Two cancer patients enter a hospital with the same symptoms, receive the same diagnosis, and are treated using the same protocols. One lives, while the other dies. Three Sydney-based academics believe the secrets hidden in the complexity of our genes may help clinicians explain why.
Over a decade, a Western Sydney University research team led by Dr Quang Vinh Nguyen in collaboration with researchers from The Children’s Hospital at Westmead and the University of Technology Sydney have developed visual tools that give oncologists unparalleled insights into understanding complex patient data that enables effective decision-making for the treatment and management of cancer.
A major hurdle to effective cancer treatment is the myriad genetic and molecular changes, almost as individual as a fingerprint, which turn healthy cells into cancer cells. Similar patient presentation often belies very different genetic activity, curtailing the ability of clinicians and researchers to learn from each new case and effectively compare patients.
To overcome this, the team has produced interactive and visual analysis tools, underpinned by machine learning algorithms that can compare a new patient’s genomic information to the wealth of historical data contained in The Children’s Hospital at Westmead’s tumour bank. Such comparisons can provide doctors with useful data on genetically similar patients, and illustrate the impact of differences in cancer’s many genetic variations.
Clinicians at Westmead have praised the technology with the research team receiving feedback such as: “The tools were useful in capturing similarities and individual difference among patient genetics” and “the visualisation tools were useful in enabling personalised medicine to compare and identify a patient among a cohort”.
Eventually, this system will provide personalised treatment recommendations based on the analysis and comparison of thousands of genes.
Need to know
- Analysis of the genetic data of cancers can lead to better prognoses
- WSU research has made a tool to visualise the data to aid diagnosis
- The work has won praise and prizes
Big data, big insights
Alongside clinical applications, the team’s data modelling could bring much-needed clarity to research. “Someone did a study into leukemia patients and found that the activity of 15 genes could predict the outcome of a group of patients,” says Daniel Catchpoole from The Children’s Hospital at Westmead. “We took the same 15 genes and found no outcome difference between our two groups. So, we expanded the list of genes to a couple of hundred, and found that the two groups were distinguished by genes associated with DNA replication. And when we looked at the treatments used by the two hospitals in the original study, the main difference was a drug that affected DNA replication.”
An essential part of the team’s work involves visualising these insights in a way that’s accessible for potential end-users, says Nguyen: “Everyone interprets differently. We discuss with clinicians and researchers what their expectations are, what they want to see, and what platforms they’ll use in their work.”
One such visualisation is the “similarity space,” a 3D cloud of data points where each point represents a patient, and the position of the points in relation to each other reflects the levels of genomic similarity between those patients’ cancers. Groups of data points reflect shared genes, which can be further investigated for clinical relevance. Thanks to a grant from the tech giant Oracle, Nguyen (pictured) has added a second doctoral student to his roster, to further investigate visualising this type of data in immersive environments, as well as
virtual and augmented realities.
The team’s work is already gathering public and professional support, with Nguyen recalling the group’s win of the Cancer Institute NSW 2015 “Big Data, Big Impact” award as one of his proudest moments of the last decade.
Meet the Academic | Dr Quang Vinh Nguyen
Dr. Quang Vinh Nguyen is a Senior Lecturer in Visual Analytics at the School of Computing, Engineering and Mathematics and MARCS Institute, the University of Western Sydney. His research areas are in Visual Analytics and Information Visualization, including Medical Data Analysis, Graph and Network Analysis, Graph Drawing, Applications with Visualization and Visual Analytics, Human Computer Interaction, and related research areas. His focus is to find effective visualisations to support the analysis of large and complex datasets, particularly genomic and biomedical data, health data, network data and other application based data. His research expertise has been built up since his PhD study, his various experiences at the University of Texas at Dallas (UTD), the University of Technology, Sydney (UTS) and the University of Western Sydney (UWS). For his academic career, he has authored and co-authored over 60 refereed publications, including edited book, book chapters, journals and conference papers relating to this research field
Higher Degree Research at Western
© oxygen/Moment/Getty, © Bill Oxford, © Markus Spiske.
Future-Makers is published for Western Sydney University by Nature Research Custom Media, part of Springer Nature.