Intelligence and Control-based BioMedicine LAB in UNIST
This figure was made by ChatGPT 4 powered by DALL E3
Wir müssen wissen. Wir werden wissen.
- David Hilbert (1930)
AI-based Medical Device, Medical Digital Twin, Clinical Decision Support, Digital Healthcare
Research Field: BioMedical Artificial Intelligence, Medical Informatics, Software as (or in) Medical Devices and Applications
Modalities: Time-series (BioMedical / Clinical) Data (with or without) Timestamps (with or without Multi-modal data)
Aim: To support Clinical Deicison/Intervention Making and To explain Operations of Medical Models/Algorithms
🤔 Our lab MAY BE looking for (prospective) students and researchers. Please check the Contact section on the website.
우리 연구실은 (잠재적) 학생 및 연구자를 추가로 찾고 있을지도 모릅니다. 자세한 사항은 Contact Section을 꼭 참고하십시오.
Our mission is to utilize and integrate data-based and knowledge-based models for defining and solving various meaningful problems in medicine and biology. We handle diverse types of data, including time series. To achieve our goals, we employ a variety of methodologies such as statistics, control, optimization theory, machine learning, and artificial intelligence. Our work spans from the science of knowledge discovery to the engineering of novel applications, aiming to address unmet demands in clinical medicine and biology and to identify and solve undiscovered needs. We engage in both methodology research and practical field applications, covering aspects from prediction and inference to decision-making and implementation in biomedical sciences and clinical fields.
본 연구팀은 데이터 기반 및 지식 기반의 모델을 활용하고 통합하여 의학과 생물학의 다양한 문제들을 정의하고 풀어내는 것을 목표로 한다. 이를 위해 여러 차원의 시계열을 포함한 다양한 데이터를 다루고, 통계와 제어공학, 최적화 이론 등과 같은 전통적인 방법론을 비롯하여 머신러닝, 인공지능과 같은 다양한 방법론을 깊게 연구하고 활용하고자 한다. 이를 바탕으로 임상 현장에서의 미발견 수요 탐색을 통해, 새롭게 제안하고 수학적 모델링, AI 및 ML 기반으로 해결하고자 한다. "code-to-clinic" 관점에서 의과학 및 임상의학 분야에서의 예측과 추론에서부터 의사결정과 실행까지 방법론 연구 및 실제 분야 적용의 영역에서 탐구하고 검증한다.
You can check our publications here: Google Scholar
We primarily focus on statistics/machine learning(ML)/artificial intelligence(AI)-based methodologies to investigate biomedical/clinical/health 'time-series/sequential' data to innovate our healthcare systems and clinical practices. (However, types of medical/clinical data are not limited to that.)
Clinical Departments of Hospitals in Collaboration: Radiology, General Surgery, Rehabilitation Medicine, Neurology, Emergency Medicine, Internal Medicine, Pediatrics, Oncology.
Projects (On-going)
Deep learning biological molecular subtype discovery for proteomic data in the extracellular endoplasmic reticulum (with Oncology)
Evaluation of driving patterens for senior drivers (with Rehabilitation Medicine)
Infant’s movement analysis by using of deep learning to predict neurological outcome (with Rehabilitation Medicine)
Development of digital twins to simulate glucose dynamics of diabetic patients (with Internal Medicine/Endocrinology)
Deep Learning-based survival analysis with multimodal clinical data for HCC treatment (with Radiology)
Explainable medical image prediction models with prior knowledge (with Radiology)
Uncovering clinical interventions/behaviors via reinforcement learning (with Emergency Medicine)
(Last updated: Nov. 24th, 2024)