image_pdfimage_print

Artificial intelligence (AI) is changing how we do practically everything, and the medical field is no exception. AI in healthcare has numerous uses, and the ways medical professionals apply concepts like machine learning are continually expanding. One of the greatest benefits of AI is its ability to process and analyze mountains of data, detect abnormal patterns, and learn from previous experiences. In healthcare, this can result in smarter decision-making, improved quality of care, and better patient outcomes. 

As medical technology improves, it can become increasingly difficult for professionals to keep pace. For example, it took 13 years—from 1990 to 2003—for multiple teams at 20 universities and research centers in six countries to map the human genome a single time. By 2019, technology and AI had advanced enough to enable scientists to offer same-day whole-genome sequencing to diagnose genetic diseases in seriously ill babies. Human brainpower simply cannot process information fast enough to achieve that kind of breakthrough without AI. 

Here are five ways AI is revolutionizing healthcare. 

1. Generative AI-powered Search Engines Can Improve Clinician Experience and Efficiency

If there’s one industry where it can be challenging to obtain standardized, complete datasets, it’s healthcare. It makes life difficult for clinicians who struggle to search massive datasets stored across multiple systems and formats. Generative AI-powered search tools like Google’s Vertex AI Search can allow healthcare providers to query datasets and get clear, fast answers in return. A simple question removes the need for extensive data analysis, instead pulling answers from clinical notes, scanned documents, insurance providers, and electronic health records and returning them in a human-readable form.

2. AI Can Improve and Enhance Medical Diagnosis

Since healthcare organizations first began using AI, it was always a tool to assist clinicians and technicians. AI would quickly process data, find correlations, and offer insights, but a human ultimately had to look at that information to make a diagnosis. 

The technology made a great leap forward, however, in 2018, when the FDA approved the IDx-DR. This was the first stand-alone AI-based system that could provide autonomous diagnoses of diabetic retinopathy without the supervision of a doctor. With the IDx-DR, a camera captures images of a patient’s eye and uploads them to the cloud, where software analyzes them. The AI compares these images to a massive bank of stored images that were previously analyzed and tagged by human professionals, then returns either a negative or positive diagnosis. The clinical study used to procure FDA approval demonstrated that the IDx-DR could detect moderate to severe diabetic retinopathy with greater than 87% accuracy. 

Another example of how AI can be a powerful tool in medical diagnosis is the deep learning convolutional neural network, or CNN. According to Science Daily, a CNN is “an artificial neural network inspired by the biological processes at work when nerve cells (neurons) in the brain are connected to each other and respond to what the eye sees.” It can quickly process, or “view,” thousands of images and teach itself to recognize patterns. 

In 2018, researchers in Germany, France, and the United States trained a CNN to detect skin cancer by showing it over 100,000 images of malignant melanomas and benign moles and teaching the system to classify them based on observable patterns. After the training period, the team tested the CNN against a group of experienced dermatologists. The CNN caught 95% of melanomas, compared with a human accuracy rate of 86.6%.

The exciting thing about improving medical diagnosis methods and accuracy is that it leads to more personalized and effective treatment overall.

3. AI Can Improve the Patient Experience

Another way AI is beneficial to healthcare is in streamlining the patient experience and making it more personalized. The more efficient organizations can be, the more patients they can see each day. 

For instance, the Olive AI platform automates some of healthcare’s more repetitive and time-consuming tasks, such as insurance eligibility checks and processing unadjudicated claims. This frees up clinicians to focus more on patients or deal with more complex administrative issues. 

Another example is Babylon Health, which provides interactive and personalized patient care through the use of its AI-powered chatbot. The tool analyzes a patient’s symptoms, then recommends one of two options: a virtual check-in with clinicians or an in-person visit at a hospital. 

Johns Hopkins Hospital has teamed up with GE to make patient flows more efficient through AI and predictive analytics. The updated command center has resulted in a 60% improvement in its ability to admit patients and a 21% increase in patient discharges before noon—which results in a quicker, more satisfying patient experience.

4. AI Can Improve Outcomes

As AI helps healthcare organizations diagnose diseases earlier and create more personalized treatment plans, the potential for a positive outcome increases significantly. 

Surgical outcomes can also be improved through AI. Despite tremendous improvement in the quality of imaging technology, surgeons still often have to rely on 2D images that are hours, days, or even weeks old. This requires them to do mental math during procedures and perform adjustments on the fly as they gauge depth, range, and accuracy based on experience and the tools at their disposal.

AI can change that. For instance, a surgical navigation system from a company called Proprio can provide a real-time, enhanced, 3D view of the surgical site anatomy. It relies heavily on artificial intelligence, augmented reality, and real-time visual rendering of what the surgeon is experiencing. This kind of system can shorten surgery times and decrease mistakes due to surgeon fatigue while increasing accuracy, reducing errors, and minimizing the amount of trauma to the surrounding areas.

5. AI Can Enhance and Accelerate Medical Research

The FDA reports that, of the drugs that enter preclinical testing, just 5 out of 5,000 make it to human testing—and just 1 of those 5 will ever gain approval for human use. With AI, however, drug discovery and repurposing are being greatly accelerated and can cut time to market for new medications.

AI algorithms, such as those used by Deep Genomics, can also help identify and develop drugs to treat genetic diseases. The platform predicts genetic alterations in protein binding, which predicts the possibility of genetic diseases. It can then discover new ways to fix those mutations and create customized treatment for people who suffer from genetic diseases. 

It’s long been known that sharing data between research organizations can save lives. The more data AI platforms can work with, the better the insights. Recently, however, the UK’s National Health Service (NHS) received widespread negative press when it announced it would give detailed health records of 55 million patients to a variety of organizations for medical research. New AI advancements, however, have enabled algorithms that can talk to each other and share insights without actually sharing highly confidential patient information. This will enable robust research while still protecting patient privacy. 

Why Is Artificial Intelligence Important in Healthcare?

Although science fiction paints a future of doctorless care, that’s not how medical professionals see AI in healthcare working out. There’s little doubt that the role of AI in healthcare will continue to expand, even reaching beyond research and diagnostics to take over some surgeries, but this will be done in such a way that augments human performance rather than replacing it.

Patients will certainly find peace of mind in the precision offered by AI and machine learning. Still, it’s hard to imagine anyone preferring to receive a diagnosis through an impersonal printout rather than an in-depth consultation with an empathetic doctor who can help them with the uniquely human aspects of medical care. 

Scripps Research Translational Institute founder Dr. Eric Topol is a cardiologist, geneticist, and digital medicine researcher. He described the impact AI will have in healthcare in this way: “It’s nice to have great minds, but we need to put an even higher priority on the humane elements, the essence of medicine, which is a connection between patients and their doctors.” He foresees AI taking over the more technical aspects of research, diagnostics, and treatment, freeing doctors to collaborate with patients in a manner that promotes the best possible health outcomes.