Production of relative clauses in Cantonese-speaking children with and without Developmental Language Disorder.

Developmental Language Disorder (DLD) has been explained as either a deficit deriving from an abstract representational deficit or as emerging from difficulties in acquiring and coordinating multiple interacting cues guiding learning. These competing explanations are often difficult to decide between when tested on European languages. This paper reports an experimental study of relative clause (RC) production in Cantonese-speaking children with and without DLD, which enabled us to test multiple developmental predictions derived from one prominent theory - emergentism. Children with DLD (N = 22; aged 6;6-9;7) were compared with age-matched typically-developing peers (N = 23) and language-matched, typically-developing children (N = 21; aged 4;7-7;6) on a sentence repetition task. Results showed that children's production across multiple RC types was influenced by structural frequency, general semantic complexity, and the linear order of constituents, with the DLD group performing worse than their age-matched and language-matched peers. The results are consistent with the emergentist explanation of DLD.

Occipital-temporal cortical tuning to semantic and affective features of natural images predicts associated behavioral responses.

In everyday life, people need to respond appropriately to many types of emotional stimuli. Here, we investigate whether human occipital-temporal cortex (OTC) shows co-representation of the semantic category and affective content of visual stimuli. We also explore whether OTC transformation of semantic and affective features extracts information of value for guiding behavior. Participants viewed 1620 emotional natural images while functional magnetic resonance imaging data were acquired. Using voxel-wise modeling we show widespread tuning to semantic and affective image features across OTC. The top three principal components underlying OTC voxel-wise responses to image features encoded stimulus animacy, stimulus arousal and interactions of animacy with stimulus valence and arousal. At low to moderate dimensionality, OTC tuning patterns predicted behavioral responses linked to each image better than regressors directly based on image features. This is consistent with OTC representing stimulus semantic category and affective content in a manner suited to guiding behavior.

Optimizing word embeddings for small dataset: a case study on patient portal messages from breast cancer patients.

Patient portal messages often relate to specific clinical phenomena (e.g., patients undergoing treatment for breast cancer) and, as a result, have received increasing attention in biomedical research. These messages require natural language processing and, while word embedding models, such as word2vec, have the potential to extract meaningful signals from text, they are not readily applicable to patient portal messages. This is because embedding models typically require millions of training samples to sufficiently represent semantics, while the volume of patient portal messages associated with a particular clinical phenomenon is often relatively small. We introduce a novel adaptation of the word2vec model, PK-word2vec (where PK stands for prior knowledge), for small-scale messages. PK-word2vec incorporates the most similar terms for medical words (including problems, treatments, and tests) and non-medical words from two pre-trained embedding models as prior knowledge to improve the training process. We applied PK-word2vec in a case study of patient portal messages in the Vanderbilt University Medical Center electric health record system sent by patients diagnosed with breast cancer from December 2004 to November 2017. We evaluated the model through a set of 1000 tasks, each of which compared the relevance of a given word to a group of the five most similar words generated by PK-word2vec and a group of the five most similar words generated by the standard word2vec model. We recruited 200 Amazon Mechanical Turk (AMT) workers and 7 medical students to perform the tasks. The dataset was composed of 1389 patient records and included 137,554 messages with 10,683 unique words. Prior knowledge was available for 7981 non-medical and 1116 medical words. In over 90% of the tasks, both reviewers indicated PK-word2vec generated more similar words than standard word2vec (p = 0.01).The difference in the evaluation by AMT workers versus medical students was negligible for all comparisons of tasks' choices between the two groups of reviewers ( p = 0.774 under a paired t-test). PK-word2vec can effectively learn word representations from a small message corpus, marking a significant advancement in processing patient portal messages.

Keystrokes: A practical exploration of semantic drift in timed word association tasks.

