A team from the University of Hong Kong (HKU) has developed an AI-driven imaging tool, Cyto-Morphology Adversarial Distillation (CytoMAD), which helps accelerate and improve the accuracy of cancer diagnosis.
The research, led by Kevin Tsia, programme director of the Biomedical Engineering Programme at HKU Faculty of Engineering, has shown CytoMAD’s effectiveness in lung cancer diagnosis and drug testing. Tsia partnered with the Li Ka Shing Faculty of Medicine and Queen Mary Hospital to conduct the research.
Go deeper with GlobalData
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
The tool leverages microfluidic technology for ‘label-free’ imaging, allowing clinicians to examine tumour cells individually and assess metastasis risks.
Tsia said: “We use generative AI technology to render much clearer label-free images with useful information such as whether a treatment has had a positive effect.”
CytoMAD’s AI capabilities correct inconsistencies in cell imaging, improve images, and extract critical information, ensuring reliable data analysis and diagnosis.
The team’s work, which includes collaborations with HKUMed associate professor James Ho, and cardiothoracic surgeon Michael Hsin, has been published in the journal Advanced Science.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataTsia said: “This technology allows us to capture cell images at great speed. Every day tens of millions of images can be generated. Therefore, leveraging this single system, we are in a unique position, among many AI innovations, to accelerate the advanced AI R&D – from training, optimisation to deployment.”
The CytoMAD tool is said to address the ‘batch effect’ challenge, where technical variations can impede cell morphology interpretation.
Current machine learning techniques often fall short due to their reliance on a priori knowledge. However, CytoMAD’s deep-learning model, combined with ultrafast optical imaging technology, is claimed to bypass these limitations.
Its application extends beyond lung cancer, with the potential to expedite drug screening processes, benefiting from the label-free method’s time efficiency and the diagnostic power of generative AI.
