Objective: To explore the reproducibility and stability of brain functional area in the response of language tasks during Chinese word processing with functional magnetic resonance imaging (fMRI) by follow-up scanning so as to provide rationales for clinical preoperative localization and the mechanisms of recovery from aphasia.
Methods: The fMRI data were collected by scanning semantic and phonologic judgments in 15 healthy volunteers. Each subject was scanned twice by the same fMRI procedure of both language tasks with an interval of 1 month. By analyzing the effective overlapping average activation maps, the reproducibility of inter-subject imaging result for two language tasks were estimated by selecting the main language areas, such as Broca's area and Wernicke's area as the region of interest (ROI). By individually calculating the spatial distance of ROI centroid coordinates in the same activating range before and after test, the inter-subject stability in between-session was calculated quantitatively.
Results: Both language tasks activated more than one language-related brain areas in cerebral hemispheres. Both language tasks induced significant BOLD response in Broca's and Wernicke's areas with a tendency of left lateralization. The number of subjects in terms of the activation of both language tasks in Broca's and Wernicke's areas accounted over a half of the total subjects. As compared with the phonologic judgment task, the semantic judgment task showed better reproducibility in Broca's area. In the same spatial distance of ROI centroid coordinates, the stability of Broca's area was higher than that of Wernicke's area while the stability of semantic judgment in Broca's area higher than that of phonologic judgment.
Conclusion: Such main language domains as Broca's and Wernicke's areas can be effectively activated by both semantic and phonologic judgments. By comparison, semantic judgment in Broca's area shows a higher level of reproducibility and stability. Thus it is applicable for clinical preoperative localization and the mechanisms of recovery from aphasia.
Download full-text PDF |
Source |
---|
Sci Rep
December 2024
Department of Computer Science, Birzeit University, P.O. Box 14, Birzeit, West Bank, Palestine.
Accurate classification of logos is a challenging task in image recognition due to variations in logo size, orientation, and background complexity. Deep learning models, such as VGG16, have demonstrated promising results in handling such tasks. However, their performance is highly dependent on optimal hyperparameter settings, whose fine-tuning is both labor-intensive and time-consuming.
View Article and Find Full Text PDFFront Comput Neurosci
December 2024
School of Health Sciences and Engineering, University of Shanghai for Science and Technology, Shanghai, China.
This study aims to enhance the classification accuracy of adverse events associated with the da Vinci surgical robot through advanced natural language processing techniques, thereby ensuring medical device safety and protecting patient health. Addressing the issues of incomplete and inconsistent adverse event records, we employed a deep learning model that combines BERT and BiLSTM to predict whether adverse event reports resulted in patient harm. We developed the Bert-BiLSTM-Att_dropout model specifically for text classification tasks with small datasets, optimizing the model's generalization ability and key information capture through the integration of dropout and attention mechanisms.
View Article and Find Full Text PDFAdv Sci (Weinh)
December 2024
Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, 07024, USA.
The performance of seven large language models (LLMs) in generating programming code using various prompt strategies, programming languages, and task difficulties is systematically evaluated. GPT-4 substantially outperforms other LLMs, including Gemini Ultra and Claude 2. The coding performance of GPT-4 varies considerably with different prompt strategies.
View Article and Find Full Text PDFCureus
November 2024
Department of Neurosurgery, Fukushima Medical University, Fukushima, JPN.
Introduction The degree to which each human brain hemisphere governs specific cognitive processes, such as language and handedness (the preference or dominance of one hand over the other), varies across individuals. Research has explored the nature of language laterality in left-handed (LH) individuals, indicating that left-hemisphere dominance for language is commonly observed across both left- and right-handed populations. Advanced imaging techniques, including functional transcranial Doppler sonography and fMRI, have revealed subtle differences in language lateralization between LH and right-handed (RH) individuals, particularly in semantic processing tasks.
View Article and Find Full Text PDFSci Rep
December 2024
School of Computer Science and Technology (School of Cyberspace Security), Xinjiang University, Urumqi, 830046, China.
Recent works have introduced prompt learning for Event Argument Extraction (EAE) since prompt-based approaches transform downstream tasks into a more consistent format with the training task of Pre-trained Language Model (PLM). This helps bridge the gap between downstream tasks and model training. However, these previous works overlooked the complex number of events and their relationships within sentences.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!