Introduction

Introduction

Biology and Medicine

  With the development of medical and health care, the average life expectancy of Taiwan's population has been increasing. In 2009, the average life expectancy of women was 79.70 years and that of men was 73.47 years. The country where the proportion of people over the age of 65 accounts for more than 7.0% is called the aging country. In 1994, Taiwan’s population over 65 years old accounted for 7.23% of the total population, and officially became the aging countries. However, the birth rate of newborns in Taiwan has been declining year by year. According to the 2014 population structure analysis, the population over 65 years old accounted for 11.6% and the age of 0-14 years accounted for 14.7%. According to the population structure analysis of 2051, the population over 65 years old will account for 35.5%. The 0-14 years old will account for 8.9%. According to the analysis of health care manpower of each person over 65 years old, in 2003, there were about 6.73 20-64 years old middle-aged people care for each person over 65 years old. By 2046, there were about 1.61 20-64 years old people for each 65-year-old or older. Health care industry is facing the crises of the increasing number of health care resource users and the decrease of the supply of medical care manpower. Aging society with fewer children are the social trends of the advanced countries. Therefore, the investment of AI resources should be based on long-term home care and telemedicine. On the other hand, image detection is an indispensable part of modern medical diagnosis and treatment. According to statistics, 80% of medical data comes from medical imaging, and 70% of clinical diagnosis requires medical imaging. By applying artificial intelligence to medical imaging, objective and quantitative results can be provided, which can effectively improve the accuracy of diagnosis and reduce the pressure on doctors. In addition to accurate diagnosis and treatment advancement, there are many unmet clinical needs - future prediction and preventive intervention. For example, developments of many chronic diseases (diseases related with living habits such as cancer or diabetes) are relatively slow and have no clinical symptoms. If the predictive model can estimate the risk of future disease, it will help to avoid complications.
  
  Another example is the unintended re-admission of patients and the unpredictable first-aid of patients in the hospital. It has always been a clinical headache, not only delaying the disease, increasing the cost of medical care, but also increasing the risk of medical disputes. How to use various clinical parameters to identify patients with high risk of various diseases in an objective manner is an important issue in the future. In addition, many diseases have symptoms or abnormal blood testing results when the disease has progressed to a considerable extent, and damage to organs has already been caused. Traditional diagnostic methods are difficult to predict future disease or prognosis of existing diseases. Our research team also hopes to construct a metabolomics-based disease occurrence and prognosis prediction model based on community and combine it with clinical data to verify the accuracy of the validation model. In the medical field, we want to develop disease prediction models through AI, and effectively make accurate risk assessments for early intervention. In terms of intensive care, the NTU medical team can complete the Yeke Membrane Emergency System device in a short period of time. The total number of cases in which the heart stops beating and the use of the Yeke Membrane First Aid in NTU is half of the world, and the patient survival rate is doubled compared with the traditional CPR technology. These medical achievements were also published in the Lancet magazine in 2008. According to the above excellent performance, The Biology and Medicine Group of National Taiwan University and NTU Hospital propose the following research directions:

  1. Using artificial intelligence to establish an interactive platform for three-dimensional spatial air pollution distribution model and home remote care

  2. Develop disease prediction models through artificial intelligence and construct disease occurrence and prognosis prediction models by metabolomics

  3. Establishment, authentication and clinical application of artificial intelligence medical image assisted diagnosis system

  4. Using deep learning method to determine the depth of anesthesia sedation platform

 
We will use AI and cloud big data to improve human health and well-being from home care, health maintenance, early disease and imaging diagnosis to critical care.

Leader
Members