This study investigates the phenomena of semantic drift through the lenses of language and situated simulation (LASS) and the word frequency effect (WFE) within a timed word association task. Our primary objectives were to determine whether semantic drift can be identified over the short time (25 seconds) of a free word association task (a predicted corollary of LASS), and whether more frequent terms are generated earlier in the process (as expected due to the WFE). Respondents were provided with five cue words (tree, dog, quality, plastic and love), and asked to write as many associations as they could. We hypothesized that terms generated later in the task (fourth time quartile, the last 19-25 seconds) would be semantically more distant (cosine similarity) from the cue word than those generated earlier (first quartile, the first 1-7 seconds), indicating semantic drift. Additionally, we explored the WFE by hypothesizing that earlier generated words would be more frequent and less diverse. Utilizing a dataset matched with GloVe 300B word embeddings, BERT and WordNet synsets, we analysed semantic distances among 1569 unique term pairs for all cue words across time. Our results supported the presence of semantic drift, with significant evidence of within-participant, semantic drift from the first to fourth time (LASS) and frequency (WFE) quartiles. In terms of the WFE, we observed a notable decrease in the diversity of terms generated earlier in the task, while more unique terms (greater diversity and relative uniqueness) were generated in the 4th time quartile, aligning with our hypothesis that more frequently used words dominate early stages of a word association task. We also found that the size of effects varied substantially across cues, suggesting that some cues might invoke stronger and more idiosyncratic situated simulations. Theoretically, our study contributes to the understanding of LASS and the WFE. It suggests that semantic drift might serve as a scalable indicator of the invocation of language versus simulation systems in LASS and might also be used to explore cognition within word association tasks more generally. The findings also add a temporal and relational dimension to the WFE. Practically, our research highlights the utility of word association tasks in understanding semantic drift and the diffusion of word usage over a sub-minute task, arguably the shortest practically feasible timeframe, offering a scalable method to explore group and individual changes in semantic relationships, whether via the targeted diffusion of influence in a marketing campaign, or seeking to understand differences in cognition more generally. Possible practical uses and opportunities for future research are discussed.

Scene semantics affects allocentric spatial coding for action in naturalistic (virtual) environments.

Interacting with objects in our environment requires determining their locations, often with respect to surrounding objects (i.e., allocentrically). According to the scene grammar framework, these usually small, local objects are movable within a scene and represent the lowest level of a scene's hierarchy. How do higher hierarchical levels of scene grammar influence allocentric coding for memory-guided actions? Here, we focused on the effect of large, immovable objects (anchors) on the encoding of local object positions. In a virtual reality study, participants (n = 30) viewed one of four possible scenes (two kitchens or two bathrooms), with two anchors connected by a shelf, onto which were presented three local objects (congruent with one anchor) (Encoding). The scene was re-presented (Test) with 1) local objects missing and 2) one of the anchors shifted (Shift) or not (No shift). Participants, then, saw a floating local object (target), which they grabbed and placed back on the shelf in its remembered position (Response). Eye-tracking data revealed that both local objects and anchors were fixated, with preference for local objects. Additionally, anchors guided allocentric coding of local objects, despite being task-irrelevant. Overall, anchors implicitly influence spatial coding of local object locations for memory-guided actions within naturalistic (virtual) environments.

A semantic feature enhanced YOLOv5-based network for polyp detection from colonoscopy images.

Colorectal cancer (CRC) is a common digestive system tumor with high morbidity and mortality worldwide. At present, the use of computer-assisted colonoscopy technology to detect polyps is relatively mature, but it still faces some challenges, such as missed or false detection of polyps. Therefore, how to improve the detection rate of polyps more accurately is the key to colonoscopy. To solve this problem, this paper proposes an improved YOLOv5-based cancer polyp detection method for colorectal cancer. The method is designed with a new structure called P-C3 incorporated into the backbone and neck network of the model to enhance the expression of features. In addition, a contextual feature augmentation module was introduced to the bottom of the backbone network to increase the receptive field for multi-scale feature information and to focus on polyp features by coordinate attention mechanism. The experimental results show that compared with some traditional target detection algorithms, the model proposed in this paper has significant advantages for the detection accuracy of polyp, especially in the recall rate, which largely solves the problem of missed detection of polyps. This study will contribute to improve the polyp/adenoma detection rate of endoscopists in the process of colonoscopy, and also has important significance for the development of clinical work.

The role of alpha oscillations in free- and goal-directed semantic associations.

