Search for: All records

Abstract ObjectiveThe use of electronic health records (EHRs) for clinical risk prediction is on the rise. However, in many practical settings, the limited availability of task-specific EHR data can restrict the application of standard machine learning pipelines. In this study, we investigate the potential of leveraging language models (LMs) as a means to incorporate supplementary domain knowledge for improving the performance of various EHR-based risk prediction tasks. MethodsWe propose two novel LM-based methods, namely “LLaMA2-EHR” and “Sent-e-Med.” Our focus is on utilizing the textual descriptions within structured EHRs to make risk predictions about future diagnoses. We conduct a comprehensive comparison with previous approaches across various data types and sizes. ResultsExperiments across 6 different methods and 3 separate risk prediction tasks reveal that employing LMs to represent structured EHRs, such as diagnostic histories, results in significant performance improvements when evaluated using standard metrics such as area under the receiver operating characteristic (ROC) curve and precision-recall (PR) curve. Additionally, they offer benefits such as few-shot learning, the ability to handle previously unseen medical concepts, and adaptability to various medical vocabularies. However, it is noteworthy that outcomes may exhibit sensitivity to a specific prompt. ConclusionLMs encompass extensive embedded knowledge, making them valuable for the analysis of EHRs in the context of risk prediction. Nevertheless, it is important to exercise caution in their application, as ongoing safety concerns related to LMs persist and require continuous consideration. 

Abstract To navigate through the environment, humans must be able to measure both the distance traveled in space, and the interval elapsed in time. Yet, how the brain holds both of these metrics simultaneously is less well known. One possibility is that participants measure how far and how long they have traveled relative to a known reference point. To measure this, we had human participants (n = 24) perform a distance estimation task in a virtual environment in which they were cued to attend to either the spatial or temporal interval traveled while responses were measured with multiband fMRI. We observed that both dimensions evoked similar frontoparietal networks, yet with a striking rostrocaudal dissociation between temporal and spatial estimation. Multivariate classifiers trained on each dimension were further able to predict the temporal or spatial interval traveled, with centers of activation within the SMA and retrosplenial cortex for time and space, respectively. Furthermore, a cross-classification approach revealed the right supramarginal gyrus and occipital place area as regions capable of decoding the general magnitude of the traveled distance. Altogether, our findings suggest the brain uses separate systems for tracking spatial and temporal distances, which are combined together along with dimension-nonspecific estimates. 

Abstract There have been significant advances in biosignal extraction techniques to drive external biomechatronic devices or to use as inputs to sophisticated human machine interfaces. The control signals are typically derived from biological signals such as myoelectric measurements made either from the surface of the skin or subcutaneously. Other biosignal sensing modalities are emerging. With improvements in sensing modalities and control algorithms, it is becoming possible to robustly control the target position of an end-effector. It remains largely unknown to what extent these improvements can lead to naturalistic human-like movement. In this paper, we sought to answer this question. We utilized a sensing paradigm called sonomyography based on continuous ultrasound imaging of forearm muscles. Unlike myoelectric control strategies which measure electrical activation and use the extracted signals to determine the velocity of an end-effector; sonomyography measures muscle deformation directly with ultrasound and uses the extracted signals to proportionally control the position of an end-effector. Previously, we showed that users were able to accurately and precisely perform a virtual target acquisition task using sonomyography. In this work, we investigate the time course of the control trajectories derived from sonomyography. We show that the time course of the sonomyography-derived trajectories that users take to reach virtual targets reflect the trajectories shown to be typical for kinematic characteristics observed in biological limbs. Specifically, during a target acquisition task, the velocity profiles followed a minimum jerk trajectory shown for point-to-point arm reaching movements, with similar time to target. In addition, the trajectories based on ultrasound imaging result in a systematic delay and scaling of peak movement velocity as the movement distance increased. We believe this is the first evaluation of similarities in control policies in coordinated movements in jointed limbs, and those based on position control signals extracted at the individual muscle level. These results have strong implications for the future development of control paradigms for assistive technologies. 

Even in our highly interconnected modern world, geographic factors play an important role in human social connections. Similarly, social relationships influence how and where we travel, and how we think about our spatial world. Here, we review the growing body of neuroscience research that is revealing multiple interactions between social and spatial processes in both humans and non-human animals. We review research on the cognitive and neural representation of spatial and social information, and highlight recent findings suggesting that underlying mechanisms might be common to both. We discuss how spatial factors can influence social behaviour, and how social concepts modify representations of space. In so doing, this review elucidates not only how neural representations of social and spatial information interact but also similarities in how the brain represents and operates on analogous information about its social and spatial surroundings. This article is part of the theme issue ‘The spatial–social interface: a theoretical and empirical integration’. 

