At Virtualitics, we feel it’s important to contribute to research and development across healthcare and environmental issues in order to make a positive impact with the help of our software.

Last year Virtualitics joined forces with a group at Columbia Medical School, led by Dr. Simon Cheng and medical student Connor Kinslow, in order to understand the link between cancer mortality as it relates to flu epidemics and how to better communicate those insights more effectively.

This particular topic is crucial given the fact that lung cancer is the leading cause of cancer-related deaths globally. While influenza illness is known to be particularly dangerous for frail and elderly patients, the relationship between influenza illness and outcomes in cancer patients remains largely unknown.

Combining the Virtualitics platform and Machine Learning expertise with the domain knowledge of Columbia’s researchers, we were able to reach profound results, which were presented at the American Society of Clinical Oncology (ASCO) Annual Meeting. The preliminary results were also published in the Journal of Clinical Oncology.

We collaboratively compared monthly mortality rates for all patients at risk, as well as newly diagnosed patients with non-small cell lung cancer (NSCLC) diagnosed between 2009 and 2015 during the high and low flu months using data from the Surveillance, Epidemiology, and End Results (SEER) Program and the Center for Disease Control and Prevention (CDC).

Influenza severity was determined by the percentage of outpatient visits to healthcare providers for influenza-like illness.

 

Figure 1: The temporal distribution of high flu months. With the exception of 2009, which saw elevated flu activity due to the H1N1 outbreak, flu severity tended to occur during the winter months. High-dimensional visualization, coupled with AI routines proved to be fundamental in understanding the complex relationships within the data.

The study was conducted using thousands of patients with NSCLC and located across 13 States.

What we found from the data in hand is that the monthly mortality rate for newly diagnosed patients was significantly higher during high flu months. The relationship between flu severity and mortality was also observed at the individual state level.

In conclusion, we were able to prove that the increased influenza severity was associated with higher mortality rates for NSCLC patients. This is the first study to estimate the effects of regional influenza outbreaks on lung cancer mortality. According to lead author Connor Kinslow: “We hope these results will inform patients and providers alike about the dangers of influenza infection for lung cancer patients and have the potential to influence national policy”.

The Virtualitics team that worked on this study was composed by David Wang (Machine Learning Engineer), Konstantin Zuev (Data Scientist), Sarthak Sahu (Head of Machine Learning), Michael Amori (CEO and Co-founder), Ciro Donalek (CTO and Co-founder).

Network graphs offer a new way to understand relationships between entities in unstructured data.

Detailed paper: 

Influenza and mortality for non-small cell lung cancer. Authors: Connor J Kinslow, Yuankun Wang, Yi Liu, Konstantin M. Zuev, Tony J. C. Wang, Ciro Donalek, Michael Amori, and Simon Cheng. Published on Journal of Clinical Oncology, Volume 37, Issue 15.

Categorised in: