For real-world medical and clinical research, the lack of data and distortion (interference, inaccuracy) has confused many medical academics, hospitals and pharmaceutical companies. Just like the “beauty” tools in the short video industry, with the “magic” of AI algorithms, they can remove impurities from photos and make blurry or defective photos clearer.
Baixinghua Technology Bx-SaaS applies semi-supervised and unsupervised learning in AI technology to deal with real-world and clinical data, which effectively complements the missing data and makes the big data closer to the real world, removing the human and force majeure interference in the data sampling process. Meanwhile this data filling and prediction is based on rigorous AI algorithms, which have successfully helped several pharmaceutical companies to get better results in clinical research, and the data analysis results have helped doctors and teachers to publish higher IF articles in clinical drug research.
Baixinghua Technology is an AI pharmaceutical Bx-SaaS platform. The company combines medical clinical data, image data, and real-world big data through machine learning, deep learning, and knowledge graph multimodality and other AI innovation technologies to produce a knowledge graph of pharmaceutical literature, breaking medical data silos and helping AI fill in medical big data to deepen research in clinical medicine, drug mechanism basic medicine, etc. At present, the company’s AI real-world data filling has helped pharmaceutical companies and physician clients successfully publish nearly 100 high-scoring SCI articles.
AI is the outlet for disruptive breakthroughs in traditional medicine
Baixinghua TechnologyAl co-founder & CEO Suhong Wang won several international algorithm competitions first, and worked in the Second Research Institute of China Aerospace Science and Technology Corporation from vision AI. The company’s co-founder and CTO, Zhifeng Xiao, is an associate professor of computer science in Pennsylvania, USA, and has published several pharmaceutical high-scoring SCI in the past two years. both have rich experience in Al algorithms.
Suhong Wang’s love affair with the medical field began in 2014. At that time, Suhong Wang co-founded an AI medical training company to provide AI training for medical students, including AI image recognition, AI-assisted diagnosis. “Rather surprisingly, medical students were quite interested in AI technology, and the demand for knowledge mapping and data fusion was particularly high. In time, we served not only medical students, but also pharmaceutical companies, CROs, etc.”
This made Wang Suhong aware of the huge potential market for AI technology in the medical field. Pharmaceutical companies, hospitals and others have sufficient medical knowledge and are in urgent need of AI technology for medical image interpretation and data processing. For example, AI technology can accurately and quickly process hundreds of thousands of real-world data after its intervention, which brings great improvement to drug development and disease research.
“During those years, the contact with a large number of doctors down the line convinced us that AI technology will move from a ‘better choice’ to a ‘must choice’ for doctors and pharmaceutical companies, and that the application of AI technology in the medical field can be seen as The invention of the steam engine in the industrial revolution. From empirical medicine to evidence-based medicine to AI medicine, AI is the exit for traditional medicine to achieve a disruptive breakthrough.”
Under the AI medical wave, Suhong Wang decided to conduct continuous in-depth research on AI technology in the medical field to produce better results, and established Baixinghua Technology in 2020.
What’s special is that Baixinghua Technology chose to focus on AI “beauty” tools for the medical industry, carry out medical real-world, clinical big data to fill in the repair, and focus on providing Bx-SaaS platform and one-stop medical research services.
“Along with the development of AI, new technologies such as NLP, speech recognition and graph convolution are blooming, and the magic of AI, like the application of beauty in short videos, is also effectively applied to the processing of medical data.” Wang Suhong said.
Clinical research applications need the support of high-quality data, which has extremely high requirements for authenticity, accuracy and granularity. All along, medical data is not standard, while real-world data is missing in the sampling process with large errors and other problems plague the Meta-analysis of medical big data. The data barriers of hospitals have not been effectively broken, and the data formats and standards are different among different hospitals, and even the data of various diseases in various departments of the same hospital are not connected to each other. With the exponential growth trend of medical data, massive data accumulation, it is difficult to analyze by human, and AI technology is needed to make the algorithm more intelligent to analyze data and get more accurate results.
Compared to the medical imaging segment, AI medicine has received less attention in real-world data processing combined with multimodal medical data, but is rapidly rising and is expected to become the second long-term growth point in AI medicine.
Baixinghua Technology has now successfully built Bx-SaaS platform, which is an AI analysis SaaS platform focusing on clinical data, real-world combined with medical and scientific literature, which can effectively and scientifically fill in the correction of medical real-world research, clinical research in the really wrong data.
At the same time, Wang Suhong said that the main threshold AI faces in the scientific research experimentation segment is the data barrier and the lack of data standardization, which is the main factor preventing AI companies from smooth layout. saaS, although encountering bottlenecks in domestic promotion, is accompanied by large-scale medical data collection, national regulatory level recognition, huge demand from pharmaceutical companies and high industry ceilings, making AI solve massive real-world data integration Medical multimodal data is increasingly becoming the immediate need of many pharmaceutical companies. And the standardization of SaaS to cut into enterprise services and provide customized one-stop high-end services has been initially validated as a business model by Baixinghua Technology.
In terms of data, Baixinghua Technology focuses on the application of data multimodality, attaches importance to the inherent logic and interpretability of data, and has established the BxKG database, including more than 1 million real-world/clinical data, more than 14 million SCI medical literature, and 12PB image data, allowing data to generate synergy and break the barriers between data.
In terms of technology, Baixinghua Technology uses its unique BxA technology for multimodal fusion and has developed AI technologies such as ML\NLP\KG to provide deeper, more comprehensive, and more accurate AI analysis for medical science and technology work, helping pharmaceutical companies, doctors, and researchers in basic medicine and drug mechanism research.
Among them, for pharmaceutical companies, long drug development time, high cost and low success rate have been a common problem in the field of drug development, Baixinghua Technology can help pharmaceutical companies to study the mechanism of action, expand new indications, predict clinical trial results and improve the efficiency of drug development. For hospitals and university researchers, Baixinghua Technology can help the research process by providing patient prognosis prediction, patient information management and query services.
At present, Baixinghua Technology has been granted 1 invention patent in the AI field in only 4 months, and 2 invention patents are under application, serving many head pharmaceutical companies and several physician customers, and the related achievements have assisted pharmaceutical workers in successfully publishing more than 50 SCI articles.
Wang Suhong said, after the new crown epidemic, the development of the AI medical field further accelerated, gradually get recognition of regulatory laws and regulations, AI plays an important role in drug indications expansion, new target new mechanism research, evidence-based research, etc. The application of AI in medical imaging is also stepping into deep water, in new scenarios of medical new mechanism research, expansion of drug indications, health management, chronic disease management, etc. The application of AI in medical imaging is worth looking forward to.
“In the future, with the increasing computing power, successive breakthroughs in algorithms and policy support, medical AI will set off a new development climax in the next 2-3 years. In the future, the data and results generated by AI will be more recognized by the regulatory authorities, helping new drugs to be approved and diseases to be treated.”
During the AI medical development opportunity period, Baixinghua Technology will continue to improve the AI medical Bx-SaaS platform next, and effectively unite with medical clinical data, imaging data, real-world big data and literature knowledge graph through AI technologies such as NLP/kg/DL/ML, to provide pharmaceutical companies, doctors and researchers in basic medical and drug mechanism We also strive for more regulatory recognition of AI-generated data and collaborative promotion of pharmaceutical R&D.