Related Experiment Video
Updated: Nov 11, 2025
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
Development and Validation of a Radiomics Model for Differentiating Bone Islands and Osteoblastic Bone Metastases at
Ji Hyun Hong1, Joon-Yong Jung1, Aram Jo1
1From the Department of Radiology, Kangdong Sacred Heart Hospital, Hallym University College of Medicine, Seoul, Republic of Korea (J.H.H.); Department of Radiology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 06591, Republic of Korea (J.Y.J., A.J., S.Y.L., H.P., S.E.L., S.K.); Division of Biomedical Engineering, Hankuk University of Foreign Studies, Gyeonggi-do, Republic of Korea (Y.N.); and Department of Biomedical Engineering, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, Republic of Korea (S.P.).
A CT radiomics machine learning model accurately differentiates bone islands from osteoblastic metastases. This AI tool demonstrated superior diagnostic performance compared to one radiologist, aiding treatment strategy.
Area of Science:
- Radiology
- Oncology
- Artificial Intelligence
Background:
- Accurate diagnosis of sclerotic bone lesions is crucial for determining appropriate treatment strategies.
- Differentiating benign bone islands from malignant osteoblastic bone metastases can be challenging.
Purpose of the Study:
- To assess the diagnostic performance of a computed tomography (CT) radiomics-based machine learning model.
- To differentiate between bone islands and osteoblastic bone metastases.
Main Methods:
- A retrospective study utilizing contrast-enhanced abdominal CT data from two institutions (training and external test sets).
- Radiomics features were extracted and used to build a random forest (RF) model.
- The RF model's performance was evaluated against three radiologists using sensitivity, specificity, accuracy, and AUC.
Main Results:
- The RF model achieved an AUC of 0.89 during cross-validation and 0.96 in the external test set.
- The model demonstrated high sensitivity (80%) and specificity (96%) in the test set.
- The radiomics model's AUC (0.96) was significantly higher than that of one radiologist (0.88).
Conclusions:
- A CT radiomics-based random forest model is effective for distinguishing bone islands from osteoblastic bone metastases.
- The model exhibits robust diagnostic performance, comparable to experienced radiologists.
- This AI approach can aid in the accurate diagnosis and management of bone lesions.
More Related Videos
12:23Multi-modal Imaging of Angiogenesis in a Nude Rat Model of Breast Cancer Bone Metastasis Using Magnetic Resonance Imaging, Volumetric Computed Tomography and Ultrasound
Published on: August 14, 2012
08:04Monitoring Tumor Metastases and Osteolytic Lesions with Bioluminescence and Micro CT Imaging
Published on: April 14, 2011