AI Australia lives and breathes data. It’s the correlations and the aggregations, which, combined with factors like the average stiletto height, can provide the most powerful insights out of everyday information. We turn our heels to the Emergency Department, where the patient visit volume on a good day can make you, and on a bad day will break you. It’s difficult to predict the changing demands of hospital resources when the visit volume can be impacted by the day of the year, holidays, weather conditions, local events, and every mosh pit gone wrong at Coachella. We can’t see the future yet… or can we? AI Australia took to the wine country with our predictive swag to work with the NSW Finance and Performance department. A rural health district was looking for an AI solution to forecast the daily visit volumes for the next 365 days. This is capital C Critical when hospitals share staff, equipment, supplies, and ambulance vehicles. Time and resources cost big money, and we know that in the health industry, money is a puddle of water on a hot day. Our data wizards took on the challenge to create a AI solution to improve the health service and resource utilisation within the rural health district.
The hospital wanted an end-to-end AI solution to predict the variations in visit volumes with the factors impacting them. A yearly forecast of this data would help hospitals within the rural health district to distribute resources to meet the visit volume demands in a more efficient and accurate way. The hospital provided over a decade of electronic medical records which our intelligence team added to the Matrix to create a time series forecast. The patterns and associating factors were seized from the data and transformed into a forecast plotting a predicted future for these factors.
AI Australia has over 20 years’ experience working with data and over 10 years’ experience providing data-related solutions in the health care industry. We are uniquely positioned to ensure you get the most powerful insights out of everyday information using health-focused intelligent agents. It’s just what we do.
Our experience demonstrating the ROI of data science in rural health districts makes us the perfect partner to transform your health initiative. While developing the solution, we collaborated and planned at Scrum meetings, allowing us to receive feedback and revise the product to provide the most precise and customised solution.
The data you collect today can inform the business success of tomorrow. It’s what you do with it that makes all the difference. The time series forecast model created to analyse daily visit volumes used indicators including seasonal patterns, holidays, and more to identify volume variables more precisely. Out of the four different prototype models created, the most optimised model was chosen for the final results. The data and model were running in real-time on SQL Server on premise with Microsoft SQL Server R Services. The final results were presented in a Microsoft Power BI report, which was transferred to the client through a SharePoint page with LiveTiles.
Time is precious, so we want to make the most of it. It took two weeks to understand the client’s needs and identity what AI solution would best suit the client’s problem; two weeks to collect data and perform data engineering; three weeks to build predictive analytics models and perform model selection and refinement; and one week to deliver the results as a product.
When data sets for the two largest hospitals in the rural district were tested against our models, the generated forecast had an overall accuracy of ninety-percent. This number decreased for small facilities due to random variations in daily visiting.
This solution proved that machine learning methods can use predictive analytics to forecast the variations of visit volumes and the potential impacting factors for a 365-day period. Hospitals can use this data to distribute resources throughout the year and meet the patient volume demands proficiently and economically.
This project demonstrated the potential for machine learning methods to provide a return on investment lost through inefficiency and preventable mistakes. AI is no longer a secret to success, but now an obtainable solution to transform yesterday’s problems into a brighter tomorrow.