How healthcare organizations can benefit from social media analysis

Artificial intelligence and machine learning are in the spotlight of many tech news websites, and it is not for nothing. These technologies are being actively implemented in an array of spheres, and some AI uses make the mind boggle.    

The below statistics show that AI is already here, and it is here to stay.

Thus, according to a 2016 Narrative Science study, 38% of enterprises are already employing AI, and this number is expected to grow to 62% by 2018. As for the AI market growth, it is anticipated to surpass $40 billion by 2020 and $100 billion by 2025.

AI is also revolutionizing healthcare, and this article is aimed at shedding light on certain aspects of this technology implementation in hospitals and care organizations.  

Healthcare and social service institutions may face bitter criticism due to the growing number of suicide cases, deaths of elderly people who didn’t get sufficient care, etc. In this case, a linguistic analysis of posts in social networks and blogs can render valuable assistance in reducing risks to a significant extent.  

To understand the tremendous importance of early detection of suicide-disposed people, let’s turn to some statistical data.  

The World Health Organization (WHO) reports say that each year nearly one million people die from suicide (one death per 40 seconds). And the death rate is predicted to increase to one death per 20 seconds by 2020.

Earlier, suicide rates have been highest among elderly people, but the situation has changed, and now in a third of all the countries youngsters are at higher risk.

Healthcare institutions and social services work with zeal to prevent suicide among adolescents and elderly people. But sometimes it’s not enough, and an AI-driven approach is needed.  

Our company, EffectiveSoft, has considerable expertise in the sphere of computer linguistics and machine learning, and we are always willing to share experience in healthcare apps development and to deliver other smart solutions.

Our firm has recently come across the challenge of identifying suicide-disposed individuals on the web. Company’s savvy-tech experts have elaborated a bevy of workflows that can be the used to conduct a linguistic analysis of social media comments, chats, posts, etc. Here are some takeaways.  

Data acquisition

First, it’s crucial to engage medical experts in creating specific questionnaires and use them to work with persons at risk. The focus group (people who suffered from the above-mentioned problems) give certain answers that have to be analyzed and systematized by physicians who will also eliminate buzz.   

As a result, customers (for instance, hospitals) get a set of professionally composed documents with the patterns describing one or another mental illness that may become the reason for suicide. Those patterns are needed for machine algorithms and further Internet search.

Machine learning

Computer linguists are building specific infrastructures giving computers the ability to learn new patterns. It’s a complicated and sophisticated process that requires high qualification.

Today, there’s an array of tools able to carry out a standard linguistic analysis. In turn, EffectiveSoft boasts a unique linguistic core that can train with predefined patterns selected by physicians. In this case, machine algorithms get an opportunity to identify specific linguistic constructions that are characteristic of individuals who are at the brink of committing suicide.

Practical use

In practice, people don’t tend to give identical answers. That’s why machine algorithms need much diverse data. Taking into account the fact that EffectiveSoft’s core allows conducting a deep linguistic analysis, the customer will be granted a chance to receive data with stand-alone sample patterns.

In addition, the algorithms can be adjusted to detect successful and unsuccessful cases for further results correction.  

Conclusion

The algorithms are then used to analyze social media posts and detect suicide-disposed persons, which becomes a valuable information source for hospitals and other care organizations in their try to reduce the number of deaths.    

The prototype described above is a bright example of EffectiveSoft’s expertise in social engineering and its ability to respond to tough challenges.    

Acknowledgements

Produced from materials originally produced by Yana Yelina, a Tech Journalist at EffectiveSoft.

References

  1. Prevention of suicide among adolescents, WHO. Available at: http://www.who.int/mental_health/en/

About EffectiveSoftEffectiveSoft Corporation

EffectiveSoft is a custom software development company headquartered in the USA (San Diego, California). We have development facilities located in Eastern Europe (Minsk, Belarus). Our team numbers 250+ experienced specialists: software developers, engineers, and scientists with expertise in different technical domains.

All of the company members have bachelor's or master's degrees in the sciences and at least five years of experience in the software development field. Our team spirit and the company's management expertise and support are successfully combined with creativity, a dedication to work, and a development culture to produce solid, effective software results. Our commitment is to provide services of the highest quality to our clients


Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.

Last updated: May 25, 2017 at 7:33 AM

Other White Papers by this Supplier