5 Real World Use Cases Of AI In Healthcare

Artificial intelligence powered medical technologies are increasingly finding applicability in clinical practice. According to a recent research report, in 2022, the software solutions segment dominated the market for artificial intelligence in healthcare and accounted for the largest revenue share of 40.2% owing to the rapidly growing adoption rate of AI-based software solutions amongst healthcare providers, payers, and patients.

Hospitals, healthcare organizations and medical practitioners are rapidly looking to adopt AI-powered solutions to take their practice to the next levels by boosting efficiency while keeping long-term costs in check. The past couple of years have seen a steep rise in technological collaborations with companies such as Arkenea that are pioneering in AI software development in the healthcare industry. 

Here are 5 real world examples of how artificial intelligence technology is being harnessed in the healthcare sector. 

1. Leveraging AI To Make Safety Huddles More Efficient – University Hospitals, Cleveland

The Problem

Safety huddles are an indispensable part of the treatment regimen that takes place at the beginning of every nursing shift. The focused exchange of information about existing or potential risks which may affect the patients requires visual feedback of data to achieve the goal of reducing harm. 

UH Cleveland Medical Center, Ohio, was looking for ways to improve the efficiency of the safety huddles and integration of AI in the daily huddle design has turned out to be a game changer. 

The Solution

University Hospital’s nurses collaborated in creation of EdgeHuddle, an AI-based platform that automates the process of collection and analysis of patient data collated from various sources including EHRs and medical devices. The pain points in the existing huddle process were identified and designing of the AI-powered tool in collaboration with the nurses has significantly improved the flow of information and data driven insights through all layers of the health system. 

By using artificial intelligence algorithms and predictive analysis, the framework identifies patterns and trends related to patient safety. The technology not only provides real-time risk assessment but also helps in formulating a plan of action for the care teams on the basis of patient needs identified. 

“The action-oriented lens supports our Zero Harm goals by providing a summary level view of Safety and Zero Harm risks for use during our organizational Tiered Safety Huddles,” stated Peter J. Pronovost, MD, PhD, FCCM, UH Chief Quality and Clinical Transformation Officer. “Displaying this clinical data for our entire health system will provide efficiency in our current quality and safety monitoring processes, allowing staff to intervene to prevent harm. This is a big advance on our journey to Zero Harm.”

The technology automates data collection and by identifying patterns and trends related to patient safety, it gives out actionable insights to support staff intervention thereby improving patient outcomes. 

2. Using Artificial Intelligence To Bring Down Unplanned Hospital Readmissions – NSW Hospital

The Problem

Readmissions are a costly affair for hospitals which often end up doubling the cost of care. Hospital readmissions are thus one of the key performance indicators. 

A rural health hospital in New South Wales was experiencing high levels of patient readmissions. Managing this required enormous amounts of resources and staff, making it a financial liability. The hospital wanted an AI solution that would parse through the existing data to determine which patients were at a high risk of readmission. 

The Solution

An AI-powered machine learning model was developed that was trained on the data of electronic medical records from the last ten years. The software took into consideration patient demographics, medical histories, treatment plans and outcomes. 

Leveraging artificial intelligence combined with clinical decision support the software showed 70 percent accuracy when it came to identifying patients with unexpected readmission within a 28 day period. The information obtained by the AI software is being used to help clinicians make real-time decisions that are directly improving the quality of care. 

“By working in advance of recovery barriers and focusing on whole-person needs, real rates of readmission can be reduced, even for people at high risk for return to acute care,” said Alejandro Quiroga Chand, MD, senior vice president, chief medical officer ambulatory care and population health.

By identifying patients at greatest risk for readmission, a focused transition support plan and targeted interventions were formulated. Proactive outreach plans and task oriented follow ups have led to a significant impact on patient care and improved the care delivery system performance metrics. 

3. Generative AI In Healthcare For Clinical Documentation – University of Kansas Health Systems 

The Problem

Clinical documentation is a cumbersome and time consuming activity that has direct correlation with increased incidences of physician burnout. It was noted that physicians in the University of Kansas Health Systems, on an average, spend about 130 minutes outside of work fulfilling their clinical documentation responsibilities. 

