Baidu released Baidu's medical brain. Is it easy for products to land on IBM and Google?

Today, under the platform of Zhang Yaqin and Wu Enda, Baidu launched its latest research on artificial intelligence in the medical field: Baidu Medical Brain. Baidu Medical Brain is built by Baidu Medical Division, Li Zheng, general manager of Baidu Medical Division, said that Baidu medical brain will be aligned with Google, IBM and other similar products.

In simple terms, the Baidu medical brain actually consists of two parts: Baidu brain + medical care.

According to Li Yanhong, Baidu's brain is divided into three blocks: artificial intelligence algorithm (very large-scale neural network), computing power (hundreds of thousands of servers to calculate, based on GPU), big data. There are four aspects of ability: the ability of speech (speech synthesis, speech recognition), image capabilities, natural language processing capabilities, and user portraits.

From Li's interpretation of Baidu's brain, it can be seen that Baidu's brain is actually an infrastructure similar to that of a power station. It has to play a role in a specific application scenario. Correspondingly, at the beginning of 2016, Zhang Yaqin also proposed â€œsmart+â€. Development ideas, and medical is one of the application scenarios.

How to combine with the application scenario? Judging from past cases, the general speech is Baidu's brain + related application scenario data = product.

The Baidu medical brain also fits this logic: It develops products through the collection and analysis of medical data and related literature. The role of the Baidu medical brain is divided into two levels: For patients, during the patient's visit, the patient's symptom descriptions are collected, summarized, classified, and organized, and the user's portrait is created for the patient to more fully understand the patient's health status. For the hospital side, the larger the medical record database is, the more helpful it is to recognize the illness and remind the doctor that the patient is more likely to have symptoms and assist the primary doctor to complete the consultation.

Li Zheng, general manager of Baidu Medical Division, said that Baidu has been involved in the medical field for 10 years, and its medical services provide online consultations from simple medical related search to accessing third-party service platforms such as Chunyu doctors and doctors online. Appointment registration and other services, and then to connect people and intelligence.

The first landing project of â€œBaidu Medical Brainâ€ was released as a â€œdialogue robotâ€. According to Baidu introduction:

"Baidu Medical Brain" is an artificially intelligent product design through the collection and analysis of massive medical data and professional literature. It simulates the doctor's consultation process and communicates with users in multiple rounds. Based on the user's symptoms, it may raise questions and verify repeatedly. Give final recommendations. During the process, the patient's symptom descriptions can be collected, summarized, classified, and organized to remind the doctors of more possibilities and to assist the primary doctors to complete the consultation.

Since Li Zheng, general manager of Baidu's medical division, said that they want to align with Google and IBM, the above project is just a test of aquatic products, and Baidu may launch more heavyweight products in the future.

Baidu believes that its own medical platform has two advantages: one is the matching and distribution of data, information and services, and the other is technology.

The first advantage is most evident in the medical dialogue robot project. Baidu's own large medical database can provide patients and physicians with more comprehensive and comprehensive information.

The second technical advantage can be subdivided: such as image recognition, speech recognition, machine learning and other technical branches. Judging from the current case of smart medical solutions, the field of artificial intelligence technology in the field of medical treatment, voice recognition is an important direction. The core revenue source of Nuance, the world's largest speech recognition company, is medical services. Its solution uses speech recognition products to record clinical case information. Statistics show that doctors use Nuance's clinical speech recognition technology products to record data of nearly 100 million patients annually. 94% of medical institutions are considering or have a strong interest in using clinical speech recognition technology.

Therefore, voice recognition is an important cutting-edge direction for Baidu, but the domestic audio giant HKUST has already had a place in this market. Assuming that Baidu is going to enter this market in the future, one of the things that needs to be done is to provide its own medical services. The level of service in the field is followed by how to persuade hospitals to adopt their products.

Back to Baidu's attempt to refer to IBM and Google, the products of both parties not only involve the software system, but also extend their tentacles into the fields of medical device manufacturing and pharmaceuticals. Whether Baidu's medical brain will enter these two pools, which are not touched by the end, cannot be known for the time being.

IBM has been working in the medical field for decades and has invested heavily in healthcare. This year, Watson spent $4 billion to raise artificial intelligence, spending $2.6 billion and $1 billion respectively. Watson â€œfed upâ€ medical data company Truven, medical imaging and clinical system provider Merge Healthcare Inc. With such a long history of IBM's R&D, and with such a large investment, Baidu, with its weak medical foundation, will need a lot of time to learn from IBM.

The original reference to Baidu was Google. It was also started with search. It also made artificial intelligence and medical care. However, in recent years, Google seems to have failed to do well in medical services, even though it has collaborated with major life science organizations. , And frequent acquisitions of related companies, but still can not avoid the reality of frequent medical products frustrated. At this time, Google does not seem to be a ready-made case for learning from Baidu.

To align with IBM and Google Medical: One is out of reach and the other is a failure. In the absence of a suitable reference for the time being, Baidu's road afterwards has to rely more on its own step by step.

In addition, Baidu, which has just experienced the medical turmoil, is handling the pressure of public opinion on the next road. It is also an important issue faced by the medical brain of Baidu.