Los Angeles Smart Healthcare

The development of unmanned driving technology has not only promoted smart transportation, but also brought unexpected gains. According to the report of "Society for Interventional Radiology: 2017 Annual Scientific Meeting", researchers in interventional radiology at UCLA discovered a cutting-edge technology from "unmanned driving". Research experts used this artificial intelligence technology to invent An intelligent medical assistant in interventional radiology. The assistant is able to communicate freely with clinicians and quickly give medically informed responses to common medical questions. With this invention, physicians can easily introduce patients to the basic concepts of interventional radiology treatment and accurate information for patients at various stages in the treatment plan.

Artificial intelligence is advancing rapidly and it is entirely possible to use it in interventional radiology as a low-cost, automated medical assistant to improve patient treatment and care. Today, artificial intelligence has transformed many industries, and it also has great potential in transforming the medical field. In this study, medical assistants using deep learning techniques were required to understand a wide range of clinical medical problems. The research specialists asked him to come up with an appropriate response to each question in a manner similar to a text message conversation. Deep learning technology is inspired by the working mechanism of the human brain. Its artificial neural network can analyze huge data sets and automatically summarize a pattern from it, which can also "unsupervised learning" without human intervention. Deep learning networks are able to analyze complex data sets and provide rich and referential suggestions in early diagnosis of diseases, treatment planning, and condition monitoring.

To perfect the smart medical assistant's functionality, experts from the research team built a knowledge base of interventional radiology by feeding the program more than 2,000 simulated cases of common interventional radiologist consultations. Through this form of learning, the program is able to immediately give the best answers to questions posed by clinicians. The answers to the program's responses contained various forms of information, including medical website addresses, medical infographics, customized medical plans, and more. If the program determines that the questioner needs a human response, it will provide relevant contact information. During the use by doctors, various question-and-answer situations will also serve as learning resources for the program, and the program will continue to learn and improve itself in practice to achieve a more intelligent level. The researchers used a technique called "natural language processing," which was implemented using IBM's Watson artificial intelligence computer. Watson can answer questions in natural language and efficiently implement other important functions of machine learning.