Screening opportunistic osteoporosis through multimodal techniques of hip joint CT images: exploring 2D and 3D deep learning, radiomics, clinical data, and their integration - Takeaways - MDSpire

Screening opportunistic osteoporosis through multimodal techniques of hip joint CT images: exploring 2D and 3D deep learning, radiomics, clinical data, and their integration

  • By

  • Xiaocong Lin

  • Xiaoling Zheng

  • Shaojian Shi

  • Rongkai Shen

  • Kaibin Fang

  • July 17, 2026

Share

  • 1

    The study investigates osteoporosis detection using hip CT data, integrating deep learning, radiomics, and clinical data.

  • 2

    A total of 567 patients with hip joint CT images were enrolled to establish an opportunistic osteoporosis screening model.

  • 3

    The GradientBoosting algorithm showed the best performance among Nomogram models, achieving an accuracy of 0.849 and AUC of 0.911.

  • 4

    Densenet201 was the top-performing 2D deep learning model, with an accuracy of 0.817 and AUC of 0.884 in screening osteoporosis.

  • 5

    The 3D deep learning model ResNet34 achieved an accuracy of 0.806 and AUC of 0.889 in the validation group for osteoporosis screening.

Original Source(s)

Screening opportunistic osteoporosis through multimodal techniques of hip joint CT images: exploring 2D and 3D deep learning, radiomics, clinical data, and their integration

  • by Xiaocong Lin, Xiaoling Zheng, Shaojian Shi, Rongkai Shen, Kaibin Fang

  • July 17, 2026

Related Content

Imaging studies for predicting hematoma expansion: from traditional imaging signs to artificial intelligence-based multimodal fusion

Editorial: Glimpse to the past – the evolution of the role of Imaging in cardio therapeutics

Study on the predictive value of femoral head HU values in hip fractures