Despite recent advances in natural language processing, many statistical models for processing text perform extremely poorly under domain shift. Processing biomedical and clinical text is a critically important application area of natural language processing, for which there are few robust, practical, publicly available models. This paper describes scis-paCy, a new Python library and models for practical biomedical/scientific text processing, which heavily leverages the spaCy library. We detail the performance of two packages of models released in scispaCy and demonstrate their robustness on several tasks and datasets. Models and code are available at https:// allenai.github.io/scispacy/.

Specific abnormalities of vision in schizophrenia have been observed to affect high-level and some low-level integration mechanisms, suggesting that people with schizophrenia may experience anomalies across different stages in the visual system affecting either early or late processing or both. Here, we review the research into visual illusion perception in schizophrenia and the issues which previous research has faced. One general finding that emerged from the literature is that those with schizophrenia are mostly immune to the effects of high-level illusory displays, but this effect is not consistent across all low-level illusions. The present review suggests that this resistance is due to the weakening of top–down perceptual mechanisms and may be relevant to the understanding of symptoms of visual distortion rather than hallucinations as previously thought.

As a step toward better document-level understanding, we explore classification of a sequence of sentences into their corresponding categories, a task that requires understanding sentences in context of the document. Recent successful models for this task have used hierarchical models to contextualize sentence representations, and Conditional Random Fields (CRFs) to incorporate dependencies between subsequent labels. In this work, we show that pretrained language models, BERT (Devlin et al., 2018) in particular, can be used for this task to capture contextual dependencies without the need for hierarchical encoding nor a CRF. Specifically, we construct a joint sentence representation that allows BERT Transformer layers to directly utilize contextual information from all words in all sentences. Our approach achieves state-of-the-art results on four datasets, including a new dataset of structured scientific abstracts.

Paediatric traumatic brain injury (pTBI) is a leading cause of disability for children and young adults. Children are a uniquely vulnerable group with the disease process that occurs following a pTBI interacting with the trajectory of normal brain development. Quantitative MRI post-injury has suggested a long-term, neurodegenerative effect of TBI on the morphometry of the brain, in both adult and childhood TBI. Changes to the brain beyond that of anticipated, age-dependant differences may allow us to estimate the state of the brain post-injury and produce clinically relevant predictions for long-term outcome. The current review synthesises the existing literature to assess whether, following pTBI, the morphology of the brain exhibits either i) longitudinal change and/or ii) differences compared to healthy controls and outcomes. The current literature suggests that morphometric differences from controls are apparent cross-sectionally at both acute and late-chronic timepoints post-injury, thus suggesting a non-transient effect of injury. Developmental trajectories of morphometry are altered in TBI groups compared to patients, and it is unlikely that typical maturation overcomes damage post-injury, or even ‘catches up’ with that of typically-developing peers. However, there is limited evidence for diverted developmental trajectories being associated with cognitive impairment post-injury. The current review also highlights the apparent challenges to the existing literature and potential methods by which these can be addressed.

Structural segmentation of T1-weighted (T1w) MRI has shown morphometric differences, both compared to controls and longitudinally, following a traumatic brain injury (TBI). While many patients with TBI present with abnormalities on structural MRI images, most neuroimaging software packages have not been systematically evaluated for accuracy in the presence of these pathology-related MRI abnormalities. The current study aimed to assess whether acute MRI lesions (MRI acquired 7–71 days post-injury) cause error in the estimates of brain volume produced by the semi-automated segmentation tool, Freesurfer. More specifically, to investigate whether this error was global, the presence of lesion-induced error in the contralesional hemisphere, where no abnormal signal was present, was measured. A dataset of 176 simulated lesion cases was generated using actual lesions from 16 pediatric TBI (pTBI) cases recruited from the emergency department and 11 typically-developing controls. Simulated lesion cases were compared to the “ground truth” of the non-lesion control-case T1w images. Using linear mixed-effects models, results showed that hemispheric measures of cortex volume were significantly lower in the contralesional-hemisphere compared to the ground truth. Interestingly, however, cortex volume (and cerebral white matter volume) were not significantly different in the lesioned hemisphere. However, percent volume difference (PVD) between the simulated lesion and ground truth showed that the magnitude of difference of cortex volume in the contralesional-hemisphere (mean PVD = 0.37%) was significantly smaller than that in the lesioned hemisphere (mean PVD = 0.47%), suggesting a small, but systematic lesion-induced error. Lesion characteristics that could explain variance in the PVD for each hemisphere were investigated. Taken together, these results suggest that the lesion-induced error caused by simulated lesions was not focal, but globally distributed. Previous post-processing approaches to adjust for lesions in structural analyses address the focal region where the lesion was located however, our results suggest that focal correction approaches are insufficient for the global error in morphometric measures of the injured brain.