Parents’ alcohol use may influence adolescent substance use and substance use intentions. Prior research has linked adolescents’ emotion reactivity with parental drinking behaviors and adolescent substance use. The present study investigated whether sub-clinical maternal alcohol use relates to adolescent neural emotion reactivity and substance use intentions in early adolescence. Early adolescents ( N = 70) viewed emotional images during a fMRI scan and completed a questionnaire about substance use intentions. Their mothers reported past 30-day alcohol use. Results showed that greater frequency of maternal alcohol use predicted adolescents’ substance use intentions. In addition, maternal alcohol use predicted adolescent blunted responses to positive emotional images in the ventromedial prefrontal cortex (vmPFC) and bilateral anterior cingulate cortex (ACC). There was no relationship between neural emotion reactivity and adolescent substance use intentions. Findings suggest that parental alcohol use may relate to adolescent’s development of reward and positive emotion processing systems, even at sub-clinical levels of drinking. 

Rehabilitation from musculoskeletal injuries focuses on reestablishing and monitoring muscle activation patterns to accurately produce force. The aim of this study is to explore the use of a novel low-powered wearable distributed Simultaneous Musculoskeletal Assessment with Real-Time Ultrasound (SMART-US) device to predict force during an isometric squat task. Participants (N = 5) performed maximum isometric squats under two medical imaging techniques; clinical musculoskeletal motion mode (m-mode) ultrasound on the dominant vastus lateralis and SMART-US sensors placed on the rectus femoris, vastus lateralis, medial hamstring, and vastus medialis. Ultrasound features were extracted, and a linear ridge regression model was used to predict ground reaction force. The performance of ultrasound features to predict measured force was tested using either the Clinical M-mode, SMART-US sensors on the vastus lateralis (SMART-US: VL), rectus femoris (SMART-US: RF), medial hamstring (SMART-US: MH), and vastus medialis (SMART-US: VMO) or utilized all four SMART-US sensors (Distributed SMART-US). Model training showed that the Clinical M-mode and the Distributed SMART-US model were both significantly different from the SMART-US: VL, SMART-US: MH, SMART-US: RF, and SMART-US: VMO models (p < 0.05). Model validation showed that the Distributed SMART-US model had an R2 of 0.80 ± 0.04 and was significantly different from SMART-US: VL but not from the Clinical M-mode model. In conclusion, a novel wearable distributed SMART-US system can predict ground reaction force using machine learning, demonstrating the feasibility of wearable ultrasound imaging for ground reaction force estimation. 

Error monitoring is an essential human ability underlying learning and metacognition. In the time domain, humans possess a remarkable ability to learn and adapt to temporal intervals, yet the neural mechanisms underlying this are not well understood. Recently, we demonstrated that humans exhibit improvements in sensorimotor time estimates when given the chance to incorporate feedback from a previous trial (Bader and Wiener, 2021), suggesting that humans are metacognitively aware of their own timing errors. To test the neural basis of this metacognitive ability, human participants of both sexes underwent fMRI while they performed a visual temporal reproduction task with randomized suprasecond intervals (1.5-6s). Crucially, each trial was repeated following feedback, allowing a “re-do” to learn from the successes or errors in the initial trial. Behaviorally, we replicated our previous finding that subjects improve their performance on re-do trials despite the feedback being temporally uninformative (i.e. early or late). For neuroimaging, we observed a dissociation between estimating and reproducing time intervals, with the former more likely to engage regions associated with the default mode network (DMN), including the superior frontal gyri, precuneus, and posterior cingulate, whereas the latter activated regions associated traditionally with the “Timing Network” (TN), including the supplementary motor area (SMA), precentral gyrus, and right supramarginal gyrus. Notably, greater DMN involvement was observed in Re-do trials. Further, the extent of the DMN was greater on re-do trials, whereas for the TN it was more constrained. Finally, Task-based connectivity between these networks demonstrated higher inter-network correlation on initial trials, but primarily when estimating trials, whereas on re-do trials communication was higher during reproduction. Overall, these results suggest the DMN and TN work in concert to mediate subjective awareness of one’s sense of time for the purpose of improving timing performance. Significance StatementA finely tuned sense of time perception is imperative for everyday motor actions (e.g., hitting a baseball). Timing self-regulation requires correct assessment and updating duration estimates if necessary. Using a modified version of a classical task of time measurement, we explored the neural regions involved in error detection, time awareness, and learning to time. Reinforcing the role of the SMA in measuring temporal information and providing evidence of co-activation with the DMN, this study demonstrates that the brain overlays sensorimotor timing with a metacognitive awareness of its passage.