- Scientists at the University of Southamptonhave developed anAIplatform called CenSegNet to uncover microscopic patterns withinbreast cancertumours. - The tool analysed over 330,000 centrosomes from tissue samples of 127 breast cancer patients treated at University Hospital Southampton.
- The AI identified two distinct centrosome defects that behave independently: an excess count of centrosomes and abnormally enlarged centrosomes.
- Tumours with high levels of enlarged centrosomes were found to be more aggressive, while lower levels correlated with higher patient survival rates.
- Researchers believe the platform could enable better forecasting of disease progression, new biomarkers and more personalised treatment strategies.
source & further reading
independent.co.uk — original article
OpenAI apologises for hacking Australian health data and pledges funding for cyber defences
How AI could help screen for type 2 diabetes in seconds
AI summaries can warp your memory, study warns