Scholarlink

The blood-brain barrier (BBB) is an essential physiological separator that controls substance movement into the central nervous system (CNS) and is crucial for CNS stability. However, the neuroprotective role of the BBB, especially in disease states, presents significant challenges for brain drug delivery. This is evident in conditions with overactive efflux transporters, which prevent drugs from reaching target brain tissues or achieving effective concentrations. These complexities have led to a rise in utilizing a variety of deep learning (DL) models, including the employment of neural network models and support vector machines to predict drug permeability across the BBB and develop personalized care for CNS disorders. Furthermore, the high specificity and accuracy of these models, along with their increased efficiency, make them more favorable than traditional models as well as machine learning. However, the complexity of DL models, data imbalances, and the opaque nature of these algorithms present ongoing challenges. Despite these challenges, DL models have the potential to truly innovate the sphere of personalized medicine, especially in terms of neurological therapies.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning and Machine Learning Algorithms: A Blood-Brain Barrier Permeability Prediction Model with Accuracy

  • 实验方案

  • 发表日期：

    2025年

查看全文
  • Arjun Ahluwalia,
  • Wireko Andrew Awuah,
  • Muhammad Hamza Shah,
  • Vivek Sanker,
  • Kwadwo Darko,
  • Adam Ben-Jaafar,
  • Joecelyn Kirani Tan,
  • Sruthi Ranganathan,
  • Tenkorang Ohenewaa Pearl,
  • Nicholas Aderinto,
  • Toufik Abdul-Rahman,
  • Oday Atallah,
  • Athanasios Alexiou,
  • Ghulam Md Ashraf

摘要

The blood-brain barrier (BBB) is an essential physiological separator that controls substance movement into the central nervous system (CNS) and is crucial for CNS stability. However, the neuroprotective role of the BBB, especially in disease states, presents significant challenges for brain drug delivery. This is evident in conditions with overactive efflux transporters, which prevent drugs from reaching target brain tissues or achieving effective concentrations. These complexities have led to a rise in utilizing a variety of deep learning (DL) models, including the employment of neural network models and support vector machines to predict drug permeability across the BBB and develop personalized care for CNS disorders. Furthermore, the high specificity and accuracy of these models, along with their increased efficiency, make them more favorable than traditional models as well as machine learning. However, the complexity of DL models, data imbalances, and the opaque nature of these algorithms present ongoing challenges. Despite these challenges, DL models have the potential to truly innovate the sphere of personalized medicine, especially in terms of neurological therapies.