Alpha oscillations are known to play a central role in several higher-order cognitive functions, especially selective attention, working memory, semantic memory, and creative thinking. Nonetheless, we still know very little about the role of alpha in the generation of more remote semantic associations, which is key to creative and semantic cognition. Furthermore, it remains unclear how these oscillations are shaped by the intention to "be creative," which is the case in most creativity tasks. We aimed to address these gaps in two experiments. In Experiment 1, we compared alpha oscillatory activity (using a method which distinguishes genuine oscillatory activity from transient events) during the generation of free associations which were more vs. less distant from a given concept. In Experiment 2, we replicated these findings and also compared alpha oscillatory activity when people were generating free associations versus associations with the instruction to be creative (i.e. goal-directed). We found that alpha was consistently higher during the generation of more distant semantic associations, in both experiments. This effect was widespread, involving areas in both left and right hemispheres. Importantly, the instruction to be creative seems to increase alpha phase synchronisation from left to right temporal brain areas, suggesting that intention to be creative changed the flux of information in the brain, likely reflecting an increase in top-down control of semantic search processes. We conclude that goal-directed generation of remote associations relies on top-down mechanisms compared to when associations are freely generated.

Co-learning companionship benefits word learning in a new language: Evidence from a dual-brain EEG examination.

Companionship refers to one's being in the presence of another individual. For adults, acquiring a new language is a highly social activity that often involves learning in the context of companionship. However, the effects of companionship on new language learning have gone relatively underexplored, particularly with respect to word learning. Using a within-subject design, the current study employs electroencephalography to examine how two types of companionship (monitored and co-learning) affect word learning (semantic and lexical) in a new language. Dyads of Chinese speakers of English as a second language participated in a pseudo-word-learning task during which they were placed in monitored and co-learning companionship contexts. The results showed that exposure to co-learning companionship affected the early attention stage of word learning. Moreover, in this early stage, evidence of a higher representation similarity between co-learners showed additional support that co-learning companionship influenced attention. Observed increases in delta and theta interbrain synchronization further revealed that co-learning companionship facilitated semantic access. In all, the similar neural representations and interbrain synchronization between co-learners suggest that co-learning companionship offers important benefits for learning words in a new language.

Creative connections: the neural correlates of semantic relatedness are associated with creativity.

The associative theory of creativity proposes that creative ideas result from connecting remotely related concepts in memory. Previous research found that higher creative individuals exhibit a more flexible organization of semantic memory, generate more uncommon word associations, and judge remote concepts as more related. In this study (N = 93), we used fMRI to investigate brain regions involved in judging the relatedness of concepts that vary in their semantic distance, and how such neural involvement relates to individual differences in creativity. Brain regions where activity increased with semantic relatedness mainly overlapped with default, control, salience, semantic control, and multiple demand networks. The default and semantic control networks exhibited increased involvement when evaluating more remote associations. Finally, higher creative people, who provided higher relatedness judgements on average, exhibited lower activity in those regions, possibly reflecting higher neural efficiency. We discuss these findings in the context of the neurocognitive processing underlying creativity. Overall, our findings indicate that judging remote concepts as related reflects a cognitive mechanism underlying creativity and shed light on the neural correlates of this mechanism.

Early parafoveal semantic integration in natural reading.

Humans can read and comprehend text rapidly, implying that readers might process multiple words per fixation. However, the extent to which parafoveal words are previewed and integrated into the evolving sentence context remains disputed. We investigated parafoveal processing during natural reading by recording brain activity and eye movements using MEG and an eye tracker while participants silently read one-line sentences. The sentences contained an unpredictable target word that was either congruent or incongruent with the sentence context. To measure parafoveal processing, we flickered the target words at 60 Hz and measured the resulting brain responses (i.e. Rapid Invisible Frequency Tagging, RIFT) during fixations on the pre-target words. Our results revealed a significantly weaker tagging response for target words that were incongruent with the previous context compared to congruent ones, even within 100ms of fixating the word immediately preceding the target. This reduction in the RIFT response was also found to be predictive of individual reading speed. We conclude that semantic information is not only extracted from the parafovea but can also be integrated with the previous context before the word is fixated. This early and extensive parafoveal processing supports the rapid word processing required for natural reading. Our study suggests that theoretical frameworks of natural reading should incorporate the concept of deep parafoveal processing.

Semantic encoding during language comprehension at single-cell resolution.

