What is AI?
Artificial intelligence (AI) refers to computer systems that simulate or exhibit a specific aspect of human intelligence or intelligent behavior, such as learning, reasoning, and problem-solving.
It is not a single technology, but rather a range of computational models and algorithms, the most important of which being Machine Learning (ML), which refers to a system that identifies patterns of data from input and makes predictions from new, never-before-seen data.
ML algorithms can automatically learn and improve from experience without being explicitly programmed, and such “learnability” represents a key feature of AI.
The benefits of AI in healthcare
Healthcare is a field in which AI is rapidly evolving, leading to a digital transformation that is:
- improving the access, quality, safety, and efficiency of various health services
- supporting evidence-based decision making
- fostering communication and coordination
- allowing for the optimization of health systems' performance
- unlocking big data to gain insights into patients
- delivering value at reduced costs, while still improving outcomes and patient experience
Current AI uses in healthcare
Recently, many ML algorithms have been approved for safe use in healthcare by the US FDA. Stanford developed an AI algorithm that can diagnose up to 14 types of medical conditions simultaneously from imaging, Mayo Clinic started using AI to find molecular biomarkers in MRI imaging instead of testing samples collected during surgery, while MIT developed an AI prediction model that can anticipate the development of breast cancer up to 5 years in advance.
Many big tech companies such as Google, Microsoft, Apple, Amazon, and IBM are providing organizations with AI cloud platforms and services as well as ML algorithms. There have also been various cross-sector collaborations, with Apple partnering with over 100 hospitals and clinics for its health records project, allowing consumers to exchange their health data with healthcare providers, or IBM partnering with various hospitals, enabling them to use Watson to make cancer diagnoses and treatment recommendations, a system that can digest information and make recommendations much more quickly and more intelligently than perhaps any machine before it, processing up to 60 million pages of text per second.
AI-assisted surgical robots are an example of this, as they can be operated both locally and remotely and are able to analyze data from pre-operative medical records to physically guide a surgeon’s instrument in real-time.
Robots can also be used in the rehabilitation of patients with stroke, to deliver equipment and medical supplies as well as to assist in the care of elderly patients.
Many hospitals in the United States have also started using ML algorithms for predictive analytics (eg, predicting adverse events, mortality rates, the number of patients in the emergency department), obtaining data that allows them to take proactive measures for the foreseen events days in advance.
EHRs (Electronic Health Records) are the backbone of this process and currently, many EHR vendors, including Epic, Cerner, Allscripts, and Athena have started to add AI into their systems with the purpose of supporting workflows, and clinical decision-making as well as patient engagement.
Another potential use that can enable clinicians to make quicker and smarter decisions about their patients uses the smartphone as a catalyst. Various companies are developing sensors that attach to phones to collect all sorts of biological data.
Ultimately, these applications can fall into 4 different categories:
The future of AI in healthcare
As more and more data are captured, and as computers become better and faster at processing them autonomously, the possibilities keep expanding. The application of AI in healthcare is disruptive, so a good question to ask is whether it will change not only how medicine is practiced, but who is practicing it. As some Silicon Valley investors are speculating, one day AI could take the place of doctors, serving as a diagnostician and even a surgeon, maybe doing the same work with better results for less money.
But physicians, after all, do more than data processing. They educate and walk the patients through the difficult times of disease and perhaps death, attend at their bedsides, and counsel them. They grasp nuance and learn to master uncertainty. No AI could replace that humanity in the eyes of a sick patient.
Chen M, Decary M. Artificial intelligence in healthcare: An essential guide for health leaders. Healthc Manage Forum. 2020 Jan;33(1):10-18. doi: 10.1177/0840470419873123. Epub 2019 Sep 24. PMID: 31550922.
About the Author
Dr. Patrizia Scali is an Italian ECFMG certified medical doctor.
Dr. Scali graduated medical school from Università degli Studi di Milano-Bicocca in Milan, Italy and has completed all 3 USMLE Step exams. She has also completed an Accelerated Certificate Program in Business Administration at University of California, Irvine (UCI).
Dr. Scali has conducted pediatric hematology research in Italy, as well as hepatology lab research at Yale School of Medicine in the US. She has worked as a primary care physician in Italy, her home country, where she also had extensive experience as a telemedicine doctor.