3 results match your criteria: "The Artificial Intelligence Research Center of Novosibirsk State University[Affiliation]"
Transfer Learning Approaches for Brain Metastases Screenings.
Biomedicines
November 2024
The Artificial Intelligence Research Center of Novosibirsk State University, Novosibirsk State University, 630090 Novosibirsk, Russia.
Background: In this study, we examined the effectiveness of transfer learning in improving automatic segmentation of brain metastases on magnetic resonance imaging scans, with potential applications in preventive exams and remote diagnostics.
Methods: We trained three deep learning models on a public dataset from the ASNR-MICCAI Brain Metastasis Challenge 2024, fine-tuned them on a small private dataset, and compared their performance to models trained from scratch.
Results: Results showed that models using transfer learning performed better than scratch-trained models, though the improvement was not statistically substantial.
AI-Assisted Identification of Primary and Secondary Metabolomic Markers for Postoperative Delirium.
Int J Mol Sci
November 2024
V. Zelman Institute for the Medicine and Psychology, Novosibirsk State University, Novosibirsk 630090, Russia.
Despite considerable investigative efforts, the molecular mechanisms of postoperative delirium (POD) remain unresolved. The present investigation employs innovative methodologies for identifying potential primary and secondary metabolic markers of POD by analyzing serum metabolomic profiles utilizing the genetic algorithm and artificial neural networks. The primary metabolomic markers constitute a combination of metabolites that optimally distinguish between POD and non-POD groups of patients.
View Article and Find Full Text PDFAn Accurate and Efficient Approach to Knowledge Extraction from Scientific Publications Using Structured Ontology Models, Graph Neural Networks, and Large Language Models.
Int J Mol Sci
November 2024
The Artificial Intelligence Research Center of Novosibirsk State University, Pirogova Street 1, Novosibirsk 630090, Russia.
Article Synopsis
- * Traditional text-mining techniques often lack accuracy due to their inability to grasp semantic and contextual details, while deep-learning models are expensive and can produce misleading information (known as hallucination).
- * This study introduces a hybrid method using text-mining, graph neural networks (GNNs), and fine-tuned large language models (LLMs) to enhance biomedical knowledge graphs, achieving high accuracy in predicting protein interactions and identifying new connections relevant to conditions like insomnia.