AI Algorithm Estimated Body Weight From CT Scans

Researchers developed a deep learning tool to predict patient weight, aiding in hip fracture risk assessments.

Updated on Sept. 25, 2026 in Weight Loss

Isometric editorial illustration of an abstract, geometric three-dimensional lattice structure representing a bone cross-section.
Researchers developed a two-branch convolutional neural network that estimates patient body weight from CT scans to refine osteoporotic hip fracture risk assessment. AI Illustration. Upload story photo >

Live Poll

Do you trust artificial intelligence to accurately assist in medical diagnostic assessments?

Scientists have created a two-branch convolutional neural network capable of estimating body weight from lower abdominal CT scans. This tool is designed to improve the accuracy of autonomous finite element analysis used for assessing osteoporotic hip fracture risk.

Why it matters

Body weight is a critical parameter for clinical fracture risk calculations, but it is not always available in imaging records. Automating this data retrieval could streamline the evaluation process for patients undergoing scans for osteoporosis.

In a preliminary study using a test set of 24 subjects, the deep learning algorithm achieved a Pearson correlation coefficient of 0.93 with actual body weight. All predictions were within 10 kg of the actual weight, though findings rely on a small training cohort of 100 CT scans.

The details

The algorithm employs a two-branch convolutional neural network that processes image features from axial CT slices while incorporating patient metadata. By synthesizing these inputs, the system performs pre-processing to estimate weight. This measurement supports autonomous finite element analysis, which simulates how bone tissue reacts to mechanical loads to determine fracture risk.

Timeline

  1. The findings were reported on September 25, 2026.

Health Landscape

This development integrates with existing research into finite element analysis for osteoporotic hip fracture risk assessment. It marks a shift toward automating complex biomechanical models that previously required manual data input.

This tool is a backend software innovation intended for clinical use, not a measurement device for personal weight monitoring. If you are concerned about your bone health or fracture risk, discuss the necessity of bone density screenings with your physician.

The takeaway

Automated weight estimation from standard medical imaging could soon play a larger role in assessing bone fracture risk. Always coordinate with your doctor regarding your specific clinical results and the appropriate imaging procedures for your bone health.

Further reading

For more on the latest developments in weight management technology, see the Weight Loss section.

Source note: This article includes information reported by Nature.

Live Poll

Do you trust artificial intelligence to accurately assist in medical diagnostic assessments?

AI Algorithm Estimated Body Weight From CT Scans