Artificial Intelligence
AI detects cancer but it’s also reading who you are
NegativeArtificial Intelligence
Recent research reveals that AI tools designed for cancer diagnosis from tissue samples are also inferring patient demographics from pathology slides, which can lead to biased results for certain groups. This bias arises not only from the absence of diverse samples but also from the training data used to develop these AI models.
AI learns to decode the diseases written in your DNA
PositiveArtificial Intelligence
A newly developed artificial intelligence (AI) has demonstrated the ability to predict specific diseases linked to genetic mutations, marking a significant advancement in understanding genetic disorders. This breakthrough could enhance the speed of diagnoses and pave the way for personalized treatment options tailored to individual genetic profiles.
AI finds a hidden stress signal inside routine CT scans
PositiveArtificial Intelligence
Researchers have utilized a deep learning AI model to identify the first imaging-based biomarker of chronic stress, known as the Adrenal Volume Index, by analyzing adrenal gland volume in routine CT scans. This new metric shows a strong correlation with cortisol levels, allostatic load, perceived stress, and long-term cardiovascular outcomes, including the risk of heart failure.