New imaging and signal-processing studies showcased AI progress toward opportunistic screening and earlier detection. One study reported that AI software can extract reliable bone mineral density from routine non-contrast CT scans, aligning closely with DXA for osteoporosis screening. In parallel, a deep learning radiomics approach trained on routine breast ultrasound images predicted sentinel lymph node metastasis preoperatively, while authors emphasized the need for larger prospective validation before clinical deployment. Together, the results reinforce a common theme: AI performance is trending toward clinically usable metrics when models leverage existing imaging pipelines, though validation and calibration remain the gating items for adoption.