Pilot Study: Using Machine Learning and Optical Coherence Tomography Angiography for Detecting Systemic Cardiovascular Illnesses and Cardiometabolic Risk Factors


This study explored the potential use of machine learning (ML) and optical coherence tomography angiography (OCTA) to detect systemic cardiovascular illnesses and cardiometabolic risk factors. Conducted at the Royal Adelaide Hospital, Australia, from January 2019 to December 2019, the study recruited patients from the Department of Cardiology and Ophthalmology. Participants were included after informed and written consent were gained, with exclusion criteria including individuals with pre-existing retinal vascular diseases, hemodynamic instability, inability to position for scanning, and severely myopic eyes. After ethics approval was obtained, the following data were collected: age, gender, height, weight, comorbidities, smoking status, current medications, and most recent blood tests.

OCTA scans were performed without the use of mydriatics using the Carl Zeiss CIRRUS HD-OCT Model 5000. Image pre-processing and machine learning involved the use of open-source Python libraries, namely Sci-Kit Learn and Tensorflow. After the images were resized to 256 × 256 pixels and re-scaled, the total dataset was randomly split into a training and a testing dataset (75%/25% split). Random transformations were then applied to the images in the training dataset, including a degree of rotation, width shifts, and horizontal flipping.

Two models were developed using convolutional neural networks, the first of which incorporated one 2D convolutional layer, one 2D maximum pooling layer, and one dense layer; the second using transfer learning and MobileNetV2. Performance was assessed on the unseen test dataset; Area under the receiver operator curve (AUC) was calculated using the trapezoidal rule and Youden’s index determined the cut-off scores.

Conducted at the Royal Adelaide Hospital and Oceanic Health Care Centre, the study was led by Dr. Mark T. Smith and Dr. Willie C. Wong, with ethics approval from the Central Adelaide Local Health Network Ethics Committee. Their team was excellent at combining quality data collection with the use of machine learning and optical coherence tomography angiography to provide a clinically useful system for predicting systemic cardiovascular illnesses and cardiometabolic risk factors.

See also  Unmasking Algorithmic Bias in Medical Imaging: New Insights and Mitigation Strategies

Frequently Asked Questions (FAQs) Related to the Above News

Please note that the FAQs provided on this page are based on the news article published. While we strive to provide accurate and up-to-date information, it is always recommended to consult relevant authorities or professionals before making any decisions or taking action based on the FAQs or the news article.

Share post:



More like this

Google Introduces Ads in AI-Generated Search Summaries To Compete with ChatGPT

Google introduces ads in AI-generated search summaries to compete with ChatGPT, enhancing user experience and revenue potential.

Cybercriminals Embrace AI for Efficient Global Operations

Discover how cybercriminals are utilizing AI to enhance their global operations, raising concerns about misuse and performance issues.

European Countries Approve Landmark AI Act for Regulation and Innovation

EU countries endorse AI Act, paving way for innovation & regulation. A groundbreaking milestone for global AI governance.

AI Threat: Will Artificial Intelligence Lead to Humanity’s Extinction?

Discover the potential threats posed by AI and the necessity of preparing for job displacement and inequality in the wake of artificial intelligence development.