Janus
Principal Investigator
Name
Nicolai Krekiehn
Degrees
Dipl.
Institution
Section Biomedical Imaging (SBMI), Intelligent Imaging Lab (i²Lab), Department of Radiology and Neuroradiology, University Hospital Schleswig-Holstein (UKSH), Kiel University
Position Title
Scientist
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1480
Initial CDAS Request Approval
Dec 2, 2025
Title
Janus
Summary
Osteoporosis is a prevalent condition characterized by low bone mineral density (BMD) and an increased risk of fractures, often leading to significant morbidity and mortality. Early identification of osteoporosis and osteoporotic fractures is critical for effective management and prevention of future fractures. This project aims to utilize low-dose computed tomography (LDCT) scans, originally acquired for lung cancer screening, to perform opportunistic screening for osteoporosis, vertebral fractures, and to estimate the risk of incident osteoporotic fractures.
The project leverages advanced AI tools developed by the research team at the i²Lab in the Section of Biomedical Imaging, University Medical Center Schleswig-Holstein (UKSH). These tools are tailored to operate on LDCT scans, providing:
- Automated detection of vertebral fractures.
- Accurate trabecular BMD estimation.
- Integration of baseline data with follow-up imaging to predict fracture risk.
Additionally, the project seeks to analyze datasets from NLST, such as "Cause of Death" and "LSS Non-Cancer Condition" to identify participants with osteoporosis-related diagnoses or fractures.
By integrating AI-driven diagnostic tools with a large-scale standardized dataset, this project will:
1. Establish a novel framework for opportunistic osteoporosis screening.
2. Validate LDCT as a feasible modality for fracture detection and BMD assessment.
3. Enable risk stratification for incident fractures, paving the way for targeted interventions in high-risk individuals.
The findings have the potential to transform LDCT from a cancer-specific screening tool into a versatile diagnostic resource, addressing multiple public health concerns simultaneously.
Aims
The primary goals of this project are:
Detection of Osteoporotic Vertebral Fractures:
Apply AI algorithms to automatically detect osteoporotic vertebral fractures on LDCT scans with high sensitivity and specificity, enabling large-scale opportunistic screening without additional radiation exposure.
Bone Mineral Density (BMD) Estimation:
The vertebral volumetric bone mineral density measurements obtained from approximately 20,000 participants will serve as the basis for a comprehensive epidemiological study. The aim is to characterize and analyze the distribution of vertebral bone density across different age groups, sexes, and additional demographic and clinical variables. This large-scale, population-based analysis will enable the derivation of normative reference values for vertebral BMD and the identification of potential risk factors associated with osteoporotic changes and vertebral fragility.
Risk Assessment for Incident Fractures:
Combine baseline BMD measures and prevalent fracture data to model the risk of future vertebral fractures through statistical and machine learning approaches. This integrative framework aims to improve individualized risk prediction and support targeted preventive strategies in high-risk populations.
Collaborators
Nicolai Krekiehn Section Biomedical Imaging, Dept. of Radiology and Neuroradiology, University Medical Center Schleswig-Holstein (UKSH), Campus Kiel
Timo Damm Section Biomedical Imaging, Dept. of Radiology and Neuroradiology, University Medical Center Schleswig-Holstein (UKSH), Campus Kiel