Artificial Intelligence: Shaping Healthcare Today and Tomorrow  

From enhancing patient care and outcomes to accelerating medical research and data management, artificial intelligence (AI) has been absorbed into a wide range of medical processes and procedures. That’s because when used prudently, it can empower physicians and clinical staff to provide a greater level of patient care with increased accuracy and efficiency. Though there are benefits to using AI, human oversight is needed to mitigate risks and ensure its application remains both safe and effective.
 
The use of AI in medicine dates to the mid-20th century, with early automation and computation tasks, development of clinical decision support systems and applications of pattern recognition for medical diagnostics. Recent technological developments have dramatically expanded potential use cases. According to a recent AMA report, in 2023 an AI large language model algorithm successfully passed the multiple-choice section of the United States Medical Licensing Examination — just one example of AI’s vast and rapidly expanding capabilities.

Artificial Intelligence Defined 

AI technologies perform tasks typically requiring human intelligence, such as visual perception, speech recognition and decision making. In clinical settings, AI software performs a series of complicated or repetitive tasks, relying on an in-depth model it has previously created and analyzed in detail.
 
AI models are developed by gathering massive amounts of information — from existing databases, books, studies, academic research, art, audio and video recordings, music, films, historical data, news and all modes of internet and online content — to emulate behaviors needed to accomplish a specific task. Then, through a process called machine learning, the system digests this knowledge and allows real-time access to data to perform actions or make decisions. AI applications range widely from autonomous driving to predictive analytics. AI used for medical purposes is trained and fine-tuned specifically with medical data. 
 
Technologies known as large language models (LLMs) and generative AI fall under the broad umbrella of AI and drive many high-profile commercial applications — such as OpenAI’s ChatGPT. LLMs generate text, while generative AI creates entirely new content, including text, images, audio, music and more. In healthcare, those products could include patient education materials, medical documentation, billing and virtual healthcare assistants. 

Easier Administration, More Accurate Diagnosis, Personalized Healthcare 

Though many AI applications build on and interact with each other, we divided AI applications in healthcare into the following five categories, along with some of their major uses:

Administration 

Physicians already recognize the value of AI in supporting administrative functions as well as clinical tasks, with 56% of those surveyed by the AMA identifying “addressing administrative burdens through automation” as the biggest area of opportunity, according to the organization’s recent report. AI applications can help practices manage administrative data and tasks much faster and with more accuracy. 
 
One of the fundamental ways automation is transforming healthcare is by simplifying routine and repetitive tasks, such as patient scheduling, billing and claims processing — all areas that are time-consuming and prone to human error. Such increased accuracy directly impacts productivity, freeing up people to focus on more patient-centric tasks. For instance, automated coding and billing ensures accurate invoicing, which in turn reduces the time spent resolving billing disputes. Administrative areas AI assists in include: 

• Access to care. Optimize scheduling and support the prior authorization process. 
• Administration and revenue cycle. Identify appropriate billing and service codes and predict and reduce claim denials. 
• Operations. Predict hospital volumes and staffing needs, track inventory to forecast medical supply orders and monitor equipment availability and predict failures. 
• Patient experience and satisfaction. Analyze feedback to identify improvements and identify what drives patient trust. 
• Patient communication. Respond to portal inquiries, generate correspondence and translate information into different languages. 
• Quality improvement and management. Track and report quality outcomes and identify quality gaps or inequities in patient care. 
• Regulatory compliance and reporting. Automate tracking and reporting of compliance measures and analyze documentation and processes to ensure adherence to laws and policies. 
• Education. Monitor training, identify learning needs and resources and provide feedback during robotic training. 

Documentation 

Another promising use of AI is enabling electronic health record (EHR) systems to become more flexible and intelligent. With the expansion of telemedicine during and post-COVID, doctors are spending as much as 68% of their time working in EHRs, one study showed. While only about 13% of physicians surveyed by the AMA currently use AI to assist with documentation, adoption is expected to grow.
 
One of AI’s main improvements to documentation is natural language processing (NLP). AI-powered algorithms can extract relevant information from unstructured clinical notes to populate structured fields in EHRs, reducing the burden of data entry and providing more accurate and complete documentation. Taking that a step further, AI-powered voice recognition technology can convert spoken words into text for inclusion in EHRs. And AI also can analyze clinical documentation to identify inaccuracies and inconsistencies to help ensure that medical records are complete.

