Osteoarthritis triaging tool: Cutting wait times and enhancing patient outcomes

Maxine Rickman, Flex Healthcare Pty Ltd.

Challenge

Osteoarthritis (OA) is a growing global health concern, with knee OA being the most common form, affecting about 4% of the global population. In Australia, an aging population and rising obesity rates contribute to increasing OA prevalence, while delays in public orthopaedic clinics—worsened by the COVID-19 pandemic—hinder timely management. There is a need for innovative solutions to expedite the triage and management of knee OA patients.

Solution

The project developed an AI-based application to detect and assess the severity of knee OA from radiographs, streamlining the triage and management process. By employing advanced deep learning models, including Convolutional Neural Networks (CNNs), the application aims to accurately classify X-rays for the presence and severity of OA, improving efficiency and decision-making in clinical settings.

Key Features

• Uses deep learning models trained on annotated datasets for high-accuracy OA detection.
• Provides automated assessment of OA severity from radiographic images.
• Validated through a real-world pilot with patients at Alice Springs Hospital, comparing AI outputs with manual assessments.
• Designed for ease of use and integration into existing clinical workflows.

Impact

The AI tool has demonstrated high accuracy in predicting knee OA severity, receiving positive feedback from clinicians regarding its usability. The solution offers the potential to expedite the triage process for knee OA patients, enhance decision-making in patient management and alleviate pressure on overburdened orthopaedic services, especially in the context of pandemic backlogs.

Next Steps

The project team will focus on integrating the AI tool into standard outpatient procedures and expanding its use in other healthcare settings. Additional user testing and feedback will guide further refinement, with the goal of achieving broader adoption and enhancing the tool’s impact on knee OA management.

Advancing health monitoring for rural and remote elderly Australians

Dr Stephanie Baker, James Cook University

Challenge

Elderly Australians living in rural, regional, and remote areas face significant barriers to accessing healthcare compared to their urban counterparts. This geographic disparity leads to higher rates of injury, chronic diseases, and premature death. The stress associated with travel for medical care exacerbates these issues, highlighting the urgent need for innovative solutions that can facilitate early detection and management of health problems for these isolated communities.

Solution

This project addresses this challenge by harnessing the power of wearable technology to monitor health parameters continuously and non-invasively. The team developed a comprehensive database by integrating physiological data from the Empatica EmbracePlus wearable device with traditional physical and mental health measurements. This innovative approach allows for real-time health monitoring, which is crucial for early intervention and improving health outcomes.

Key Features

• Wearable device data: The project collected continuous physiological data from the Empatica EmbracePlus, including heart rate, sweat levels, and motion activity. Additional metrics such as steps and temperature were also recorded.

• Reference measurements: Participants underwent physical and cognitive tasks while wearing the device, with vital signs (heart rate, blood pressure, respiratory rate, blood oxygen saturation, and temperature) recorded at different stages of activity. Self-assessments of stress and fatigue were also included.

• Timestamped data: All data points were timestamped, allowing for precise linkage between wearable and reference measurements.

Impact

This database provides a rich resource for developing new methods of health monitoring using wearable technology. It supports the exploration of relationships between wearable data and various health parameters, potentially leading to the identification of novel biomarkers. Future research will focus on continuously and non-invasively measuring blood pressure and respiratory rate through AI models, and investigating how wearable data correlates with stress and fatigue.

Next Steps

The team will work on refining AI models to enhance continuous monitoring capabilities and deepen our understanding of how wearable data can improve health management for elderly Australians, irrespective of their location. By advancing wearable technology and data analysis, this project aims to contribute to a future where high-quality, accessible health monitoring is a reality for all elderly Australians.

A mobile app and dashboard for effective management of early-stage chronic kidney disease

Dr Euijoon Ahn, James Cook University

Challenge

Chronic kidney disease (CKD) is a growing health concern in Australia, particularly in rural and regional communities where access to specialised care is limited. Early-stage CKD management is often delayed due to a lack of tailored resources and digital solutions, leading to poor health outcomes and increased healthcare costs.

Solution

The CKDFree project addresses these challenges by developing a user-friendly mobile application designed to support early-stage CKD management. Bringing together healthcare professionals, IT specialists, and community stakeholders, the app offers innovative digital tools to enhance patient engagement, improve health literacy, and empower users to manage their condition more effectively.

Key Features

• Personalised health dashboards to track key health metrics.
• Educational resources tailored to CKD management.
• A digital ‘health priorities’ summary to guide patient-provider discussions.
• Tools for recording and monitoring health data, designed with input from end-users to ensure the app meets their needs.

Impact

Preliminary feedback from the pilot testing phase shows positive results, with participants indicating improved engagement and understanding of their condition. The project is expected to contribute to the body of knowledge on digital health interventions for chronic diseases, enhance the skills and knowledge of healthcare providers and patients, and improve health outcomes, reduce healthcare costs, and boost quality of life for individuals with CKD, especially in rural and regional areas.

Next Steps

The CKDFree project will continue to refine and expand the app, incorporating feedback from ongoing pilot testing. The next phase includes scaling the app’s implementation to reach a broader population, further enhancing its functionality, and solidifying its role as a scalable and sustainable solution for CKD management across diverse communities

Mapping statewide EMR observation records to OMOP CDM to increase interoperability

Dr Han Cham Lim and Professor Clair Sullivan, The University of Queensland

Challenge

Many greenfield sites transitioning to the OMOP Common Data Model (CDM) face challenges in establishing a suitable development and test environment. Organisations also require guidance for generating synthetic Electronic Medical Records (EMR) data and mapping existing data systems, such as Cerner, to the OMOP CDM.

Solution

The project produced a comprehensive guidance document detailing how to set up an OMOP CDM environment, along with scripts to generate synthetic EMR data. This solution enables sites to begin development work and training while waiting for ethical and regulatory approval to access real production data.

Key Features

• A step-by-step guide for setting up an OMOP CDM environment for research and development.
• Scripts for creating synthetic EMR data to facilitate training, education, and development.
• Detailed mapping guidance for organisations using Cerner to transition smoothly to the OMOP CDM.

Impact

The guidance document addresses a significant gap by helping organisations transition smoothly to the OMOP CDM, supporting data integration and analysis. The ability to generate synthetic data allows early development work and familiarises users with the OMOP environment before using real data.

Next Steps

Continue refining the guidance based on user feedback, expanding the support materials for other EMR systems beyond Cerner, and develop additional resources to enhance training and capacity-building efforts for organisations adopting the OMOP CDM.