The Establishment of a Hypoxia Cellular Morphology Model Based on Deep Convolutional Neural Networks and Intelligent Screening of Anti-Hypoxia Drugs

EasyChair Preprint 13229

Abstract

The rapid development of artificial intelligence has brought many innovative achievements to fields such as drug design and drug discovery . Combining traditional graphics methods with deep learning methods can significantly improve the accuracy of drug screening.

Hypoxia refers to the process in which tissues or cells in the body undergo abnormal changes in morphology, physiological functions, and metabolism due to insufficient oxygen supply or oxygen utilization obstacles. Hypoxia is a major characteristic of high-altitude environments, so it is crucial to prevent and treat cardiovascular diseases related to hypoxia. 

Enlargement of the cell nucleus is a recognizable morphological feature of hypoxic cells. Based on a deep learning multi-cell image classification model, this study constructed a high throughput compound screening system for discriminating the anti-hypoxic activity of thousands of compounds. By simultaneously performing prediction scoring using AC16 and H9C2 models, the anti-hypoxic activity of thousands of compounds was predicted, and some compound molecules with anti-hypoxic

effects were successfully screened. Therefore, morphology-based CNN systems can become powerful tools for screening antioxidant drugs.

Keyphrases: Convolutional Neural Networks, Screening of anti-hypoxia drugs, deep learning

@booklet{EasyChair:13229,
  author    = {Xinyi Zhang and Zheng Wang},
  title     = {The Establishment of a Hypoxia Cellular Morphology Model Based on Deep Convolutional Neural Networks and Intelligent Screening of Anti-Hypoxia Drugs},
  howpublished = {EasyChair Preprint 13229},
  year      = {EasyChair, 2024}}