Medical Information Extraction With NLP-Powered QABots: A Real-World Scenario.

The advent of computerized medical recording systems in healthcare facilities has made data retrieval tasks easier, compared to manual recording. Nevertheless, the potential of the information contained within medical records remains largely untapped, mostly due to the time and effort required to extract data from unstructured documents. Natural Language Processing (NLP) represents a promising solution to this challenge, as it enables the use of automated text-mining tools for clinical practitioners. In this work, we present the architecture of the Virtual Dementia Institute (IVD), a consortium of sixteen Italian hospitals, using the NLP Extraction and Management Tool (NEMT), a (semi-) automated end-to-end pipeline that extracts relevant information from clinical documents and stores it in a centralized REDCap database. After defining a common Case Report Form (CRF) across the IVD hospitals, we implemented NEMT, the core of which is a Question Answering Bot (QABot) based on a modern NLP model. This QABot is fine-tuned on thousands of examples from IVD centers. Detailed descriptions of the process to define a common minimum dataset, Inter-Annotator Agreement calculated on clinical documents, and NEMT results are provided. The best QABot performance show an Exact Match score (EM) of 78.1%, a F1-score of 84.7%, a Lenient Accuracy (LAcc) of 0.834, and a Mean Reciprocal Rank (MRR) of 0.810. EM and F1 scores outperform the same metrics obtained with ChatGPTv3.5 (68.9% and 52.5%, respectively). With NEMT the IVD has been able to populate a database that will contain data from thousands of Italian patients, all screened with the same procedure. NEMT represents an efficient tool that paves the way for medical information extraction and exploitation for new research studies.

Dissecting the Clinical Heterogeneity and Genotype-Phenotype Correlations of MAPT Mutations: A Systematic Review.

BACKGROUND: Microtubule-associated protein tau (MAPT) mutations are one of the main causes of genetic Frontotemporal dementia (FTD) and are characterised by high clinical heterogeneity. A behavioural variant of FTD is the principal phenotype, but other rarer phenotypes are described, mostly reported as single cases. In this review, we provide an overview of the clinical phenotypes associated with MAPT mutations in order to define their characteristics and explore genotype-phenotype correlations. METHODS: We performed systematic bibliographic research on the Pubmed database, focusing on articles published between 1998 and 2022. We analysed the clinical phenotype of 177 patients carrying MAPT mutations, focusing on the rarest ones. We performed a narrative synthesis of the results. RESULTS: Regarding language phenotypes, the most frequent were the non-fluent variant and the semantic variant of Primary Progressive Aphasia (nfvPPA, svPPA), approximately in the same proportion. Almost 20% of the whole group of patients present a clinical phenotype belonging to the corticobasal syndrome-progressive supranuclear palsy (CBS-PSP) spectrum. While no clear genotype-phenotype correlation could be identified, some mutations were associated with a specific phenotype, while others gave origin to multiple clinical pictures and mixed phenotypes. CONCLUSIONS: A high clinical heterogeneity exists in FTD associated with MAPT mutations without a clear phenotype-genotype correlation in most cases. However, some characteristics can be helpful to drive genetic testing. Deep phenotyping of patients, together with functional studies of single mutations, particularly those associated with atypical phenotypes, are necessary to better understand the biological mechanisms underlying this clinical variability.

Space for power: feeling powerful over others' behavior affects peri-personal space representation.

We investigated whether and how social power affects the representation of peri-personal space (PPS). We applied a multisensory interaction task to assess PPS representation and the Personal Sense of Power Scale to assess participants' feelings of power over others' behaviors and over others' opinions. In Study 1, we probed PPS representation in a virtual social context. Participants with a higher sense of power showed a less defined differentiation between the close and far space as compared to participants with a lower sense of power. This effect was replicated in Study 2 when participants performed the task in a non-social context (with no person in the scene), but only after they were reminded of an episode of power. Thus, social power-the perception of power over others' behavior-affects the multisensory representation of the self in space by blurring the differentiation between one's own PPS and the space of others.

Spontaneous intracerebral haemorrhage associated with early-onset cerebral amyloid angiopathy and Alzheimer's disease neuropathological changes five decades after cadaveric dura mater graft.

