From experience to data How can AI continue to write that "the medicine is not as effective as the camphor tree"?

China News Service, Jiangxi Zhangshu, October 19 (Xiong Jinyang, Nie Yanjing) A Chinese herbal medicine is put into the testing equipment. Industrial vision scans the appearance, X-rays look at the inside, and hyperspectral analyzes the ingredients. After a few dozen seconds, a digital "physical examination report" can be generated. At the 57th National Medicinal Materials and Drugs Fair in Zhangshu, held from October 16 to 18, a large AI medicinal material testing model jointly developed by the team of Zhang Boli, an academician of the Chinese Academy of Engineering, and Yangzijiang Pharmaceutical Group was officially released, and intelligent testing equipment equipped with related technologies was simultaneously unveiled at the exhibition.

"If the medicine is not enough, the camphor tree will not be enough, and the medicine will not be effective if the camphor tree is not enough." This proverb that has been passed down for thousands of years points out the unique status of the camphor tree in the "Medicine Capital of China". Why does camphor tree medicine continue to be "effective" after thousands of years? How can an AI model turn thousands of years of experience and reputation into a verifiable standard?

What's the difference between "Qi" and "Ling"?

The quality judgment of Chinese herbal medicines has long relied on the "eyes, hands, noses and mouths" of experienced herbalists. Sensory descriptions such as color, aroma, taste and shape lack unified quantitative indicators, and the judgments of different masters and different batches often vary.

This skill has been passed down by camphor trees for more than 1,800 years. Zhangbang's processing skills are unique, and the reputation of "the medicine is not as good as the camphor tree" is precisely due to the skills of these generations of pharmacists. However, this also means that the quality determination of medicinal materials cannot be standardized and scaled, nor can the source be traced.

Camphor trees are facing this dilemma. At present, the city has gathered 502 pharmaceutical companies, with a planting area of ​​more than 500,000 acres of traditional Chinese medicine. The output of traditional Chinese medicine pieces accounts for 80% of Jiangxi Province, and pharmaceutical distribution and wholesale companies account for about one-third of Jiangxi Province. "Qi" is the traditional advantage of camphor tree, but the other side of "Qi" is the many varieties of medicinal materials, wide sources, mixed batches, and uneven quality. For a long time, the local modern quality inspection and traceability system has not been fully established, making it difficult for good medicinal materials to "prove their innocence."

The "spirit" mentioned by the ancients is the spirit of experience and skill. The "spirit" that we are going to continue to write today needs the support of the spirit of data and the spirit of standards. This constitutes the practical logic of Zhangshu’s introduction of AI testing projects: when the industrial scale reaches this point, quality control has changed from “icing on the cake” to a “must-answer” real test question.

Can the "intelligent physical examination" understand the touch of an old pharmacist?

"Everyone can understand this set of equipment as a 'smart physical examination' for Chinese herbal medicines." Ji Changquan, head of the supply chain management department of Yangzijiang Pharmaceutical Group and general manager of Jiangsu Longfengtang Chinese Medicine Segment, said that industrial vision can identify mildew, insects and surface foreign matter .

According to reports, the equipment takes a proprietary multi-modal AI large model as the core and integrates industrial vision, hyperspectral analysis and X-ray penetration imaging technology, with a comprehensive detection accuracy of more than 90%.

Accuracy is only the basis. This large AI model explores and cracks the most difficult part of the traditional Chinese medicine industry - experience inheritance. "Spectral characteristics, composition indicators, and authenticity determinations, once digitized, can be precipitated, passed on, and iterated. The experience of old herbalists no longer disappears as people age, but becomes callable and traceable data assets." Liu Changqing, senior manager of Tianjin Modern Innovative Chinese Medicine Technology Co., Ltd. said that AI testing does not replace statutory inspection and professional judgment, but combines traditional experience with modern technology to promote more standardized, stable, and transparent quality management of Chinese herbal medicines.

This boundary awareness not only respects the professional rules of the traditional Chinese medicine industry, but also provides a clear framework for the implementation of technology.

According to Xu Haoyu, Secretary of the Party Committee, Chairman and President of Yangtze River Pharmaceutical Group, this model is a "digital intelligence answer sheet handed over by the collaboration of industry, academia and research", which can realize the rapid identification of medicinal materials entering the market and the auxiliary detection of multiple indicators. It can leave traces in real time and trace the entire process, so that high quality and low prices can be evidence-based and build a solid defense line for the quality of medicinal materials from the source of transactions. "It is these front-line accumulation of 'controlling the source and strengthening digital intelligence' that allow us to better understand the needs of the industry and have the conditions and responsibility to apply AI technology to the source of medicinal materials trading."

How far is it from the camphor tree to "buy with eyes closed"?

In September this year, the Longfengtang (Jiangxi) International Traditional Chinese Medicine City project officially started construction in Zhangshu. Jichangquan said that in the future, AI-assisted testing, spot warehousing, logistics distribution, digital traceability and smart trading platforms will be coordinated and promoted to build a modern Chinese herbal medicine trading service system with "spot stock, quality control and chain".

This is a key step for Zhangshu to use technological innovation to empower new productive forces in traditional Chinese medicine. AI detection solves the problem of "verification", warehousing spot solves the problem of "supply", digital traceability solves the problem of "checking", and smart transactions solves the problem of "selling clearly". The combination of several things can form a complete chain of trust.

According to the plan, the project will build a credible, controllable, and traceable physical transaction scenario offline and an integrated smart transaction service platform online, connecting the entire chain of production areas, markets, enterprises, logistics, and finance. It will focus on solving pain points such as scattered sources of high-quality goods, insufficient guarantee of spot goods, shoddy goods, and disordered prices, so that people can "buy medicinal materials with closed eyes" in Zhangshu.

Talking about why he chose Zhangshu to invest, Jichangquan said that Zhangshu has a history of more than 1,800 years of development of traditional Chinese medicine. It not only has Zhangbang's processing skills, but also has a national-level professional market for traditional Chinese medicinal materials, a pharmaceutical trade fair platform and relatively complete industrial supporting facilities. “Not only does it have history, but it also has a strong desire to promote the transformation and upgrading of traditional industries.”

"Buying with eyes closed" expresses consumers' expectations for good medicinal materials. Jichang Quan said: "What we have to do is to implement standards, data and responsibilities in every link, so that consumers' trust can be based on."

The release of a model does not mean that the industry's chronic problems will be automatically eliminated. Deep-seated pain points such as scattered high-quality supply, insufficient spot protection, and disordered prices still require the joint efforts of technology, systems, and markets to gradually solve them. The scientific and technological answer sheet for the "Millennium Pharmaceutical City" has just begun. From "based on experience" to "based on data", from "famous endorsement" to "quality and credibility", the old proverb "the medicine is not as effective as the camphor tree" is ushering in a new footnote.