Privacy Protection and Cybersecurity in Healthcare. Is AI the Answer?
As more healthcare systems become digitized, the risk of cyberattacks increases, causing concern among those involved with these systems. Patient health records are high-ranking on the black market, and identity theft as well as falsifying medical records is on the rise. These attacks can result in severe consequences for patients and hospitals. The COVID-19 pandemic has also caused pressures to increase, along with a growth in the number of ransomware attacks on hospital systems. You can read more about how COVID-19 has affected cybersecurity in the WIRED article, ‘AI in Healthcare: Protecting the Systems that Protect Us’.
However, positive technology can counteract negative technology, and the development of artificial intelligence (AI) and machine learning (ML) has proven to decrease these risks and attacks by anticipating and preventing them. With enough financial support and research, AI could prevent the loss of millions of dollars, the tarnishing of hospital reputations and the invasion of patients’ privacy.
Machine Learning and Virus Detection
Emerj has outlined the major possibilities involved with machine learning and preventing cyberattacks. Their article ‘Machine Learning in Healthcare Cybersecurity - Current Applications’ walks readers through technologies such as anomaly detection, predictive analytics and visualizing cybersecurity threats in a user interface. All of these are viable solutions - yet, the article doesn’t ignore the complications that could be involved with installing these kinds of AI-ML technologies. AI can be developed and utilized to the advantage of the cyberattackers, as well, so developers must take this into consideration.
Darktrace, a leading company that develops AI for cybersecurity, has developed a system that models the human immune system. This is done through the understanding of “self”. The human immune system “learns what’s normal in our body and, therefore, what’s not normal, and that’s how Darktrace works - so we can automatically detect threats within any enterprise network ” states CEO of Darktrace, Nicole Eagan. This is based on the idea that viruses are already inside and need to be found. Eagan highlights the need to focus on the inside as well as on the outside in order to detect threats, creating a more holistic approach.
Successful Implementation in Healthcare
Darktrace has successfully detected Maze ransomware in a healthcare system, enabling the threat to be stopped before encryption began. Their case study document outlines the process of this success, as well as their wider role within healthcare.
Artificial intelligence can overcome many obstacles within the healthcare industry, including making digitized systems and protocols more secure. While the industry has seen effective developments in this sector, we can expect rapid evolution in this area to protect all parties that come together to make healthcare systems flourish. Successes such as Darktrace are planting the seeds of trust that are essential for systems to move towards and implement AI-ML within their organizations.
Written by:
Maxine Wesley
