We are living in a data-driven era and so we keep hearing about the terrific innovations made possible by data science week after week. Last week was no different!
So, let's take a detailed look at a few stories from last week that showcase the impact of data science in a healthcare organization, such as Penn Medicine, for the federal government of the United States and the US Army.
Such stories will reinstate the fact that the emergence of big data, machine learning, business intelligence and more has led to good breakthroughs for the welfare of mankind.
Clinicians and medical informatics experts at Penn Medicine, developed a new technology to identify patients who are under the risk of developing a critical condition like sepsis.
This new breakthrough has improved their ability to detect this risk in patients 24 hours faster than was possible before.
The Penn team started out with a model labeled as Systemic Inflammatory Response Syndrome (SIRS). It helped them identify the key symptoms of the sepsis condition such as specific limits of temperature, respiratory rate, heart rate, and white blood count.
All this data gets loaded into the computer for each patient. The computer then uses the new algorithm to check the resemblance of a new patient's characteristics with those who have suffered from sepsis in the past.
When a close resemblance is found, the concerned doctor and nurse acting on the relevant patient are alerted. Based on their action and experience with such patients, the clinicians also give some feedback to the algorithm in order to improve it.
An Interesting Fact
Dhanurjay Patil, the first chief data scientist of the United States Office of Science and Technology Policy, has described Barack Obama as the most data-driven president, which America has ever had.
Patil has initiated the website www.data.gov for maintenance of US government data, that is shared with public for better transparency.
For example, they are using a data-centric approach that considers the differences between people's genes, lifestyles and environments to prevent or treat diseases. This is bound to improve the health of civilians and even extend their lives to a certain extent.
Identifying Possible Violent Offenders
Researchers of the Harvard Medical School have come up with an algorithm that assists in identifying US army soldiers who are more likely to carry out violent crimes, including robbery, murder, and kidnapping.
They have developed this program on the basis of US military records of more than 975,000 soldiers who served between 2004 and 2009.
They used the machine learning technique to identify patterns among the soldiers who were found guilty of committing violent crimes during that period.
The researchers highlighted a few common symptoms within such high-risk individuals. These were prior sufferings from mental health disorders, previous offenses, unfortunate social background (with a lack of opportunities), and so on.
The aim of this “early predictor” algorithm, is to avoid the crimes being committed by US military personnel due to the presence of some high-risk soldiers.
This can be achieved by offering timely counseling to such soldiers who are likely to offend. This model is almost ready and could be employed by the US military very soon.
These stories prove how data science is revolutionizing our lives every day. Do you have any story to share or have any comments to add? Please feel free to leave your comments below and thanks for reading!