Business challenge

To identify cancer trends, advise policy-makers and guide further research, researchers at The Chinese University of Hong Kong needed to find a way to analyze large, complex cancer data-sets.

Transformation

The university uploaded data on 27 different cancers from 102 cancer registries into IBM Watson Analytics. The cloud-based solution helps to explore and visualize the data quickly and intuitively.

Results

Cuts

time spent on data analysis from a semester to a few weeks

Identifies

cancer trends, highlighting areas to focus limited resources

Harnesses

larger data-sets than is possible with traditional desktop-based tools

Business challenge story

Targeting research to achieve the greatest impact

Cancer is the number one killer in Hong Kong, and one of the most significant issues in global healthcare. In addition to the human cost, cancer research, treatment and nursing place a significant financial burden on governments and healthcare companies alike.

CUHK wanted to harness big data to identify cancer trends and advise policy-makers on how to use their limited resources to make the biggest impact in the fight against cancer.

Professor Helen Meng, Professor and Director of the CUHK Big Data Decision Analytics Research Centre, explains: “Data analytics can empower us to study cancer incidence trends, which can potentially inform us about which types of cancer to focus on with limited healthcare resources.”

Professor Kelvin Tsoi, Research Associate Professor of the CUHK Big Data Decision Analytics Research Centre, adds: “With the interactive platform for data visualization, we can more easily identify trends and patterns, which are very important to help us extend our work to other research topics.”

For example, in recent years, the incidence of colorectal cancer (CRC) in the US has successfully been reduced by the introduction of a large-scale screening program. Similar programs could be potentially be valuable in other countries and for preventing other types of cancer – but to decide what courses of action would have the greatest positive impact, it is first vital to understand which cancers are most prevalent among which populations, age-groups and demographics, and how those figures are changing over time.

Professor Meng comments: “CUHK wanted to study data collected by the International Agency for Research on Cancer (IARC), which collects incidence data for 27 types of cancer from 102 cancer registries worldwide. It is a large and complex data-set, and it took a great deal of time just to complete the initial data cleansing. We expected that with general tools, analyzing the data was going to be a very time-consuming task.”

The streamlined support of Watson Analytics is a significant accelerator for our analysis of over 20 years of cancer data.

—Professor Helen Meng,Professor and Director of the CUHK Big Data Decision Analytics Research Centre,The Chinese University of Hong Kong

Transformation story

Harnessing the power of Watson Analytics

IBM Hong Kong, with whom CUHK has a supportive long-term working relationship, approached the university with a possible solution to the data analytics challenge.

“We were trying to decide how best to analyze the large data-sets when IBM came to us and suggested IBM Watson Analytics,” notes Professor Meng. “The timing was perfect.”

A cloud-based data discovery service, IBM Watson Analytics automates predictive analytics, guides data exploration, and provides a wide range of effortless visualization tools.

“IBM Watson Analytics enables us to realize our data exploration aims, and enables seamless exploration and analysis of far larger volumes of data than traditional tools,” says Professor Helen Meng. “The IBM Analytics solution is very intuitive and interactive – it’s almost more of a research companion than a tool. Once we upload the data-sets, it provides strong analytical support that streamlines our investigation and help us uncover our research trajectory as we interact with the data.”

She continues: “The platform also offers a host of dashboards and visualization tools, from word clouds to choropleth maps to radar graphs, which enhance human interpretability of complex data. With big data, you generally want an aggregate view, but you also want to look at various deep specific questions—Watson Analytics offers several different views of the data on one screen, giving us a 360-degree view.

“It is a great credit to the IBM Hong Kong Team who introduced us to this solution at the exact time we needed it. The partnership has greatly accelerated our work in analyzing 20 years of data, by leveraging the capabilities of Watson Analytics.”

Results story

Finding smarter ways to fight cancer

With the ability to explore and analyze big data through IBM Watson Analytics, CUHK has already gained a better understanding of cancer incidence and trends across eight representative global regions: Hong Kong, China and Japan, Croatia, Sweden and the UK, and Costa Rica and the US. By comparing incidence data from high-income economies with lower- and middle-income economies, the research reveals a clearer picture of cancer trends, which will help to decide the upcoming research focus.

Professor Meng explains: “By analyzing cancer trends, we can identify areas where scientists and clinicians can best focus further research efforts. These trends are vital when deciding how to distribute limited resources—for example, we can see that some cancers are strongly linked to certain age groups or socioeconomic factors.”

Professor Tsoi adds: “In particular, the solution has highlighted important trends, such as a gradual decrease in stomach cancer over the years; a recent decrease in the incidence of CRC in the US but not elsewhere; and a higher incidence of breast cancer in Europe and America, compared to Asia.

“By shedding light on the disease, we can make recommendations for future policy-making. Some cancer types are preventable if detected early, so our research can encourage a move towards cancer screening programs where appropriate. Ultimately, as well as saving lives, this could potentially help to reduce costs for governments and healthcare organizations, because preventive measures are generally less expensive than treatment.”

CUHK was also able to accelerate the analysis phase of the project significantly. For example, once researchers had built an analytical dashboard for the first cancer in the study, they were able to re-use the same dashboard for the other 26 data-sets with just a few mouse-clicks. By contrast, if they had been using other tools, such as spreadsheets, they would have needed to do the same work many times over.

“It’s difficult to estimate the time savings because this type of data analytics would have been near-impossible using traditional tools,” comments Professor Meng. “What I can say is that we were able to complete the analysis in a few weeks, and we’re sure it would have taken at least one whole semester the old way.

“Watson Analytics makes data exploration much faster, and also makes it much easier to collaborate. Because it’s a cloud platform, we can access the data from any Internet-connected computer and multiple people can work on it concurrently—which also helps to speed up the process.

“Furthermore, with IBM’s support, we will train our students in data analytics, a valuable skill when entering the job market. Like most modern economies, we’re seeing a huge opening in the market for data scientists and data engineers in Hong Kong. By giving our students the opportunity to use Watson Analytics in their studies, we’re equipping them with skills that our economy needs. Whether they continue as healthcare researchers or move into a completely different field, we’re sure that the analytics skills they learn here will give them an advantage in the future.”

CUHK is also interested in working closely with IBM to add new algorithms and modeling capabilities to Watson Analytics in the future.

“To put our research in context, understanding cancer incidence is just one part of a larger initiative to study and plan for the needs of an aging population,” comments Professor Meng. “In the coming years, more than one third of Hong Kong citizens will be over the age of 65. The study of healthy aging is a top priority, and it’s a very complex problem, involving many different types of data. We are using big data tools like Watson Analytics to solve many large-scale challenges for the future.”

business logo

About The Chinese University of Hong Kong

Founded in 1963, The Chinese University of Hong Kong (CUHK) is a public research university in Hong Kong. The territory’s second oldest university, the education experience is distinguished by a college system, bilingualism and multiculturalism, across nine constituent colleges and eight academic faculties. Four Nobel laureates are associated with the university, which has a student population of around 16,000 undergraduates and 14,000 postgraduates.
 

Solution Components

  • Educ: Innovation in Research
  • Watson Analytics

Take the next step

IBM Analytics offers one of the world's deepest and broadest analytics platform, domain and industry solutions that deliver new value to businesses, governments and individuals. For more information about how IBM Analytics helps to transform industries and professions with data, visit ibm.com/analytics. Follow us on Twitter at @IBMAnalytics, on our blog at ibmbigdatahub.com and join the conversation #IBMAnalytics.

View more client stories or learn more about IBM Analytics