
Healthcare organizations that adopt AI will become more efficient; their professional judgments will become more accurate and their health predictions better informed.
In the near future, applications of artificial intelligence (AI) will play an increasingly important role in healthcare services. AI’s ability to assist in managing electronic health records, as well as in diagnosing and treating patients, will prove too valuable for healthcare providers and administrators to ignore. By adopting AI, their operations will become more efficient, their professional judgments more accurate, and their health predictions better informed.
Making Electronic Health Records Management More Flexible
Electronic health record (EHR) systems are expensive to maintain and cumbersome to use. AI is already being applied to make EHR systems more efficient. AI can extract and index information from provider notes and help personalize treatment plans. In the near future, AI could make it easier for providers to continually customize their EHR systems to better meet the changing needs of their practices to save time and improve patient outcomes.
Improving Diagnoses and Treatment
Computers can analyze massive amounts of data. They also excel at recognizing patterns. Put these two abilities together, and you get AI’s extraordinary power to successfully perform tasks that previously required human intelligence—in some cases, even surpassing humans in accuracy.
In a process called “deep learning,” scientists train AI systems using large amounts of labeled data. The AI is then able to identify patterns by itself when given data to which it hasn’t yet been exposed. In healthcare, deep learning will train AI systems to help clinicians provide more accurate diagnoses, identify patients at risk of various diseases and conditions, and create individualized treatment plans.
A recent study found that AI outperformed radiologists at finding cancer on CT scans used to screen smokers for lung cancer. When prior scans were not available, the AI did better than all six radiologists in the study, coming up with both fewer false positives and fewer false negatives. When there were prior images, the AI and the radiologists were equally accurate. Although this was a preliminary study, it shows how near-future applications of AI can provide clinicians with more accurate diagnostic and predictive information than is available to them now. The results of this study also may enable radiologists to make more frequent life-saving identifications of early-stage cancer and other diseases.
In addition to reading images, AI will be able to extract useful information from patients’ medical records. By finding patterns in the medical histories, AI could provide warnings when patients are at risk of developing conditions such as sepsis, diabetes or heart disease. An area of intense interest now is using AI to identify which patients are most at risk of being re-admitted to a hospital after being discharged.
AI may also reduce the amount of trial and error involved in prescribing medications and other treatments. It can help identify the treatments most likely to succeed based on each patient’s unique combination of genes, medical history and environmental influences.
The Time Is Right
In the past, there was more resistance to using AI in healthcare. Now, providers and the public are increasingly ready to accept the use of this cutting-edge technology to make healthcare administration more efficient and providers’ decisions more accurate.
Healthcare professionals are seeing what AI can bring to their practices. At the same time, the public has gotten used to the idea of self-monitoring their health using smartphones and smart watches. More work still needs to be done to implement industrywide standards and to safeguard patient privacy, but the benefits of AI in healthcare now appear overwhelming. It is inevitable that healthcare providers and organizations will soon come to increasingly rely on AI applications.
This could be just the beginning. How well AI can “think” depends in large part on how much computing power is available — and that power is increasing exponentially.
Healthcare in the more distant future may include AI applications that we can’t even imagine now. In the meantime, healthcare providers and administrators may soon enjoy greater efficiency and cost containments, while patients could benefit from more accurate diagnoses and effective treatments.
Sources
https://www.nytimes.com/2019/05/20/health/cancer-artificial-intelligence-ct-scans.html
https://www.nature.com/articles/s41591-019-0447-x
https://hbr.org/2018/12/using-ai-to-improve-electronic-health-records
https://becominghuman.ai/the-role-of-ai-in-healthcare-technology-6c33a6eee18c
https://www.internationalsos.com/client-magazines/in-this-issue-3/how-ai-is-transforming-the-future-of-healthcare
https://www.appian.com/blog/the-growing-role-of-ai-in-healthcare/
About the Author: Steve Bradshaw began working in the medical gas and environmental industry in 1991, starting Evergreen Medical Services, Inc. in the Carolinas in 1997.