Sign in Subscribe
Machine learning

Breakthrough in Artificial Intelligence: Generative AI Tool for Biomedical Nanoengineering

Strategic Insights

— 1 min read

A team of researchers has developed a generative artificial intelligence (AI) tool, NanoSafari, to automatically extract knowledge from the biomedical nanoscience literature and address scientific queries. The tool, which uses the Grouped Iterative Validation based Information Extraction (GIVE) method, was able to extract contextual information on nanoparticle characteristics from 20,000 published articles and provide accurate nanomaterial design parameters. In a blinded evaluation, biomedical nanoscientists found that NanoSafari outperformed the baseline model in providing more reliable parameters for nanomaterial design tasks, as further validated by bench experiments.

Key Takeaways:

  • A generative AI tool, NanoSafari, has been developed to automate knowledge extraction from biomedical nanoscience literature and provide accurate nanomaterial design parameters.
  • The tool uses the Grouped Iterative Validation based Information Extraction (GIVE) method to extract contextual information on nanoparticle characteristics from 20,000 published articles.
  • In a blinded evaluation, NanoSafari outperformed the baseline model in providing more reliable parameters for nanomaterial design tasks.
  • The research demonstrated the utility of AI-based methods for automated learning from published work to provide accurate scientific references for biomaterial and bioengineering applications.
  • The tool has been incorporated into a database to provide accurate nanomaterial design parameters.
  • Biomedical nanoscientists have validated the accuracy of NanoSafari through bench experiments.
  • The research was conducted by a team of researchers led by Haitao Song and includes Yifan Wang, Yue Teng, Guan Huang, Jingzhe Qian, Hongyu Wang, Shiyan Dong, JongHoon Ha, Yifan Ma, Mengyu Chang, Seong Dong Jeong, Weiye Deng, Benjamin R. Schrank, Adam Grippin, Annette Wu, Jared L. Edwards, Yixiang Zhang, Yuanyuan Lin, and Wilson.

Statistics:

  • 20,000 published articles were used to train the NanoSafari tool.
  • The tool was able to extract contextual information on nanoparticle characteristics from these articles.
  • 117 pertain to the research in the Robots & Machine Learning journal.
  • The research was conducted by a team of researchers from the Shanghai Artificial Intelligence Research Institute Co. Ltd.

Sources:

  • A Generative Artificial Intelligence Copilot for Biomedical Nanoengineering. ACS Nano, 2025.
  • NewsRx. Data on Artificial Intelligence Reported by Haitao Song and Colleagues (A Generative Artificial Intelligence Copilot for Biomedical Nanoengineering). Robotics & Machine Learning. May 26, 2025; p 117.

Read more