From sequences of speech sounds1,2 or letters3, humans can extract rich and nuanced meaning through language. This capacity is essential for human communication. Yet, despite a growing understanding of the brain areas that support linguistic and semantic processing4-12, the derivation of linguistic meaning in neural tissue at the cellular level and over the timescale of action potentials remains largely unknown. Here we recorded from single cells in the left language-dominant prefrontal cortex as participants listened to semantically diverse sentences and naturalistic stories. By tracking their activities during natural speech processing, we discover a fine-scale cortical representation of semantic information by individual neurons. These neurons responded selectively to specific word meanings and reliably distinguished words from nonwords. Moreover, rather than responding to the words as fixed memory representations, their activities were highly dynamic, reflecting the words' meanings based on their specific sentence contexts and independent of their phonetic form. Collectively, we show how these cell ensembles accurately predicted the broad semantic categories of the words as they were heard in real time during speech and how they tracked the sentences in which they appeared. We also show how they encoded the hierarchical structure of these meaning representations and how these representations mapped onto the cell population. Together, these findings reveal a finely detailed cortical organization of semantic representations at the neuron scale in humans and begin to illuminate the cellular-level processing of meaning during language comprehension.

Optimized continuous homecare provisioning through distributed data-driven semantic services and cross-organizational workflows.

BACKGROUND: In healthcare, an increasing collaboration can be noticed between different caregivers, especially considering the shift to homecare. To provide optimal patient care, efficient coordination of data and workflows between these different stakeholders is required. To achieve this, data should be exposed in a machine-interpretable, reusable manner. In addition, there is a need for smart, dynamic, personalized and performant services provided on top of this data. Flexible workflows should be defined that realize their desired functionality, adhere to use case specific quality constraints and improve coordination across stakeholders. User interfaces should allow configuring all of this in an easy, user-friendly way. METHODS: A distributed, generic, cascading reasoning reference architecture can solve the presented challenges. It can be instantiated with existing tools built upon Semantic Web technologies that provide data-driven semantic services and constructing cross-organizational workflows. These tools include RMLStreamer to generate Linked Data, DIVIDE to adaptively manage contextually relevant local queries, Streaming MASSIF to deploy reusable services, AMADEUS to compose semantic workflows, and RMLEditor and Matey to configure rules to generate Linked Data. RESULTS: A use case demonstrator is built on a scenario that focuses on personalized smart monitoring and cross-organizational treatment planning. The performance and usability of the demonstrator's implementation is evaluated. The former shows that the monitoring pipeline efficiently processes a stream of 14 observations per second: RMLStreamer maps JSON observations to RDF in 13.5 ms, a C-SPARQL query to generate fever alarms is executed on a window of 5 s in 26.4 ms, and Streaming MASSIF generates a smart notification for fever alarms based on severity and urgency in 1539.5 ms. DIVIDE derives the C-SPARQL queries in 7249.5 ms, while AMADEUS constructs a colon cancer treatment plan and performs conflict detection with it in 190.8 ms and 1335.7 ms, respectively. CONCLUSIONS: Existing tools built upon Semantic Web technologies can be leveraged to optimize continuous care provisioning. The evaluation of the building blocks on a realistic homecare monitoring use case demonstrates their applicability, usability and good performance. Further extending the available user interfaces for some tools is required to increase their adoption.

Mind the suffix: Pseudoword processing in children and adults.

Previous evidence has shown that pseudowords made up of real morphemes take more time to process and generate more errors than pseudowords without morphemes in a lexical decision task. The explanation for these results is controversial because two possible arguments may be posited; the first is related to the morphological composition of the stimuli, and the second is related to the larger semantic interpretability of pseudowords with morphemes in comparison with pseudowords without morphemes (a semantic-based explanation). To disentangle this issue, we conducted an experiment with 92 children and 42 adults. For this purpose, a lexical decision task was implemented, controlling for semantic interpretability while manipulating the morphological status of pseudowords. The results show that the morphological composition of pseudowords generates larger latencies and more errors than pseudowords without morphemes, thereby corroborating that morphemes are activated during pseudoword processing even in the case of young readers.

Verbal semantic expertise is associated with reduced functional connectivity between left and right anterior temporal lobes.

The left and right anterior temporal lobes (ATLs) encode semantic representations. They show graded hemispheric specialization in function, with the left ATL contributing preferentially to verbal semantic processing. We investigated the cognitive correlates of this organization, using resting-state functional connectivity as a measure of functional segregation between ATLs. We analyzed two independent resting-state fMRI datasets (n = 86 and n = 642) in which participants' verbal semantic expertise was measured using vocabulary tests. In both datasets, people with more advanced verbal semantic knowledge showed weaker functional connectivity between left and right ventral ATLs. This effect was highly specific. It was not observed for within-hemisphere connections between semantic regions (ventral ATL and inferior frontal gyrus (IFG), though it was found for left-right IFG connectivity in one dataset). Effects were not found for tasks probing semantic control, nonsemantic cognition, or face recognition. Our results suggest that hemispheric specialization in the ATLs is not an innate property but rather emerges as people develop highly detailed verbal semantic representations. We speculate that this effect is a consequence of the left ATL's greater connectivity with left-lateralized written word recognition regions, which causes it to preferentially represent meaning for advanced vocabulary acquired primarily through reading.

