Unlike many people who are impressed by Philips, this Dutch-based company that sells and services in more than 100 countries around the world is no longer an all-encompassing traditional business giant, but is gradually transforming into a commitment to health. A technology company that provides solutions.
As we all know, as early as 1920, Philips' products have entered the Chinese market. It not only brought thousands of homes to lighting, home appliances, medical , electronic communication and other products, but also integrated technology and design into daily life solutions. In 2016, under the new business structure, Philips has focused its attention on health technology and lighting, and fully grasped the trend of integration of personal health and professional medical care. Not long ago, the company announced the "AI Health Care" strategy. So, what is the motivation for this multinational company to rush into this new track?
He Guowei, CEO of Philips Greater China, told Mobile Technology, “We made a decision five years ago, taking out 10% of annual sales – 1.7 billion euros in research and development, and another 60% of the investment for software. Development. Within the company, one out of every four scientists is engaged in research related to AI big data, and the total number of relevant research is almost 250."
Why "AI+ Medical Imaging" became popular
In China, medical imaging data is increasing at an annual rate of 30%, and the number, work experience and efficiency of imaging doctors are simply not enough to cope with such growth trends. At present, most of the medical image data still depends on manual analysis, and the most obvious defect is inaccuracy. According to Philips' 2018 China Health Outlook survey: the number of professional medical staff in China is scarce, and the distribution of high-quality resources is uneven. The “first accurate diagnosis and treatment†has become a bottleneck affecting the efficacy rate and effect of major diseases, resulting in a higher rate of surgical treatment. 50%. The deep learning method in the field of computer vision has become more and more mature, providing a new opportunity for automatic analysis of medical images and assisting doctors in intelligent and precise diagnosis and treatment. Therefore, “AI+ Medical Imaging†has become the most popular branch of current AI medical.
Relying on the role of Philips equipment supplier in the world, it cooperates with more than 4,000 top hospitals and scientific research institutions to embed artificial intelligence technology into the equipment. From the actual clinical needs of customers, it has become a series of intelligent intelligence development. The best "incubator" for tools. Experts from the Philips Research Institute also disclosed these experimental data to us.
In the research topic of Zhongshan Hospital, radiologists are handwritten in the report, but it takes time for doctors to go back and use the naked eye to identify key information. If you use AI to get a radiology report to the surgery report, and you can do vertical comparison and timeline alignment, let the doctor can do more with less at a glance. The final scientific statistics show that the doctors participating in the experiment saved nearly 50 hours. Among them, the time required for doctors to obtain clinical patients with the naked eye was reduced by 85%, and the efficiency of extracting key points from the complete information of patient cases increased by an average of 50%. This is not a case -
There is also a risk model established through a database of more than 70,000 ICU patients over 15 years, which can even predict organ lesions that occur 3 hours after a critically ill patient. Through the sampling analysis of its own unsupervised equipment, Philips can predict the probability of a fall in the home for one month in advance, thus reminding the family to prevent it.
This is not to exaggerate the effect of AI or data prediction. Apple Watch has previously sent out several times to inform users that the heartbeat is risky and needs to be treated quickly, thus saving a real news like a life.
As a long-established professional medical company with a wide range of clinical pathways, its diagnostic imaging products are devices that generate and collect medical imaging data, and in specific cases, it has formed an experimental alliance with eight top three hospitals, advancing a PercuNav integrated interventional navigation technology, with powerful and intelligent fusion alignment method - anatomical intelligent alignment, using vascular alignment method and liver surface alignment method, quickly complete the fusion of various image data in 15-30 seconds, and accurate High degree. In addition, the US Only approach allows the clinician's operating experience to exceed expectations. PercuNav technology on EPIQ can help doctors perform ultrasound interventions and preoperative and postoperative comparisons in liver, lung, superficial, and musculoskeletal;
Another Ingenia CX Lightspeed Smart Magnetic Resonance is also Philips' "experimental specimen" with an 80/200 Alpha Smart Gradient and revolutionary whole body compressed sensing imaging CS SENSE, which can greatly enhance the body's more than 30 clinical and scientific sequences. Scan speed and the throughput of the scan. Light-speed intelligent magnetic resonance and ISD2.0 artificial intelligence platform seamlessly combine to support clinical research in the fields of nerve, tumor, heart and imaging, and accelerate innovation and transformation.
This cloud is unusual: build AI health care sharing platform
In addition to clinical equipment, Philips is also actively deploying the industry ecosystem – “HealthSuite Insightsâ€, which brings together the artificial intelligence assets of Philips and other companies that can be shared and licensed for use by data scientists, software developers, clinicians. A variety of shared tools and technologies, such as health care providers, and help save time, resources, and costs in developing and deploying medical artificial intelligence solutions. And its project to land in China also has a very local name: Shenfeiyun.
“The platform created by Shenfeiyun and Shenzhou Medical is one of the important achievements of our strategy in China. With 'Shenfeiyun' and 'Philips Nebula System', the 'cloud platform' service model with flexible payment mechanism is realized. Remote means, so that the basic hospitals can provide high-quality health care services based on 'precise diagnosis and treatment' for the general public, thus helping the construction of medical associations and promoting the homogenization of quality medical resources in China's society." He Guowei introduced.
“Philips Nebula 3D Image Data Center†is an intelligent clinical imaging diagnostic system that can realize image processing of different brands and different types of imaging equipment, and provide advanced visual post-processing of multi-modal images to help radiologists and clinicians better. Identify lesions, monitor, diagnose, and follow up on disease treatment. It covers multiple clinical fields of radiology, including cardiology, oncology (liver, lung, breast, prostate, etc.), neurology, emergency, orthopedics, dentistry, etc. There are more than 70 clinical application modules in the field. Based on clinical guidelines, advanced 3D image post-processing technology provides a clearer representation of the body's tissue structure, helping clinicians make rapid, accurate clinical diagnosis decisions based on images, and planning individualized treatment options.
In addition, through the "Nebula platform", it is also possible to build a remote medical imaging center across departments, across hospitals, and across regions, optimize clinical procedures, achieve sharing of image resources across departments, achieve homogenization of quality medical resources, and promote graded diagnosis and treatment. So the big picture of AI's health is naturally a good prospect, but the challenge is not too small -
After all, medicine is a serious science that emphasizes practice, relies on evidence, and is related to the quality of life. How to turn rich image big data into a realistic guiding plan is the next topic that Philips will face.
Serious science is the last step of innovation
As mentioned above, although China has a large amount of medical image data, the whole industry is still in the stage of transition from traditional film to electronic data. A large amount of image data has not yet been digitized, and the data sharing and intercommunication between hospitals is low. The data is a big bottleneck. Only by working with clinically experienced experts and doctors can we train "smart artificial intelligence." Therefore, AI medical imaging requires scientific assessment criteria and systems from data sources, modeling, training to outcome evaluation. For decades, Philips has strictly followed clinical guidelines and medical pathways in these areas, and based on scientific assessment standards and systems, it has repeatedly and rigorously verified to ensure a safe and reliable solution for the clinic.
Of course, due to the large number of barriers in China's medical system, there are a large number of non-systematic, unstructured data in medical data. How to learn these data through health medical machines to get through, how to obtain training data through scientific research combined with actual medical scenarios and further landing? Behind this, it is necessary for the relevant industry and companies to meet the challenges of accepting business models, the assessment of interdisciplinary compound talents, and strict clinical verification, which are unprecedented challenges in the health care industry !
Anesthesia Medical Co., Ltd. , https://www.jssinoanesthesia.com