Cerebral amyloid angiopathy (CAA) is a small vessel disease, causing spontaneous intracerebral hemorrhage (ICH) in the elderly. It is strongly associated with Alzheimer disease (AD), as most CAA patients show deposition of Aß-i.e. the basic component of parenchymal Alzheimer amyloid deposits-in the cerebral vessels. Iatrogenic early-onset CAA has been recently identified in patients with a history of traumatic brain injury or other cerebral as well as extra-cerebral lesions that led to neurosurgery or other medical procedures as intravascular embolization by cadaveric dura mater extracts many years before the first ICH event. In those patients, a transmission of Aß seeds from neurosurgical instruments or from cadaveric dura mater exposure was suggested. We report a 51-year-old woman with unremarkable family history who presented abruptly with aphasia and right hemiparesis. A cerebral left lobar haemorrhagic stroke was documented by neuroimaging. Accurate anamnesis revealed a neurosurgical procedure with cadaveric dura mater graft at the age of 2 years for an arachnoid cyst. The neuropathological examination of the cerebral parietal biopsy showed severe amyloid angiopathy in many leptomeningeal and cortical vessels, as well as abundant parenchymal Aß deposits, neurofibrillary tangles and neuropil threads. The mechanism involved in the human-to-human transmission of the Aß proteinopathy remains to be clarified. In our patient the cadaver derived dura used for grafting is a very strong candidate as the source of the transmission. A systematic monitoring of individuals who have had neurosurgical procedures in early life, especially those involving cadaveric dural grafts, is required to determine the ratio of those affected by CAA many years later and unaffected. Moreover, our report confirms that in addition to vascular and parenchymal Aß pathology, neurofibrillary changes indistinguishable from AD may develop in specific conditions with long latency period from the neurosurgical or embolization procedure.

Rapid Recalibration of Peri-Personal Space: Psychophysical, Electrophysiological, and Neural Network Modeling Evidence.

Interactions between individuals and the environment occur within the peri-personal space (PPS). The encoding of this space plastically adapts to bodily constraints and stimuli features. However, these remapping effects have not been demonstrated on an adaptive time-scale, trial-to-trial. Here, we test this idea first via a visuo-tactile reaction time (RT) paradigm in augmented reality where participants are asked to respond as fast as possible to touch, as visual objects approach them. Results demonstrate that RTs to touch are facilitated as a function of visual proximity, and the sigmoidal function describing this facilitation shifts closer to the body if the immediately precedent trial had indexed a smaller visuo-tactile disparity. Next, we derive the electroencephalographic correlates of PPS and demonstrate that this multisensory measure is equally shaped by recent sensory history. Finally, we demonstrate that a validated neural network model of PPS is able to account for the present results via a simple Hebbian plasticity rule. The present findings suggest that PPS encoding remaps on a very rapid time-scale and, more generally, that it is sensitive to sensory history, a key feature for any process contextualizing subsequent incoming sensory information (e.g., a Bayesian prior).

Predictive coding of action intentions in dorsal and ventral visual stream is based on visual anticipations, memory-based information and motor preparation.

Predictions of upcoming movements are based on several types of neural signals that span the visual, somatosensory, motor and cognitive system. Thus far, pre-movement signals have been investigated while participants viewed the object to be acted upon. Here, we studied the contribution of information other than vision to the classification of preparatory signals for action, even in the absence of online visual information. We used functional magnetic resonance imaging (fMRI) and multivoxel pattern analysis (MVPA) to test whether the neural signals evoked by visual, memory-based and somato-motor information can be reliably used to predict upcoming actions in areas of the dorsal and ventral visual stream during the preparatory phase preceding the action, while participants were lying still. Nineteen human participants (nine women) performed one of two actions towards an object with their eyes open or closed. Despite the well-known role of ventral stream areas in visual recognition tasks and the specialization of dorsal stream areas in somato-motor processes, we decoded action intention in areas of both streams based on visual, memory-based and somato-motor signals. Interestingly, we could reliably decode action intention in absence of visual information based on neural activity evoked when visual information was available and vice versa. Our results show a similar visual, memory and somato-motor representation of action planning in dorsal and ventral visual stream areas that allows predicting action intention across domains, regardless of the availability of visual information.

Social perception of others shapes one's own multisensory peripersonal space.

The perception of our self is not restricted to our physical boundaries, but it extends beyond the body to incorporate the space where individual-environment interactions occur, i.e., the peripersonal space (PPS). PPS is generally conceived as a low-level multisensory-motor interface mediating hand-object interactions. Recent studies, however, showed that PPS representation is affected by higher-level cognitive factors. Here we asked whether the multisensory representation of PPS is influenced by high-level mechanisms implied in social interactions, such as the social perception of others. To this aim, in Experiment 1, we developed and validated a new multisensory interaction task in mixed reality (i.e., the Social PPS task). This task allows measuring the boundaries of PPS between one self and another person in a fully controlled, yet highly ecological, set-up. In the Experiment 2, we used this task to measure how participants' PPS varied when facing another person. The social perception of this person was manipulated via a classic social psychology procedure, so that, in two conditions, she was perceived either as a moral or an immoral character. We found that PPS representation is sensitive to the social perception of the other, being more extended when participants were facing a moral than when facing an immoral person. This effect was specific for social context, as no change in PPS was found if participants were facing an object, instead of the person. Interestingly, the social manipulation affected also attitude, identification, willingness to interact with the other, so as interpersonal distance. Together these findings show that social perception of others affects both the psychological representation of the others in relation to oneself and the multisensory representations of the space between oneself and the other, offering new insights about the role of social cognition in body representation.