Yitu Healthcare Draws 'Cancer Prevention Map' to Bring AI-Powered Early Screening to Grassroots Communities
With early detection, the 10-year survival rate for early-stage breast cancer can exceed 80%, and the breast-conserving surgery rate can reach 40–50%.
Unfortunately, 75% of lung cancer patients in China are already at an advanced stage at the time of diagnosis, with a five-year survival rate of less than 20%; furthermore, fewer than 10% of breast cancer cases are detected at an early stage. The large number of patients with advanced-stage cancer not only imposes a heavy economic burden on patients and their families but also causes immense psychological distress, leading to countless social tragedies.
How can we ensure that early-stage cancers in their nascent phase are detected without fail? How can we enable more cancer patients to enjoy a higher quality of life and extended survival? How can large-scale early screening initiatives support frontier research into the pathogenesis of lung cancer?
From May to July 2018, in Shilou Town, Panyu District, Guangzhou, the Second Affiliated Hospital of Guangzhou Medical University conducted a low-dose spiral CT screening program for early detection of lung cancer. Within just three months, health screenings were performed on tens of thousands of residents, with 1,323 individuals completing early lung cancer screening. Ten cases of lung cancer were subsequently identified, nine of which were early-stage and confirmed by pathology.
At a time when radiology departments in large tertiary hospitals are already severely understaffed and overwhelmed by the daily demand for image interpretation, who exactly supported the launch of this large-scale lung cancer early screening campaign, which lasted several months and covered tens of thousands of people? Did the doctors somehow grow three heads and six arms?
On November 9, at the launch event of the “AI Cancer Prevention Map” initiative hosted by Yitu Healthcare, Professor Zhang Zhenfeng, Deputy Director of the Department of Radiology and Director of the Minimally Invasive Intervention Center at the Panyu Campus of the Second Affiliated Hospital of Guangzhou Medical University, revealed the answer—artificial intelligence.
Professor Zhang Zhenfeng revealed that, supported by advanced medical artificial intelligence technologies, the “AI + physician” early screening model has significantly improved screening efficiency, enabling tumor early screening to be conducted efficiently, accurately, and reliably. This approach facilitates the timely detection of early-stage lung cancer, extends survival duration and improves quality of life for cancer patients, and provides abundant data for future large-scale frontier research on the pathogenesis of lung cancer.
This visit to Panyu served as the catalyst for Yitu Healthcare’s Anti-Cancer Map. While tiered diagnosis and treatment may have previously remained merely a rhetorical concept, repeated grassroots initiatives have profoundly highlighted to Yitu the scarcity and inadequacy of primary healthcare resources. Consequently, Yitu has re-evaluated the pivotal role that AI technology can play in this era—not only by assisting in diagnosis but also by leveraging big data to uncover the underlying causes behind high disease prevalence.
At the 2018 CCR Conference, Yitu Healthcare announced the official launch of the “AI Cancer Prevention Map” project. The company plans to invest RMB 100 million in project funding over the next five years, collaborate with hundreds of medical institutions, and cover 19 provinces, municipalities, and autonomous regions. By leveraging AI applications to enhance the service delivery capacity of medical institutions, strengthen research on cancer populations, and optimize early screening methods, the initiative aims to propel China’s tumor screening into the “AI+” era.
“While RMB 100 million is a modest amount, as a startup, we will make every effort to advance the development of cancer screening,” said Ni Hao, President of Yitu Healthcare, at the press conference. “The ‘AI Cancer Prevention Map’ primarily consists of two components. First, we aim to provide technological and financial support to hospitals offering screening services, thereby enhancing their screening capabilities and accelerating the expansion of cancer screening coverage. Second, we will collaborate with hospitals on certain research initiatives, such as establishing prospective cohorts to conduct follow-up studies on cancer patients, and attempting to derive insights from a genetic perspective to uncover patterns of disease among cancer populations. Through these research projects, we hope to guide and optimize the implementation of early cancer screening.”