Simple autonomous agents can enhance creative semantic discovery by human groups.

Innovation is challenging, and theory and experiments indicate that groups may be better able to identify and preserve innovations than individuals. But innovation within groups faces its own challenges, including groupthink and truncated diffusion. We performed experiments involving a game in which people search for ideas in various conditions: alone, in networked social groups, or in networked groups featuring autonomous agents (bots). The objective was to search a semantic space of 20,000 nouns with defined similarities for an arbitrary noun with the highest point value. Participants (N = 1875) were embedded in networks (n = 125) of 15 nodes to which we sometimes added 2 bots. The bots had 3 possible strategies: they shared a random noun generated by their immediate neighbors, or a noun most similar from among those identified, or a noun least similar. We first confirm that groups are better able to explore a semantic space than isolated individuals. Then we show that when bots that share the most similar noun operate in groups facing a semantic space that is relatively easy to navigate, group performance is superior. Simple autonomous agents with interpretable behavior can affect the capacity for creative discovery of human groups.

Enhancing semantic segmentation in chest X-ray images through image preprocessing: ps-KDE for pixel-wise substitution by kernel density estimation.

BACKGROUND: In medical imaging, the integration of deep-learning-based semantic segmentation algorithms with preprocessing techniques can reduce the need for human annotation and advance disease classification. Among established preprocessing techniques, Contrast Limited Adaptive Histogram Equalization (CLAHE) has demonstrated efficacy in improving segmentation algorithms across various modalities, such as X-rays and CT. However, there remains a demand for improved contrast enhancement methods considering the heterogeneity of datasets and the various contrasts across different anatomic structures. METHOD: This study proposes a novel preprocessing technique, ps-KDE, to investigate its impact on deep learning algorithms to segment major organs in posterior-anterior chest X-rays. Ps-KDE augments image contrast by substituting pixel values based on their normalized frequency across all images. We evaluate our approach on a U-Net architecture with ResNet34 backbone pre-trained on ImageNet. Five separate models are trained to segment the heart, left lung, right lung, left clavicle, and right clavicle. RESULTS: The model trained to segment the left lung using ps-KDE achieved a Dice score of 0.780 (SD = 0.13), while that of trained on CLAHE achieved a Dice score of 0.717 (SD = 0.19), p<0.01. ps-KDE also appears to be more robust as CLAHE-based models misclassified right lungs in select test images for the left lung model. The algorithm for performing ps-KDE is available at https://github.com/wyc79/ps-KDE. DISCUSSION: Our results suggest that ps-KDE offers advantages over current preprocessing techniques when segmenting certain lung regions. This could be beneficial in subsequent analyses such as disease classification and risk stratification.

Creation of Standardized Common Data Elements for Diagnostic Tests in Infectious Disease Studies: Semantic and Syntactic Mapping.

