yet little is known about how physicians translate their training to this healthcare medium. We report the constraints, workarounds, and opportunities highlighted by eight primary care physicians who used such a platform in simulated medical scenarios with standardized patients. We found that due to the perceived inefficiency of communicating over text, the physicians made subconscious use of double-barreled questions and action multiplexing to streamline the conversation. In addition, the physicians overcame the lack of missing verbal and visual cues by adding explicit messages to convey empathy and active listening. We also identify several affordances of text-based platforms, such as the ability for users to reference the conversation history and for patients to feel a sense of privacy during sensitive disclosure. From these findings, we propose design opportunities for how future synchronous text-based platforms can better support medical consultations.

Recently, digital scribe systems have been gaining popularity as a possible work-around solution to the Electronic Medical Record (EMR) documentation burden that affects many physicians. The proposed system would automate the clinical summary physicians take by capturing and extracting the patient-physician conversation during the consultation. While promising in concept, how this system would apply to real-world use and its limitations are still not well understood. To examine these issues, we designed a digital scribe prototype to generate notes of different qualities ranging from the reality of current state-of-the-art technology to the potential of future implementations. We conducted a ”Wizard of Oz” study with 24 primary care physicians using our digital scribe prototype in 4 simulated medical encounters followed by a semi-structured interview. This exploratory study provides an understanding of physicians’ interaction with digitally scribed notes, their perceptions on note quality, their perceived workflow impact and several directions for improvements.

How neuronal diversity emerges from complex patterns of gene expression remains poorly understood. Here we present an approach to understand electrophysiological diversity through gene expression by integrating pooled- and single-cell transcriptomics with intracellular electrophysiology. Using neuroinformatics methods, we compiled a brain-wide dataset of 34 neuron types with paired gene expression and intrinsic electrophysiological features from publically accessible sources, the largest such collection to date. We identified 420 genes whose expression levels significantly correlated with variability in one or more of 11 physiological parameters. We next trained statistical models to infer cellular features from multivariate gene expression patterns. Such models were predictive of gene-electrophysiological relationships in an independent collection of 12 visual cortex cell types from the Allen Institute, suggesting that these correlations might reflect general principles relating expression patterns to phenotypic diversity across very different cell types. Many associations reported here have the potential to provide new insights into how neurons generate functional diversity, and correlations of ion channel genes like Gabrd and Scn1a (Nav1.1) with resting potential and spiking frequency are consistent with known causal mechanisms. Our work highlights the promise and inherent challenges in using cell type-specific transcriptomics to understand the mechanistic origins of neuronal diversity.