Simple speech to text software and medical scribes were both unreliable and expensive so the health system wanted to leverage smart tech at its 140 hospital and clinic locations and saw the potential artificial intelligence had for clinical documentation. 

The Solution

University of Kansas Medical Systems partnered with a leading healthcare technology company to roll out an AI-based medical transcription solution for its 1,500 physicians. The software records the visit, converts it into text and summarizes the most important parts for both physicians and patients. The breakthrough comes in the form of AI technology that identifies over 90 percent of key points in the provider-patient conversations during the visit. An initial transcript generated within minutes of the conversation ending which can be converted into the summary format the physician prefers. 

AI powered interactive editing tools then accelerate the editing process. The software also integrates with other healthcare software such as EHRs to streamline documentation and promote interoperability. 

Dr. Gregory Ator, Chief Medical Information Officer and Head and Neck Surgeon at The University of Kansas Health System sees great potential in the breakthrough that generative AI offers in the field of health. “This cutting-edge technology will not only close the documentation cycle in real-time but also improve the overall quality and consistency of our clinical notes. This technology represents a major step forward in reducing burnout, improving provider satisfaction, and ultimately enhancing the delivery of patient care.”

4. Clinical Decision Tool To Identify Pregnant Women At Risk – Tommy’s National Center For Maternity Improvement

The Problem

Routine pregnancy checks often miss out on life threatening growth restrictions which is a major contributing factor for stillbirth and premature births. Lowering the risk of preterm birth or stillbirth was the challenge that Tommy’s national center for maternity improvement wanted to address. 

The growing need for personalized care during pregnancy and the need to eliminate inequalities in care in ethnic minorities, led to the development of an AI-powered clinical decision tool – Tommy’s app. 

The Solution

The information gathered at pregnancy checkups is processed by the AI algorithms to assign a risk score to every patient. The tool enables midwives and doctors to assess each woman’s needs during pregnancy more accurately and to personalize their care, using the latest evidence based practices. 

The healthcare providers can identify which pregnant women are at the risk of developing complications during pregnancy that can lead to stillbirth. The clinical decision tool also enables identification of women who are at most risk for premature birth. 

The app itself has a dual interface, both for providers as well as patients. The comprehensive risk assessment and decision support enables the healthcare practitioners to deliver more effective care. The interface for maternity service users contains the information hub which has allowed them access to information they need in accessible and understandable terminology. This has proved to empower the pregnant women to be more engaged in their own care regimen. 

The tool has also shown promise in reducing health inequalities in ethnic minorities. The perinatal death rate was three times greater in ethnic minorities. However, using this tool in combination with targeted care, the perinatal death rates came down and became the same across all ethnicities. 

5. Artificial Intelligence For Identification Of Surgical Implants – Kethan Solutions 

The Problem

Orthopedic implants are a go to treatment to avoid disability and impairment of function in conditions such as fractures, spondylosis and disc herniation. A number of cases require repeat surgeries which take place years after the index surgery and patient records are unavailable. 

Accurate identification of the existing implant is imperative for preoperative evaluation but manually identifying the brand and model of the implant often proves to be time consuming. There is also a huge margin of error involved which adversely affects surgical planning. Seemingly minor differences in screw head shape or depth can make removal impossible without the correct tool. 

The Solution

Using AI to automate identification was what Brain Murray of Kethan Solutions LLC targeted when he approached Arkenea for developing an artificial intelligence based iOS application. After several brainstorming sessions and whiteboard discussions, a solution roadmap was implemented to develop a mobile application that ran on artificial intelligence and used machine learning to identify implants.

The implant and its attributes like the manufacturer’s name could be identified based on the algorithm’s analysis of the radiological image. The AI not only successfully identified the implant in the X Ray, but also matched the uploaded image correctly with the existing dataset it was trained on, irrespective of the image’s orientation. 

By managing a consolidated dataset of implant information, the software made it easier for radiologists and surgeons to plan their surgeries by successfully identifying the implant preoperatively. 

Have an idea that leverages artificial intelligence in healthcare?

Looking for a custom software solution to solve a problem for your medical practice or have an bespoke software idea that leverages artificial intelligence? We can help!

Arkenea is one of the pioneers in the healthcare software industry that leverages cutting edge technologies like artificial intelligence to deliver innovative software solutions. Get in touch with our consultants to discuss your project today.