Researchers have developed a deep learning ensemble capable of estimating adult sex, age, height, and weight from CT scans. The model, which combines ConvNeXt-Base, ViT-Base/16, and MaxViT-Base architectures, achieved high accuracy in sex classification and low mean absolute errors for age, height, and weight. The study, involving over 128,000 CT examinations from Japan, demonstrated that body surface area calculated from estimated values could reproduce trends observed with true measurements, suggesting potential applications in medical analysis. AI
IMPACT This research demonstrates a novel application of deep learning in medical imaging, potentially improving efficiency and accuracy in patient data estimation from CT scans.
RANK_REASON Academic paper detailing a new deep learning model and its performance on a specific task. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
