Evolution of AI Healthcare

From Treatment to Prevention: The Evolution of AI in Healthcare

Artificial intelligence has a vast range of use cases in healthcare, from research to clinical care to financial applications and more. AI can be used to predict the onset of diseases, for example, enabling care providers to more easily identify and improve care for high-risk patients. It can also support clinicians in analyzing medical images and even be used to help discover and develop medications.

How did artificial intelligence get its start in healthcare, and how will it continue impacting the healthcare industry?

Here’s a look at the evolution of AI in healthcare.

Evolution of AI in Healthcare

The early days of artificial intelligence 

In 1950, British logician and mathematician Alan Turing began wondering if machines could think. Six years later, scientists and mathematicians discussed the simulation of intelligence by machines during a Dartmouth Summer Research Project. AI began to be used for healthcare purposes in the early 1970s, where it was applied to biomedical problems. During this decade, there was also an increase in AI research. In 1980, the American Association for Artificial Intelligence was established and included a subgroup on medical applications. In the following decades, AI would be introduced in clinical settings. [i]

In 2012, Geoffrey Hinton and his colleagues released AlexNet, which showed that deep neural networks perform better than traditional AI pipelines. Two branches of AI in healthcare eventually evolved: physical, which includes using devices to improve care, and virtual, which includes machine learning and deep learning. [ii]

Several companies have created AI tools that are exploring the benefits of deep learning, including IBM Watson, which is being used to investigate for diabetes management, drug discovery, and advanced cancer care, and Google’s Deep Mind, which is being considered for applications like mobile medical assistant, using medical imaging to determine diagnostics, and predicting patient deterioration. [iii]

Increased computer processing speed, a large talent pool of AI professionals, and growing data collection data libraries have served as catalysts for the speedy development of AI technology and tools, including those used in healthcare. [iv]

In clinical studies, AI can quickly analyze images and datasets and compare them to other studies to identify patterns and other connections that may not be easily noticed by the human eye. With AI, medical imaging professionals can quickly track important information. [vii]

Medicine development 

As artificial intelligence evolved, it became useful in identifying new drug applications and developing new medicines. Using databases of molecular structures, supercomputers have been used to predict the effectiveness of potential medicines against different diseases. AI algorithms can also trace a drug’s toxic potential and mechanisms of action to identify new drug applications. Additionally, artificial intelligence can be used to help quickly and successfully discover and develop medicine. [viii]

Disease prevention 

Changing the healthcare model from a reactive approach to a proactive approach may help reduce annual healthcare costs in the US by $150B in 2026 as well as lead to fewer hospitalizations, doctor visits, and treatments. [ix]

Artificial intelligence can support a proactive approach by helping care providers predict diseases, identify high-risk patients, diagnose patients faster and with increased accuracy, and improve the quality of care they provide. 

Lumiata, for example, uses AI and machine learning to predict disease onset, hospitalizations and readmissions, diagnoses, medical events, and more. Using a vast dataset and by looking at several risk factors, our AI solutions can identify patients at risk of developing chronic kidney disease, asthma, congestive heart failure, dementia, obesity, primary hypertension, and many other acute and chronic health issues.

Our predictions empower providers to address each patient’s risk factors sooner than later to help keep them healthy. This proactive approach to care helps patients avoid disease onset and the high costs associated with treating diseases. This cost savings extends to care providers and payers, too. 

Financial applications for AI in healthcare 

While artificial intelligence has important uses on the clinical side of healthcare, it’s also a valuable tool on the financial side. For example, providers can use Lumiata’s AI products to improve payment integrity. Our AI technology looks for outliers in the patterns of care to identify suspicious healthcare claims. 

Payers can then investigate these claims to make sure they’re correct before paying them. Not only does this allow payers to catch more suspicious, higher-cost claims, it also allows payers to send fewer claims to vendors for review, resulting in fewer vendor fees. Lumiata can also help providers identify suspicious claims so their team can review them before submitting them to payers. This helps ensure that claims are paid the first time. 

Artificial intelligence can also be used to support payers’ actuarial and underwriting efforts. Our AI and ML tools can identify high-cost members, cost bloomers, group costs, and more to give payers clearer insight into financial risks. Payers can then use these insights to set premiums that are high enough to cover predicted costs yet low enough to attract new members and prevent existing members from leaving. 

Learn more Over the past 70+ years, artificial intelligence has evolved from a concept to a powerful resource that can help keep patients healthy and reduce healthcare costs. To see what AI can do for your organization, request a demo of Lumiata.