Sun Yat-sen University Cancer Center: Providing Advanced Digital Pathology Services for Southern China

In the famous story of "Bian Que Meeting Duke Huan of Cai", Bian Que demonstrated his extraordinary talent for diagnosing diseases by observing the Duke during their four encounters and detecting the progression of his disease. As he explained to the Duke, the illness developed in stages—first affecting the veins of the skin, then the muscles and skin, followed by the digestive system, and ultimately reaching the bone marrow, making it incurable.

While traditional diagnostic methods are crucial in modern TCM, Sun Yat-sen University Cancer Center based in Guangzhou, Guangdong Province, is also a pioneer of digital pathology, which is transforming cancer diagnosis.

Freeing Pathology and Its Potential

Esophageal cancer is one of the deadliest forms of tumors, and according to the World Health Organization, half of the global cases occur in China.

Conventional clinical treatment using dual-drug chemotherapy achieves only about a 30% objective response rate (ORR), with less than one year of survival for advanced cancer patients.

In this ongoing struggle, someone always needs to step forward. A research team conducted a clinical study lasting several years, and found that adding PD-1 antibody combined with paclitaxel and cisplatin to the chemotherapy group could significantly extend the overall survival of patients to over 15 months, with an ORR of up to 72.1%.

This China-developed solution using chemoimmunotherapy became a milestone that changed the global treatment landscape for esophageal cancer.

It was supported by the research team from Sun Yat-sen University Cancer Center, which is one of the largest cancer hospitals in China and the backbone of the nation's cancer prevention and control efforts.

For a long time, pathology was a critical, indispensable part of clinical diagnosis and scientific research in cancer.

As the saying goes, "seeing the big picture from small details and knowing the end from the beginning," pathologists have always been making big decisions based on even the smallest of details, and this is highlighted by the fact that over 90% of cancer diagnoses rely on pathological examination results.

Pathologists sample, prepare, and analyze tumor tissues to determine the nature of the tumor, its primary site, degree of differentiation, and potential responsiveness to treatment. Based on these findings, they help guide the treatment plan and prognostic assessment.

The adoption of digital pathology is rapidly expanding in the current digital era, and Sun Yat-sen University Cancer Center is leading the way in this field. It pioneered the shift from traditional microscope-based slide analysis to digital diagnosis through the implementation of slide scanning and online viewing platforms. Digital slide images are easier to collect, retrieve, and manage, overcoming time and space limitations.

The center did not achieve this transformation to digital pathology overnight; rather, it faced numerous requirements and challenges along the way.

Adequate system performance: Pathological diagnosis requires a detailed analysis of each slide, and a single digital slide can easily exceed 3 GB. The poor read performance of the existing data infrastructure significantly slowed retrieval speeds, as a single slide would take at least 30 seconds to load. Performance deteriorated even more when multiple doctors within the hospital simultaneously retrieved slide images, often resulting in pixelation and frame freezing.

Large storage capacity: A digital slide image, despite being the size of a fingernail, typically has a file size comparable to that of a high-definition movie. Currently, Sun Yat-sen University Cancer Center generates more than 6,000 pathology slides daily, with an annual growth rate of 30%. As image resolution gets increasingly higher and retention periods gradually increase, the data center will face significant long-term challenges to storage space and energy consumption.

Smooth data sharing: The rise of smart hospitals is driving hospital expansion, as well as the widespread adoption of Internet healthcare and diagnosis across hospitals. However, in China, there is a severe shortage of pathologists, and some hospitals are severely understaffed. With traditional image viewing operations, pathology file interoperability and cross-region data sharing are still difficult.

Needless to say, an innovative storage foundation is necessary to further unlock the power of pathology data.

The Quiet Transformation Protecting Our Health

The level of digitalization in the pathology department determines the extent of a hospital's development. Sun Yat-sen University Cancer Center urgently needed a powerful data storage foundation and confidently opted for Huawei OceanStor Pacific scale-out storage.

With Huawei's solution, it resolved a series of challenges and reaped multiple benefits.

Leading performance with over 1,000 images retrieved in one second: Huawei's solution uses advanced all-flash media to enhance the performance in handling hybrid workloads of the image retrieval model. In this way, it supports informatization of Sun Yat-sen University Cancer Center with high concurrent read/write performance, and boosts system performance by nearly 20-fold, enabling retrieval of over 1,000 slide images within a second with no frame freezing.

Efficient storage of large datasets: Huawei OceanStor Pacific scale-out storage adopts an unstructured data compression algorithm that is uniquely designed for pathology files, in addition to a compact 2 U chassis housing 25 innovative 30 TB high-density SSDs. It consumes as little as 0.5 watts per TB and requires only one third of the storage space compared to what was previously needed, resulting in a 20% reduction in TCO.

Agile data transfer facilitated by visualized management: Huawei-developed DME (Omni-Dataverse) is one of a kind. It provides a unified global data view that enables visualized management and on-demand flow of data across the Cancer Center's Yuexiu, Huangpu, and Tianhe data centers, helping build fully automated pathology departments in these branch hospitals.

In fact, the healthcare industry urgently needs converged storage of imaging service data. A unified data lake solution has enjoyed success supporting digital platforms running big data and AI, such as digital pathology, picture archiving and communication system (PACS), as well as bioinformatics platforms. Huawei, building on its expertise in data storage and numerous success cases in healthcare industries, is set to provide cutting-edge storage, encompassing elastic scaling, protocol interworking, and intelligent tiering.

The OceanStor Pacific scale-out storage device runs silently while processing data efficiently, helping patients enjoy even greater care. The power indicator glows quietly, illuminating the path to future healthcare services.

Thanks to the Cancer Center's determination, transformation, and commitment to patient care, the story has a happy ending despite the twists and turns as every bit of data is used for protecting patients' health.

Since its establishment in 1964, Sun Yat-sen University Cancer Center has steadily established itself as a major player in cancer diagnosis and treatment over the past 60 years. As it climbs to even greater heights, it is joined by Huawei.

"Healthcare is the art of preserving and sustaining human life."

If Bian Que were to witness modern healthcare, he'd be astounded by the power of digital pathology, the rapid advancements in health technology, and its vast potential to improve human lives.