LG Group is producing visible results in the field of bio artificial intelligence (AI), which it has identified as a future growth area. It is reported that they have showcased research and development (R&D) achievements at an international conference and have begun full-scale commercialization activities.
LG Group's AI think tank, LG AI Research Institute, installed an exhibition booth at the American Society of Clinical Oncology (ASCO) 2025 conference held in Chicago, United States, from 3rd to 7th (local time). They presented the cancer diagnostic AI model 'Exa One Pass' to cancer researchers worldwide. This is the first time LG AI Research Institute has participated in ASCO, the largest cancer conference in the world.
Regarding their participation in ASCO, an LG Group official noted, "The purpose is to systematically promote the cancer diagnostic AI platform developed by LG in the global market and to enhance collaboration opportunities with global pharmaceutical companies."
'Exa One Pass' has learned from 73,000 images of patient tissue biopsies. Genetic decoding data was also analyzed alongside. The company explained that the AI was able to determine which genes had abnormalities and how they would respond to chemotherapy just by looking at the images.
Even for the same type of cancer, the response to chemotherapy varies depending on each patient's genetic mutations. Typically, genetic mutations are analyzed to find suitable chemotherapy drugs, but this means that by simply providing the tissue pathology images to the AI, the treatment response can be predicted immediately.
In the past, when actual cancer patients participated in clinical trials of new chemotherapy candidates, pathology analyses were conducted over 2 to 8 weeks to determine if they were suitable subjects, but the company explained that with 'Exa One Pass', results can be derived within one minute.
LG Group's subsidiary, LG CHEM, has independently developed an AI model to predict growth after treatment with growth hormone. This AI model combines and recombines several deep learning models, having learned from treatment data of 3,045 children with growth hormone deficiency. Deep learning is a field of AI machine learning that uses multilayer neural networks mimicking the structure of the human brain to learn and predict data.
The growth hormone treatment 'Eutropin', which LG CHEM launched in 1993, is a core drug driving revenue growth in its life sciences business. By incorporating AI technology, they have expanded their reach in the growth hormone treatment field to include diagnosis and prediction.
Last month, LG CHEM announced its technological development achievements through a poster presentation on 'Development of an AI-based Growth Prediction Model' at the European Society for Pediatric Endocrinology and the European Endocrinology Congress held in Copenhagen, Denmark.
The research utilized actual growth data from 550 patients as AI performance validation data to evaluate the stability of predictions for treatments within 1 to 3 years. According to the research results, the AI model was able to predict the growth extent after one year of treatment with an average error margin of 1.95 cm, based solely on the height measured during the first consultation and the prescribed dosage.
LG CHEM plans to further stabilize the prediction performance of its AI model for commercialization as a physician-exclusive AI platform. Yoon Soo-young, head of life sciences and management innovation at LG CHEM, stated, "It is significant that we have developed a growth prediction model tailored to our country’s children, based on data from domestic growth hormone deficiency patients," and added, "We expect that this achievement will serve as a compass for treatment in actual clinical practice."
Since the appointment of Chairman Koo Kwang-mo, LG has identified 'ABC' as a future growth area. ABC stands for AI (artificial intelligence), Bio (biotechnology), and Clean tech (clean technology), and LG plans to invest 100 trillion won in these areas over the next five years, from 2024 to 2028. Recently, evaluations have emerged that they have begun to produce visible results in AI, integrated with the pharmaceutical and biotech sectors.
In his New Year's address this year, Chairman Koo expressed the intent to foster the bio business, one of the future growth engines, saying, "I will challenge the future, to be able to spend more time with loved ones through innovative medicines that treat incurable diseases."
According to group officials, various institutional sectors within the subsidiaries, as well as external organizations, are pursuing a strategy of extensive collaboration. The LG AI Research Institute has been jointly developing AI for diagnostic and therapeutic applications in Alzheimer's disease and cancer with the Jackson Laboratory (JAX), a renowned nonprofit genomic research institute in the United States, since early last year.
They are developing next-generation protein structure prediction AI in collaboration with Professor Baek Min-kyung from Seoul National University. Professor Baek developed the AI protein structure prediction program 'RoseTTAFold' along with Professor David Baker from the University of Washington, who won the Nobel Prize in Chemistry last year.