Imaging and Testing 

The most extensive use of AI in diagnostics so far has been in imaging. As of October 2023, the U.S. Food and Drug Administration (FDA) had approved 692 artificial intelligence/machine learning medical devices, of which 531, or 77%, were for use in radiology, the AMA says. Uses include: 

AI-enhanced software that can recognize complex features in medical images nearly impossible for humans to see. AI algorithms can offer a new perspective about which image features should be considered valuable in supporting clinical decisions. They also can analyze medical images with speed and precision, aiding in the identification of early-stage diseases that may be difficult to detect through traditional methods.

New possibilities in image segmentation and quantification. By employing sophisticated algorithms, AI can accurately delineate structures of interest within medical images, such as tumors, blood vessels or cells, helping clinicians precisely target areas for procedures and therapies.

Advances in image-guided interventions and surgical procedures. By combining preoperative imaging data with real-time imaging during surgery, AI algorithms can provide surgeons with augmented visualization, navigation assistance and decision support.
 
Similarly, AI is used in pathology for speedy and accurate image analysis, helping pathologists identify abnormalities and assisting in diagnosis by highlighting areas of concern on slides or providing differential diagnoses. By analyzing large datasets of patient information and pathology reports, AI can help predict disease progression, treatment response and patient outcomes. And it can streamline pathology workflows by automating repetitive tasks such as slide preparation, staining and data entry, allowing pathologists to focus more on complex cases.

Clinical Decision Support 

One of AI’s most promising roles is in clinical decision support at the point of patient care. AI algorithms can analyze a vast amount of patient data to assist providers in making accurate diagnoses and better-informed decisions about treatment. Using the same NLP that assists in administration and documentation, AI can analyze patient data in the electronic record in real time to provide personalized treatment recommendations, alert healthcare providers to potential drug interactions or adverse events and suggest evidence-based guidelines for diagnosis and management. Other applications include:  

Risk assessment and disease prevention. With its ability to process and evaluate large volumes of patient data, AI can identify patterns, trends and risk factors for various diseases and conditions, helping providers proactively intervene to prevent adverse outcomes.

Genomic analysis. AI is used to analyze genomic data to identify genetic mutations associated with certain diseases or predict patients' response to treatments. This information can guide personalized treatment plans. 

Virtual health assistants. AI-powered virtual assistants can answer health-related questions, schedule appointments, remind patients to take medications and offer personalized health advice. These assistants can also collect symptom information and assist in triaging patients.

Remote Monitoring 

Smart watches and fitness trackers that use AI algorithms to track various health metrics have been joined by more sophisticated AI-enabled wearable devices such as continuous glucose monitors, ECG and respiratory monitors, as well as seizure or fall detection and pain management devices.

AI algorithms can analyze the continuous streams of medical data such devices produce to detect patterns and trends that may indicate changes in a patient's health status. AI analysis also can provide personalized health insights such as predicting disease risks, recommending lifestyle changes and optimizing medication schedules. Because monitoring is continuous, AI can serve as an early warning system, alerting healthcare providers to potential health emergencies.

The Power to Transform Healthcare 

As AI continues to evolve, its role in healthcare will undoubtedly expand, reshaping the landscape of medical practice and fostering a future where healthcare is more accessible, efficient and effective. While artificial intelligence has the potential to transform the healthcare sector, the technology also brings risks — such as errors, bias and privacy and security concerns. It will change the traditional role of physicians, who face both opportunities and challenges in adapting to this new environment. Some specialties, such as radiology, may be more impacted by AI than others, but doctors will not become obsolete. Rather, the power of artificial intelligence lies not in standalone applications, but in its use alongside a physician’s expertise and judgment.

As chief technology officer of MagMutual, Nevarda Smith has spearheaded multiple successful AI projects.

Disclaimer: The information in this article concerning artificial intelligence and its use in healthcare is intended for general informational purposes only. While efforts are made to ensure the accuracy and reliability of the information presented, MagMutual cannot guarantee its completeness, suitability or validity for any particular purpose. Users are advised to verify information about the use of artificial intelligence in healthcare with other credible sources and to exercise their own judgment when applying it to specific situations.