Jin Zhengyu, Director of the Department of Radiology at Peking Union Medical College Hospital, also addressed the issue of early cancer screening at the conference: “On one hand, factors such as the stress induced by today’s fast-paced lifestyle and environmental pollution resulting from urban development are significant contributors to cancer. On the other hand, public awareness regarding cancer, as well as the emphasis on early detection and prevention, still needs to be enhanced. Both Yitu and our hospital must work together to address these challenges.”
Not every enterprise is capable of undertaking anti-cancer initiatives on such a scale; achieving these goals requires robust core competencies as foundational support.
Yitu has long been committed to innovative research in artificial intelligence, integrating advanced AI technologies with industry applications to build a safer, healthier, and more convenient world. Its products currently serve numerous sectors, including urban public safety, healthcare, and finance. In the fight against cancer, Yitu has selected the following products as its weapons.
In the AI Cancer Prevention Map, lung cancer is the greatest enemy. Data show that the 10-year survival rate for early-stage lung cancer reaches 92%, whereas 75% of patients are diagnosed at an advanced stage. There is a significant difference in five-year survival rates between early- and late-stage disease. Therefore, early diagnosis and treatment can substantially reduce patient mortality, carrying immeasurable significance for patients themselves, their families, and society as a whole.
The care.ai™ intelligent imaging diagnostic system for lung cancer, developed by Yitu Healthcare, was created in collaboration with over 300 physicians. Adhering to the Fleischner Society Guidelines, NCCN Guidelines, and Chinese expert consensus recommendations, and trained on more than one million datasets, the system has achieved high detection sensitivity and a high rate of report adoption.
Today, this system not only enables rapid, precise, and reliable detection of pulmonary nodules within seconds but also provides functionalities such as precise lesion localization, differentiation between benign and malignant nodules, and automated comparison with historical imaging. It seamlessly integrates into clinical workflows by generating structured reports, achieving a sensitivity of over 95% and a clinical adoption rate of 92% for these structured reports. Truly serving as an assistant and “second brain” for clinicians, its overall performance surpasses that of most competing products.
Statistical data from the National Cancer Center of China shows that among all malignant tumors potentially affecting women, breast cancer ranks first, accounting for 16.51%. Globally, breast cancer has the highest incidence rate among malignant tumors in women in both developed and developing countries, making the breast an important application scenario for AI in medical imaging.
However, the development of AI products for breast imaging is significantly more challenging than that for pulmonary nodules. This is particularly true for mammography, as there is an extreme scarcity of radiologists in China who specialize in interpreting mammograms, far failing to meet clinical demand.
Yitu Healthcare’s intelligent diagnostic system for mammography, developed over several years, delivers robust performance. Leveraging advanced algorithmic innovations and a vast dataset of tens of thousands of real-world mammographic images from multiple Grade A tertiary hospitals, this AI system achieves second-level interpretation of mammograms. It offers multiple functionalities, including glandular tissue classification, lesion detection, feature description, and intelligent BI-RADS categorization, and can automatically generate structured reports for use by radiologists.
These two product categories will play a significant role in oncology treatment. With the support of Yitu AI products, more primary healthcare institutions will gain access to therapeutic capabilities that were previously unattainable. Through these products, more people will be able to experience AI firsthand, thereby accelerating the widespread adoption of early cancer screening in China.
The advancement of healthcare is a long journey that cannot be accomplished overnight. The “AI Cancer Prevention Map” project marks a promising start. Ni Hao stated, “The first batch of 19 cities falls within the national framework for cancer prevention. Wherever the country launches early cancer screening initiatives, we provide our support. However, we have also invested significant financial resources in many regions outside these 19 provinces and municipalities, particularly in more remote areas. As long as there are plans to launch early cancer screening projects, Yitu is willing to provide technology, funding, and systems to support them. Meanwhile, our research efforts will continue, as human understanding of cancer remains far from sufficient. We hope to identify underlying issues through data analysis.”
“Yitu is fully prepared in terms of mindset. Of course, Yitu alone is not enough to change the current state of healthcare. We hope to work together with doctors and numerous life science companies to jointly promote early cancer screening, enabling society as a whole to experience the value that artificial intelligence brings to healthcare.”