How AI is Making Healthcare More Efficient and Affordable
Artificial Intelligence (AI) is becoming more and more prevalent in our daily lives, whether we are aware of it or not. Virtual assistants like Alexa, facial recognition software, and the Google search engine are just some examples of how we use AI every day. All of these examples make our lives more convenient, but there is one particular part of our lives that can be improved drastically through the use of AI: healthcare.
The healthcare industry is continually evolving as AI and Machine Learning (ML) are playing more prominent roles. In fact, projections indicate that healthcare spending on AI will grow 48% annually until 2023.
How is AI Used in Healthcare?
ML, a subset of AI, uses algorithms and data to offer automated insights to doctors and healthcare providers, the end result of which is providing data-driven clinical decision support (CDS).
AI can also foster preventative medicine and new drug discoveries. For example, IBM’s Watson performs big data analysis so that it can determine treatment options for people with tumors and cancer. In fact, researchers at the University of North Carolina Comprehensive Cancer Center use Watson to identify treatment for over one thousand patients.
Another example is Google Cloud’s Healthcare app, which enables health organizations to collect, store, and access data all in one secure place. It also offers CDS functions and other AI solutions designed to help doctors make informed decisions regarding each patient. One of the ways it does this is by using ML to collect people’s electronic health records, which it then uses to create intelligent insights for healthcare providers.
For example, Google teamed up with several universities to create an AI system that predicts the outcome of hospital visits. The predictions can help patients avoid readmission (by foreseeing the outcome of the first visit) and to shorten the amount of time patients need to be in the hospital (a win-win for hospitals and patients).
According to Time, 34% of healthcare expenditures are administrative. With AI, a lot of these tasks can be automated, including pre-authorizing insurance, maintaining records, and following-up on unpaid bills. As AI shoulders a good portion of the administrative load, a lot of time and money can be saved.
The AI that people wear also plays a part in the health industry. FitBits and smartwatches, for example, can analyze data to alert users and healthcare professionals to potential health risks. Actually, the initial information provided by these devices goes directly to the consumer, which means they can assess their own health first before turning to healthcare professionals, thereby easing their workload and minimizing unnecessary doctor and hospital visits.
The Risk of Using AI in Healthcare
The benefits of using AI in the field of healthcare are clear, from CDS to shortening hospital stays to administrative assistance. However, AI in healthcare is not as cut and dry as it seems. The risks with implementing AI in a field of life and death is that the data analyzed is based on the data that we, humans, provide. And when humans provide data, there is a risk of an unconscious bias. If AI is analyzing biased data sets, there is a risk of producing biased solutions.
John Giannandrea, senior vice president of Machine Learning and AI Strategy at Apple and former AI chief at Google, has commented about unconscious bias, “The real safety question, if you want to call it that, is that if we give these systems biased data, they will be biased.”
In the field of healthcare specifically, in which the people using AI are doctors and nurses, not tech workers, the risk of using AI and ML with biases is greater. People who don’t have a deep technical understanding of the software are more at risk to use it incorrectly.
The Future of AI in Healthcare
Despite the risks, the use of AI in healthcare is growing quickly, both because of the prevalence of wearable technology and the general trend toward the digitalization of everything.
Overall, the use of AI in hospital and healthcare settings allow doctors to better treat their patients. This includes storing patient information on private clouds, like Google, which allows easier access to health information, and helping pinpoint the right treatment for cancer patients.
As it stands now, health tech companies, startups, and healthcare professionals are actively seeking ways to implement AI into the healthcare industry. The speed with which new technology becomes part and parcel of current medical treatment will depend on how well programmers minimize the risks and how well the healthcare system can adapt to new ways of treating patients.