In recent years, medical technology has entered the clinical, home, and care fields in large numbers, allowing smart medical applications that were originally only electronic and mobile, because of ultra-low latency 5G technology, artificial intelligence (AI) image recognition, and virtual reality. A major step forward has been made in technological breakthroughs such as virtual surgery and IoT medical equipment. How can medical institutions, medical material manufacturers, health care providers, and technological innovations use the latest medical technology to create a healthier world for everyone? ?
Although the emergence of 5G has solved the problems of smart medical transmission delay, insufficient bandwidth, and limited Internet of Things device connections, “only highways are useless, and more importantly, they have to carry the corresponding services.” Then, what scenarios are What about smart medical applications that may be implemented in the next 10 years?
5G realizes remote surgery without waiting for a bed
Kang Junhong, director of the Artificial Intelligence Medical Research Center of Taipei Medical University, believes that de-localization may be the biggest change. Today, 400,000 people go to see doctors in Taiwan every day. In the future, when the environment of home medical care, mobile medical care, and teletherapy is mature, everyone can see a doctor without going to a hospital or clinic, which can greatly reduce transportation time and costs. In fact, Japan has already After 10 years of work, the focus is not on remote consultations, but on creating interactive scenarios that hospitals don’t have. For example, home medical care can be used with data collected by AIoT sensors at home, so that doctors can get the patient’s home like an immersive experience. Data, master more accurate and complete health and medical information.
In addition, the extension of the emergency medical field is also a major change. In the past, emergency medical treatment could only be done in the emergency room. However, if 5G ultra-low latency characteristics can be used, emergency medical treatment and passport protection can be carried out remotely. , Can improve treatment efficiency.
If telemedicine can be popularized smoothly, physicians can better grasp the golden rescue opportunity for patients in emergencies.
Kang Junhong believes that telemedicine or telesurgery must be fast and accurate, and information exchange is the key, but all data transmission delays must be minimized, and the network must maintain high stability in order to provide timely feedback to complete the operation; Since medical treatment and surgery are feasible, hospitals no longer have to carry out an arms race of hundreds or thousands of beds, and some patients can be well cared for at home.
However, there are still many obstacles in clinical application of 5G medical care. Citing statistics from PC Magazine, Kang Junhong stated that 61% of respondents believe that cost is the biggest consideration. In addition, medical terminal equipment, law, medical culture, and medical evidence will also affect the acceptance of medical staff and patients; these factors Among them, the background culture and application field practice are still the primary considerations. After all, these are “not applicable everywhere”, and very “local” design thinking and data must be adopted.
AI helps assist diagnosis and improve the quality of medical decision-making
With the wave of AI and deep learning sweeping into the medical field, many medical research institutions or technological innovations have invested in the development of AI image recognition and interpretation models for various diseases to assist clinicians in interpreting images, improving the quality of medical decision-making and the accuracy of diagnosis degree.
Changjia Intelligence, derived from China Medical University, combines the professions of AI solution architects, clinicians, and a large number of data scientists to create smart medical technology services using AI as tools, and cooperates with Acer, HP, Shanghai Bank, and Huida (Nvidia), Trend Micro, Yunda and other technology manufacturers, built their own high-speed computing center and AI development software, and cooperated with the affiliated hospital of China Medical University, using a large number of medical databases, training and developing more than 80 medical AI interpretation models, The medical AI-assisted diagnosis system has been introduced to 20 medical institutions.
The powerful analytical power gives AI the opportunity to become the most powerful assistant of clinicians.
Huang Zongqi, CEO of Changjia Intelligence, said that the application of medical AI to image recognition is developing rapidly, which can effectively assist clinicians to quickly identify diseases, reduce the time and medical costs spent in the traditional process of doctors judging lesions, and also increase clinicians Interpret the image efficiency, reduce the difference between different readers in the examination, and provide more accurate diagnosis and quality of consultation.
For example, diabetic retinopathy must be evaluated by fundoscopy images, and then based on the degree of blood vessels, hemorrhage and spots. Usually doctors will spend a few minutes to evaluate based on experience, but different doctors do There may be some differences in the evaluation level; now the AI diabetic retinopathy interpretation system can help physicians more quickly determine whether there is diabetic retinopathy, and it can also evaluate glaucoma, cataracts and other eye problems. If this system can be applied It is even more useful when you go to non-ophthalmology departments such as home medicine, internal medicine, and metabolism.
At present, Changjia has developed the “Automated Bone Age Assisted Interpretation System”, “Ultrasonic Breast Tumor Assisted Classification System”, “Liver Health Assessment Management System”, “Flatfoot X-ray AI Assisted Interpretation System”, “Liver Cancer Computed Tomography-Tumor Range Detection Medical AI-assisted applications such as “Retinopathy”, “Retinopathy Auxiliary Judgment System”, “Thoracic X-ray Abnormal Classification Screening”, and “ECG Assisted Interpretation System” have all entered the verification or clinical trial stage.
In addition, a set of “Smart Medical Aided Diagnosis Solutions” has been developed. Whether it is computed radiography, computed tomography, ultrasound, magnetic resonance imaging, etc., or blood tests, electrocardiograms and other test values, it can be used as smart medical assistance The model training material of the diagnostic system can be used in all parts of the human body.
Use MR to learn precise anatomy, flip the medical teaching mode
Many medical students have to go through a very difficult general anatomy course, but now with virtual reality and augmented reality (AR) technology, whether it is a student or an intern, just wear VR glasses or use a mobile phone. Interactive learning in the lifelike human body model and anatomy teaching environment completely flips the single and boring teaching method of the past flat teaching materials.
Use MR to learn precise anatomy, flip the medical teaching mode
Many medical students have to go through a very difficult general anatomy course, but now with virtual reality and augmented reality (AR) technology, whether it is a student or an intern, just wear VR glasses or use a mobile phone. Interactive learning in the lifelike human body model and anatomy teaching environment completely flips the single and boring teaching method of the past flat teaching materials.
The continuous advancement of virtual reality technology in recent years has also brought new opportunities for medical care.
The DeepQ team of the health care division of HTC has joined hands with the National Defense Medical College to create the largest Mixed Reality (MR) anatomy classroom in Taiwan. As long as students wear HTC VIVE head-mounted displays, they will be able to Quickly switch with the virtual anatomy classroom environment, and can be immersed and repeated autonomously to learn accurate anatomy textbooks, and can be matched with anatomy atlas to quickly learn the corresponding relationship of the human body.
“The MR mixed reality learning method can release the teacher’s teaching imagination and provide a key tool for teachers to flip teaching, so that teachers and students can have closer communication and interaction.” said Zhang Zhiwei, general manager of DeepQ, HTC’s health care division. When teachers are in class, they can combine the virtual space with the real environment to break through the shortcomings that VR cannot directly interact with students. Observe the students’ understanding of the course from the teaching interaction, and make the process of learning anatomy more effective.
Currently, DeepQ’s 3D Organon mixed reality anatomy teaching application can present thousands of static human models. Through disassembly and interaction, they can fully understand the three-dimensional concept of the human body’s spatial structure. There are also hundreds of dynamic models for students Learn the movement of joints and muscles and even the beating pattern of the heart. You can also learn microanatomy. The villi, chylous ducts, muscle fibers, and glomeruli of the small intestine are clearly visible; on the other hand, students can also evaluate their own learning status , Through the VR anatomy test questions, you can dissect the treadmill test in the three-dimensional anatomical space.
