There seems to be no doubt left among healthcare professionals on the benefit that Artificial Intelligence (AI) can bring to the table. Whether its clinical applications, administrative systems, or claims management, the solutions range across the board. Every hospital now knows that AI is the answer to the challenge of staying ahead in the healthcare industry. The number of vendors that offer AI and machine learning solutions are numerous and their effects come with the potential to transform operations across medical practices of variant sizes. According to a recent study titled ‘Artificial Intelligence for Health and Healthcare’ by The MITRE Corporation, it was observed that, “Unlike previous eras of excitement over AI, the potential of AI applications in health may make this era different because the confluence of the following three forces has primed our society to embrace new health-centric approaches that may be enabled by advances in AI:
- Frustration with the legacy medical system
- Ubiquity of networked smart devices in our society
- Acclimation to convenience and at-home services
Artificial Intelligence in a Nut Shell:
Although AI is a broad umbrella that embraces many different types of next-generation technologies, specific to the healthcare industry, it offers the capacity to mimic superior human intelligence and tasks that cover knowledge management, recognition of processes, processing of administrative tasks, data automation, etc.
The automation angle of Artificial Intelligence allows for mundane tasks to be performed more efficiently and faster, with a reduced possibility of errors. It can be likened to the adoption of an employee with superior robotic capacities. In this article, we will explore the following questions:
- What are the ways in which hospitals have begun to adopt these technologies?
- What are the benefits of these adoptions?
- How can hospitals leverage artificial intelligence?
Adoption of AI technologies:
AI technologies can be leveraged to improve functionality across the following areas:
- Improved Patient Access
- Intelligent Outpatient Scheduling
- Enhanced MRI Scans
- Management of ORs and Lab Timings
- Automated Preauthorization Procedures
- Predictive Claims Denials/ Rejects
- Identification of Medicine Variants
Improved Patient Access:
In a hospital, one of the primary drivers toward success is the streamlining of processes related to appointments and the speedy manner in which patients gain access to their caregivers. This accessibility remains a priority for every hospital across the country. Referral rates, revenue, and satisfaction of patients are all dependent on the efficiency of this process. There are several AI technologies that can be deployed to efficiently manage patient appointment scheduling in such wise that the patient receives the access to healthcare in the quickest possible way. The analytical tools of these solutions reduce the need for human intervention and can save hospital hours of manual work.
According to ODPHP, “Access to health services means “the timely use of personal health services to achieve the best health outcomes.” It requires 3 distinct steps:
- Gaining entry into the health care system (usually through insurance coverage)
- Accessing a location where needed health care services are provided (geographic availability)
- Finding a health care provider whom the patient trusts and can communicate with (personal relationship)
Intelligent Outpatient Scheduling:
Patient appointment scheduling is a process that demands considerable hours of labor and usually results in dissatisfied patients and frustrated staff. Set timelines fail to identify the true need of patients and the manual process is not geared to catch cancellations and no-shows. Further, follow-ups are usually even more tedious and disorganized. Every no-show is costly and can affect the overall revenue of a hospital.
AI solutions address these pain points through:
- Through the use of past data, demographics, complaints, locations and other factors, AI technologies use predictive analytics to detect delays and no-shows, allowing for schedule optimization.
- These same analytic tools can intelligently determine the length of each appointment and allow scheduling to maximize gains.
Enhanced MRI Scans:
Caregivers rely heavily on MRI scans in order to gauge the level and kind of care the patient requires for brain-related and other issues. The patients are required to stay still for these scans for efficient outcomes.
“A research team led by Dr. Matthew S. Rosen of Massachusetts General Hospital, Martinos Center for Biomedical Imaging, and Harvard University set out to improve image reconstruction by harnessing the power of machine learning. The work, which was funded by several NIH components, was published on March 21, 2018, in Nature.”
Naming the technology AUTOMAP, the researchers found that machine learning was capable of producing a more powerful version of the image than the conventional scanner. Further, they “found that AUTOMAP enabled better images with less noise than the conventional MRI. The signal-to-noise ratio was better for AUTOMAP than conventional reconstruction (21.6 vs. 17.6). AUTOMAP also performed better on a statistical measure of error known as root-mean-squared-error (6.7% versus 10.8%). In addition, AUTOMAP was faster than the manual tweaking now done by MRI experts.”
Imaging capabilities that can be enhanced through artificial intelligence are not limited to MRIs and significant improvements have been observed in retinal scans and classification of skin cancer as well.
Management of ORs and Lab Timings:
Operating Rooms (ORs) and labs are important service areas within a hospital that require prior scheduling for relevant procedures. The management of these areas is a tedious task and if not executed efficiently, can end up costing the hospital more revenue that earned. Mismanagement of these rooms is a hurdle faced by most hospitals across the country. Extended time slots and the effect of that on next-in-line schedules, under-utilization, etc. are all detrimental factors toward hospital revenue.
AI tools can efficiently schedule usage of these rooms, and also offer insights into maximizing profits through analytics. This is done through:
- Tracking arrivals
- Analysis of time for preparation
- Analysis of time for procedure relative to the physician
Automated Preauthorization Procedures:
The pre-authorization process is generally one that requires a large amount of administrative staff involvement with immense room for errors. According to studies by the American Medical Association and the Journal of the American Board of Family Medicine, “today’s pre-authorization methods can take 20 hours of labor per week per physician and lead to a total annual cost of $600-700M in the US.” Not only is the process labor intensive, but the length of the preauthorization procedure discourages patients and delays usually lead to decreased revenues. When preauthorization is ignored, reimbursement turns into a risk.
AI technologies are geared to automate this process through:
- Language processing
- Context interpretation
- Identification of clinical information
- Management of patient records
Predictive Claims Denials/ Rejects:
As every hospital caregiver and an owner would agree, this is the lifeblood of the revenue cycle. This is where intelligent software is an imperative requirement. Tracking claims and denials is a task that is overburdening administrative staff and is now virtually impossible to execute without AI technologies. However, while choosing the right technology, it is of primary importance to select one that offers predictive analysis that will assist in making the required modifications to avoid rejections and denials.
The benefits of adopting this technology are as follows:
- Providers get reimbursed faster
- Rectification of errors before being claims are processed
- Problem areas are flagged and corrections are recommended
- Reduced denials and rejects
- Increased revenue
Identification of Medication Variants:
An analysis of the operations and performance of the hospital system is the first step towards implementing change that will optimize the key practices. This analysis can be achieved through machine learning, with an insight into expenses, profits, and costs related to medications. The pharmacy, usually, ends up being among the highest cost factors in a hospital. Automated technologies analyze medications that offer the same results but are less in cost, thereby increasing savings.
AI technology offer solutions for medication variation through:
- Patient profile analysis
- Physician records
- Past medication documentation
- Relative costs
In the healthcare space, with special relevance to hospitals, it is now of paramount importance to leverage these technologies and reduce costs, while increasing revenues. Further, these significant technologies will improve the overall efficiency of operations, and finally result in overall patient satisfaction.