Co-training AI tool can mimic the second opinion process
A new co-training AI algorithm for medical imaging, developed by researchers from Monash University, is said to mimic the process of seeking a second opinion.
The research aims to address the limited availability of human annotated, or labelled, medical images by using an adversarial, or competitive, learning approach against unlabelled data.
PhD candidate Himashi Peiris of the Faculty of Engineering said the research design had set out to create a competition between the two components of a “dual-view” AI system.
“One part of the AI system tries to mimic how radiologists read medical images by labelling them, while the other part of the system judges the quality of the AI-generated labelled scans by benchmarking them against the limited labelled scans provided by radiologists,” Peiris said.
“Traditionally, radiologists and other medical experts annotate, or label, medical scans by hand, highlighting specific areas of interest, such as tumours or other lesions. These labels provide guidance or supervision for training AI models.
“This method relies on the subjective interpretation of individuals, is time-consuming, and prone to errors and extended waiting periods for patients seeking treatments.”
The availability of large-scale annotated medical image datasets is often limited, as it requires significant effort, time and expertise to annotate many images manually, the researchers said.
The algorithm is designed to allow multiple AI models to leverage the unique advantages of labelled and unlabelled data, and learn from each other’s predictions to help improve overall accuracy.
“Across the three publicly accessible medical datasets, utilising a 10% labelled data setting, we achieved an average improvement of 3% compared to the most recent state-of-the-art approach under identical conditions,” said Peiris, claiming the algorithm demonstrates remarkable performance even with limited annotations, unlike algorithms that rely on large volumes of annotated data.
“This enables AI models to make more informed decisions, validate their initial assessments, and uncover more accurate diagnoses and treatment decisions.”
The next phase of the research will focus on expanding the application to work with different types of medical images and developing a dedicated end-to-end product that radiologists can use in their practices.
The study, published in Nature Machine Intelligence, was led by A/Professor Mehrtash Harandi and conducted by principal researcher Himashi Peiris, a PhD candidate at Monash University’s Faculty of Engineering, together with A/Professor Zhaolin Chen, Dr Munawar Hayat and Professor Gary Egan, from Monash Biomedical Imaging and the Faculty of Information Technology.
The five-year strategy plan is aimed at achieving four key outcomes for digital health: digitally...
With Australia choosing HL7 FHIR as its healthcare standard, the partnership will also support...
Virtus will roll out these solutions over the next six months at its 62 fertility clinics,...