BACKGROUND: It is necessary to harmonize and standardize data variables used in case report forms (CRFs) of clinical studies to facilitate the merging and sharing of the collected patient data across several clinical studies. This is particularly true for clinical studies that focus on infectious diseases. Public health may be highly dependent on the findings of such studies. Hence, there is an elevated urgency to generate meaningful, reliable insights, ideally based on a high sample number and quality data. The implementation of core data elements and the incorporation of interoperability standards can facilitate the creation of harmonized clinical data sets. OBJECTIVE: This study's objective was to compare, harmonize, and standardize variables focused on diagnostic tests used as part of CRFs in 6 international clinical studies of infectious diseases in order to, ultimately, then make available the panstudy common data elements (CDEs) for ongoing and future studies to foster interoperability and comparability of collected data across trials. METHODS: We reviewed and compared the metadata that comprised the CRFs used for data collection in and across all 6 infectious disease studies under consideration in order to identify CDEs. We examined the availability of international semantic standard codes within the Systemized Nomenclature of Medicine - Clinical Terms, the National Cancer Institute Thesaurus, and the Logical Observation Identifiers Names and Codes system for the unambiguous representation of diagnostic testing information that makes up the CDEs. We then proposed 2 data models that incorporate semantic and syntactic standards for the identified CDEs. RESULTS: Of 216 variables that were considered in the scope of the analysis, we identified 11 CDEs to describe diagnostic tests (in particular, serology and sequencing) for infectious diseases: viral lineage/clade; test date, type, performer, and manufacturer; target gene; quantitative and qualitative results; and specimen identifier, type, and collection date. CONCLUSIONS: The identification of CDEs for infectious diseases is the first step in facilitating the exchange and possible merging of a subset of data across clinical studies (and with that, large research projects) for possible shared analysis to increase the power of findings. The path to harmonization and standardization of clinical study data in the interest of interoperability can be paved in 2 ways. First, a map to standard terminologies ensures that each data element's (variable's) definition is unambiguous and that it has a single, unique interpretation across studies. Second, the exchange of these data is assisted by "wrapping" them in a standard exchange format, such as Fast Health care Interoperability Resources or the Clinical Data Interchange Standards Consortium's Clinical Data Acquisition Standards Harmonization Model.

The role of embodied cognition in action language comprehension in L1 and L2.

In this study we carried out a behavioral experiment comparing action language comprehension in L1 (Italian) and L2 (English). Participants were Italian native speakers who had acquired the second language late (after the age of 10). They performed semantic judgments on L1 and L2 literal, idiomatic and metaphorical action sentences after viewing a video of a hand performing an action that was related or unrelated to the verb used in the sentence. Results showed that responses to literal and metaphorical L1 sentences were faster when the action depicted was related to the verb used rather than when the action depicted was unrelated to the verb used. No differences were found for the idiomatic condition. In L2 we found that all responses to the three conditions were facilitated when the action depicted was related to the verb used. Moreover, we found that the difference between the unrelated and the related modalities was greater in L2 than in L1 for the literal and the idiomatic condition but not for the metaphorical condition. These findings are consistent with the embodied cognition hypothesis of language comprehension.

Semantic associative abilities and executive control functions predict novelty and appropriateness of idea generation.

Novelty and appropriateness are two fundamental components of creativity. However, the way in which novelty and appropriateness are separated at behavioral and neural levels remains poorly understood. In the present study, we aim to distinguish behavioral and neural bases of novelty and appropriateness of creative idea generation. In alignment with two established theories of creative thinking, which respectively, emphasize semantic association and executive control, behavioral results indicate that novelty relies more on associative abilities, while appropriateness relies more on executive functions. Next, employing a connectome predictive modeling (CPM) approach in resting-state fMRI data, we define two functional network-based models-dominated by interactions within the default network and by interactions within the limbic network-that respectively, predict novelty and appropriateness (i.e., cross-brain prediction). Furthermore, the generalizability and specificity of the two functional connectivity patterns are verified in additional resting-state fMRI and task fMRI. Finally, the two functional connectivity patterns, respectively mediate the relationship between semantic association/executive control and novelty/appropriateness. These findings provide global and predictive distinctions between novelty and appropriateness in creative idea generation.

Multi-task transfer learning for the prediction of entity modifiers in clinical text: application to opioid use disorder case detection.

BACKGROUND: The semantics of entities extracted from a clinical text can be dramatically altered by modifiers, including entity negation, uncertainty, conditionality, severity, and subject. Existing models for determining modifiers of clinical entities involve regular expression or features weights that are trained independently for each modifier. METHODS: We develop and evaluate a multi-task transformer architecture design where modifiers are learned and predicted jointly using the publicly available SemEval 2015 Task 14 corpus and a new Opioid Use Disorder (OUD) data set that contains modifiers shared with SemEval as well as novel modifiers specific for OUD. We evaluate the effectiveness of our multi-task learning approach versus previously published systems and assess the feasibility of transfer learning for clinical entity modifiers when only a portion of clinical modifiers are shared. RESULTS: Our approach achieved state-of-the-art results on the ShARe corpus from SemEval 2015 Task 14, showing an increase of 1.1% on weighted accuracy, 1.7% on unweighted accuracy, and 10% on micro F1 scores. CONCLUSIONS: We show that learned weights from our shared model can be effectively transferred to a new partially matched data set, validating the use of transfer learning for clinical text modifiers.