Seoul: LG AI Research, the artificial intelligence lab under South Korea's LG Group, announced the launch of its upgraded pathology AI model, EXAONE Path 2.0. This new version promises significant improvements in the analysis and prediction of genetic mutations, gene expression patterns, and subtle structural changes in human cells and tissues, all derived from pathological tissue images.
According to Yonhap News Agency, the EXAONE Path 2.0 model has been trained on a robust dataset comprising over 10,000 entries, where whole slide images are paired with multiomics information. This extensive training allows the model to predict gene activity through image analysis alone, eliminating the need for costly genomic testing. Park Yong-min, a lead researcher at LG AI Research, highlighted the efficiency of the model, stating that genetic testing can now be completed in under one minute, a significant reduction from the previous timeline of more than two weeks. This advancement is expected to critically impact cancer patient treatment by securing the 'golden time' for interventions.
In addition to its speed, the EXAONE Path 2.0 model offers practical applications for doctors and pharmaceutical companies. It aids in the rapid analysis of cancer patients' pathological images, facilitating the identification of gene mutations and the selection of suitable targeted therapies. This capability is expected to streamline processes within the medical field, enhancing the delivery of personalized medicine.
Meanwhile, LG AI Research has also announced a partnership with Professor Hwang Tae-hyun from Vanderbilt University Medical Center. This collaboration aims to develop a multimodal medical AI platform. Professor Hwang, who is the founding director of the Center for Molecular AI at Vanderbilt, focuses on integrating artificial intelligence with molecular biology to advance areas such as cancer research, transplant science, and precision medicine.