Big data transforms the way entire industries operate and leverage their information. One of the most promising fields for big data is healthcare analytics. Data management tools can save lives, reduce costs, improve resource allocation, and make predictability the new normal.
According to a WHO report, by 2050, there will be 1.5 billion people older than 65 (in contrast with 524 million in 2010). The average age of the earth’s population will be older, which will increase healthcare market growth and the number of people with chronic diseases.
“Primary care providers will play an important role in providing preventive services and caring for this growing elderly population. However, the expansion of medical knowledge and treatment options for these diseases and others has contributed to a proliferation of medical and surgical specialties and subspecialties. More than one-third of patients are referred to a specialist annually, and these specialists play essential roles in the diagnosis, treatment, and monitoring of patients afflicted with diseases and adverse medical events.”
— Health Affairs Magazine
Given our present circumstances, alerting systems for intensive care will be indispensable in the near future.
Let’s go through the current trends in patient monitoring and real-time analytics:
- the aging population and increase in chronic cases
- shift to patient-centric and value-based healthcare
- expanding volumes of big data, opening new horizons for research
- advancements in the Internet of Things solutions and apps
- rapid adoption of wearable technology
“The development and usage of wellness monitoring devices and related software that can generate alerts and share the health-related data of a patient with the respective health care providers has gained momentum, especially in establishing a real-time biomedical and health monitoring system. These devices are generating a huge amount of data that can be analyzed to provide real-time clinical or medical care. The use of big data from healthcare shows promise for improving health outcomes and controlling costs.”
— Journal of Big Data
Alert Management Software In A Nutshell
This article focuses on one area of healthcare analytics—real-time alerting. In many clinics across the globe, clinical decision support software has been introduced. The software analyzes patient data in real-time to give medical professionals data-driven advice on treatment. The patient data is collected from the array of medical devices and wearables that accumulate data and send it to the cloud.
As healthcare businesses explore their options for leveraging data, the market for specialized software solutions is growing at a fast pace. Nowadays, a healthcare provider can choose among self-hosted alert services and open-source solutions. There are a huge variety of libraries; for example, you can pick one android alerting library among many options.
Now let’s see what the benefits of real-time alert systems are:
- the patient gets access to doctors when necessary
- instant patient-doctor communication
- remote and real-time health monitoring
- smart systems of alerts and notifications
An additional benefit is real-time patient monitoring, which has a far greater and more profound impact on how doctors work. Historically, medical professionals have experienced alert fatigue due to an abundance of false positives.
“Because of a rapidly aging population, the Japanese in particular are at the forefront of combining advanced robotics with caregiving and treatment. Robots can be used for everything from monitoring elderly patients who live alone, to helping doctors provide care from a distance to rural patients, to even robotic pets that help calm and soothe dementia and Alzheimer’s patients. The potential to improve patient outcomes, understand disease—even cure cancer—all seem just around the corner with these advances in the quantity and quality of the data we collect along with the computing power to analyze and understand it.”
— Bernard Marr, Enterprise Tech
Mobile Alerting Systems: The Architecture
Simply put, the architectural design of a mobile alerting system is this:
The patient database contains condition-related data such as profiles, parameters, information on the condition, and other information gathered by medical staff. Doctors can choose the profiles and conditions for receiving alerts.
“If you go to the American Telemedicine Association annual conference, you’re like a kid in a candy store—there’s so much wonderful technology and equipment. But there’s a real need for big data and telehealth to merge into one. There are many opportunities for improvement.”
— Heather Zumpano, IMST Telehealth Consulting
We should differentiate doctor-based, patient-based, and nurse-based alerting functions.
By doctor-based we mean:
- critical readings and parameter values that contain abnormalities in them
- non-compliance in measuring parameters
- adverse effects of medicine
Patient-based functions include:
- the correct time, type, and amount of medicine to take
- new prescription
- critical values
- appointments
And by nurse-based functions we mean:
- the correct type, time, and amount of medicine to give
- sending patients to a doctor
- the availability of technical devices for examination
- new educational material on patients, etc.
Open-Source Software Solutions
In cases of smaller teams and streamlined development pipelines, open-source solutions are the best choice. However, before turning to open-source software solutions, you should be aware of existing pitfalls:
- Transparency and Visibility — reliability on mishmash third-party communication tools
- Documentation — open-source solutions do not provide post-incident data and reviews
- Context — too little or too much info
- Scalability — substantial human error involved
- Alert Fatigue — open-source solutions tend to have a very plain alert rules engine, or might entirely lack one
“Custom-tailored and purpose-built solutions definitely win the battle with open-source software. Since we are talking about the transformation of the healthcare industry per se, it takes every business involved to rethink operations and tackle specific issues. There is no room for a one-size-fits-all approach—it has to be a surgical strike.”
— Vlad Medvedovsky at Proxet, software development solutions company
Monitoring Dashboard
Let’s take Azure alerts as an example of a monitoring dashboard. One of the most popular cloud platforms holds 31% of the global market. The typical monitoring dashboard looks like this:
The dashboard provides a summarized view of different components as well as historical data on each metric. The dashboard allows real-time system tracking. The charts and graphs are used to give the big picture.
“How do you convert smart monitoring and telemonitoring into outcomes? You can dazzle with the amount of data but the challenges arise when you can’t show real impact on the outcomes.”
— Avner Halperin, CEO of EarlySense
Alerting Visualization Tools
Google Charts provides a wide variety of visualizations to deploy. The line chart is the most popular visualization, and gives a clear idea of metrics over time. With line charts, it is easy to trace anomalous behavior and deviations from norms. Line charts also display relationships and can be combined with heatmaps, gauges, flame graphs, etc.
Here’s an example line graph:
If this chart helps you see how a real-time alerts system can help a healthcare facility, you are one step closer to big data and its power. In case you have any questions, the Proxet team is always here to provide you with answers.
