Advancements in artificial intelligence (AI) and machine learning have paved the way for innovative applications in various fields, including healthcare. One such application is the development of AI algorithms capable of predicting an individual’s health outcomes later in life. This emerging technology holds the potential to revolutionize healthcare by enabling personalized risk assessments and preventive interventions.
With the simple press of a button, individuals could gain valuable insights into their future health, empowering them to make informed decisions and take proactive measures to mitigate potential health risks. This study explores the current landscape of AI-based health prediction and its implications for personalized healthcare, highlighting the opportunities and challenges associated with this transformative approach.
Thanks to artificial intelligence (AI), predicting the risk of developing severe health conditions later in life will be as easy as pressing a button. Abdominal aortic calcification (AAC), a marker associated with cardiovascular disease, falls, fractures, and late-life dementia, can be detected through bone density machine scans commonly used for osteoporosis diagnosis. However, analyzing these scans requires highly trained experts and can be time-consuming.
Researchers from Edith Cowan University have collaborated to develop software that can analyze approximately 60,000 images in a single day, significantly improving efficiency. This advancement is expected to facilitate the widespread use of AAC in research and assist individuals in preventing future health problems.
Researcher and Heart Foundation Future Leader Fellow Associate Professor Joshua Lewis said, “This significant boost in efficiency will be crucial for the widespread use of AAC in research and helping people avoid developing health problems later in life. Since these images and automated scores can be rapidly and easily acquired during bone density testing, this may lead to new approaches for early cardiovascular disease detection and disease monitoring during routine clinical practice.”
The study was a significant international collaboration involving Edith Cowan University (ECU), the University of WA, the University of Minnesota, Southampton, the University of Manitoba, the Marcus Institute for Aging Research, and Hebrew Senior Life Harvard Medical School, demonstrating a genuinely multidisciplinary global effort. While previous algorithms have been developed to assess abdominal aortic calcification (AAC) from bone density machine images, this study stands out as the largest of its kind, based on commonly used machine models. It is the first to be tested using routine bone density scans in a real-world setting.
The software developed by the team achieved an 80% agreement with expert analysis in determining the extent of AAC (low, moderate, or high), a remarkable outcome considering it was the initial version of the software and over 5,000 images were analyzed.
The software developed in the study demonstrated high accuracy in diagnosing abdominal aortic calcification (AAC) levels. Only 3% of individuals with high AAC levels were misdiagnosed as having low levels by the software. This is crucial as these individuals are at the most significant risk of cardiovascular events and mortality. The research team has improved the software’s accuracy in more recent versions.
This automated assessment allows large-scale screening for cardiovascular disease and other conditions, enabling early intervention and lifestyle changes. The Heart Foundation funded the project through Professor Lewis’ Future Leadership Fellowship.
Journal Reference:
- Naeha Sharif, Syed Zulqarnain Gilani, et al., Machine learning for abdominal aortic calcification assessment from bone density machine-derived lateral spine images. eBioMedicine.DOI:10.1016/j.ebiom.2023.104676.