We are thrilled to announce the publication of the paper “Accelerating discoveries in medicine using distributed vector representations of words” in the Expert Systems with Applications [link]. Authored by Matheus V. V. Berto, Breno L. Freitas, Carolina Scarton, João A. Machado-Neto, and Tiago A. Almeida, this groundbreaking study introduces an innovative system leveraging word embeddings to transform biomedical research.
The system uncovers latent knowledge from vast medical literature, focusing on Acute Myeloid Leukemia (AML). It demonstrates the ability to anticipate critical discoveries years before their formal publication. By coherently encoding biomedical knowledge, the research opens new pathways for accelerating medical advancements, including a data-driven strategy for faster and more efficient drug testing.
Highlights of the research include:
- A novel embedding-based framework to extract hidden insights from medical texts.
- Demonstrated efficacy in predicting therapies up to 11 years before their official proposal.
- The potential to revolutionize drug discovery and testing in the biomedical field.
This research was supported by FAPESP (grants 21/13054-8 and 22/07236-9). We congratulate the authors for their remarkable contribution to advancing medical science and Natural Language Processing!