- 아이젠사이언스-휴니버스글로벌, 병원 업무 보조 대형언어모델 보급을 위한 연구 협력 양해각서 체결 > 뉴스 - 스타트업 커뮤니티 씬디스
- 클라우드 병원정보시스템 ‘PHIS’와 의료 특화 대형언어모델(LLM)의 융합 아이젠사이언스와 휴니버스글로벌이 병원 업무 보조 대형언어모델(LL…, 스타트업에 종사하시는 여러분들의 놀이터 씬디스는 스타트업 커뮤니티 입니다.
Integration of Cloud Hospital Information System ‘PHIS’ and Medical-Specialized Large Language Model (LLM)
Aigen Sciences and Huniverse Global announced on the 23rd that they have signed a memorandum of understanding for research cooperation to develop and distribute a large language model (LLM) for assisting hospital work. The main purpose of this agreement is to apply Aigen Sciences' medical-specialized LLM technology to Huniverse's cloud-based hospital information system (PHIS) to provide advanced medical services.
Huniverse's cloud hospital information system, PHIS, is the first system to be introduced to a super-specialized hospital in Korea. It began with the world's first introduction of a cloud hospital information system to a super-specialized hospital with over 1,000 beds, implemented in three super-specialized hospitals of the Korea University Medical Center (Guro, Ansan, and Anam hospitals). PHIS is equipped with all the core functions of a large hospital's computer system and has a modular structure that can meet the diverse needs of different hospitals.
PHIS is a system based on cloud technology that enhances the efficiency of hospital operations and drives digital innovation. It establishes a PHIS clinical terminology standard system that meets international standards, and its flexibility and scalability allow it to be applied to various medical institutions.
Through this research collaboration, applying Aigen Sciences' medical-specialized LLM to PHIS is expected to significantly improve the efficiency of patient care and medical administration within hospitals.
Aigen Sciences' ‘AIGEN InSight’ is a small-scale LLM that minimizes costs by significantly reducing the number of parameters and improves accuracy through fine-tuning. This model consists of 7 billion parameters, making it lightweight enough to operate on a single computer. It was the first to overcome the US Medical Licensing Exam, which models of similar size have failed to achieve. AIGEN InSight achieved a high score of 74, comfortably surpassing the average passing score of 60, compared to MediTron-7B, the previous best-performing 7 billion parameter model, which received a score of 52.
AIGEN InSight possesses multi-step reasoning capabilities necessary to solve complex medical problems, and in seven medical benchmark performance evaluations, it showed an average of 13% higher performance than OpenAI's GPT-3.5 (175 billion parameters), which has a significantly larger scale.
Through this collaboration, the integration of PHIS and a medical-specialized LLM is expected to give rise to innovative hospital work support AI that assists medical professionals with clinical decision support in medical settings, the organization of non-standardized medical charts, patient consultations, scheduling, and other medical administrative tasks.
Aigen Sciences CEO Jae-Woo Kang stated, “We expect the integration of Huniverse's PHIS and Aigen's medical-specialized LLM technology to significantly improve the efficiency and quality of medical services.” He added, “This will enable hospitals to not only achieve cost savings but also implement more accurate and efficient healthcare services, while providing patients with a faster and more friendly new user experience.”
Huniverse Global CEO Sang-Heon Lee said, “Aigen Sciences' AIGEN InSight is a small but exceptionally powerful language model that addresses the issues of computing power and energy consumption that have plagued conventional large language models, and we expect it to be useful for repetitive tasks and information generation in hospital settings.” He also said, “We believe that the synergy between AIGEN InSight and PHIS will reduce the workload of medical staff, enhance the quality of patient care through accurate diagnostic support, and contribute to medical innovation that is felt by the public.”
Huniverse Global Introduction
Huniverse Global developed PHIS, the world's first cloud-based precision medicine hospital information system for super-specialized hospitals, through the national strategic project ‘Precision Medicine Hospital Information System Development Project’ in 2017. It is evolving into a big data platform for personalized precision medicine. The PHIS developed by Huniverse Global has developed all information generated in hospitals, including outpatient and inpatient care, medical services, and electronic medical record systems (EMR), into 38 standard modules and applies them to various sizes of medical institutions. Starting with three super-specialized hospitals of the Korea University Medical Center (Anam, Guro, and Ansan hospitals) in 2021, PHIS has expanded to Asan/Cheonan Chungmu Hospital in 2022, Seoul Samsung Hospital and Donggang Hospital in 2023, and Wonju Severance Christian Hospital in 2024. By continuously generating multi-level and multi-institutional standardized medical big data through PHIS's clinical terminology/standard system that complies with international standards, it aims to improve the domestic medical big data ecosystem to the highest level globally.
Aigen Sciences Introduction
Aigen Sciences is an artificial intelligence (AI)-based new drug development company established on April 13, 2021. It was founded based on core technologies developed by CEO Jae-Woo Kang who led the DMIS research team at Korea University's Department of Computer Science, achieving seven victories in various international bio-medical AI competitions over the past seven years, outperforming global leading AI research institutions such as Google, Stanford, MIT, Janssen, NVIDIA, and Siemens. Aigen Sciences is currently undertaking 14 in-house drug development projects, including cancer and rare/difficult-to-treat disease treatments. It aims to grow the company through the creation of value from technology transfer (license-out) of self-developed drug candidates and securing profits (profit sharing) from the commercialization of co-development pipelines, playing a vital role in opening the path for AI-driven work innovation across the pharmaceutical and medical industries.
Huniverse Global: https://huniverse.co.kr/
Website: https://www.aigensciences.com/
Contact
Aigen Sciences
Jae-Woo Kang, CEO
02-6952-3530
Comments0