Back

Bone Densitometry Dataset for Computer Aided Osteoporosis Disease Detection

masnabadi, n.; Sadeghi-Niaraki, A.; Karimi, M.; AbuHmed, T.; Azarbani, N.; Choi, S.-M.

2025-01-28 rheumatology
10.1101/2025.01.25.24319689 medRxiv
Show abstract

Recently, automatic disease diagnosis based on medical images has become an integral part of digital pathology packages. To create, develop, evaluate, and compare these systems, we need diverse data sets. One of the key features in the diagnosis of bone diseases is measuring bone mineral density (BMD). Most research in this field uses manual methods to directly extract bone image features despite the underlying correlation between diseased and healthy bones, which explains the limited results. Detection of significant changes in bone mineral density (BMD) relies on minimally invasive dual energy x-ray absorptiometry (DXA) scanners. This article presents a collection of bone density test results along with a patient profile called Arak Bone Densitometry Center data. The patient profile includes height and weight and information about the patient, along with photos of the imaging areas. The number of these patients is 3,643, with about 4,020 photos stored next to them. Which can be used to develop automatic disease diagnosis methods and software. Datasethttps://drive.google.com/drive/folders/1HmLTG4GFgB2s4D0x7TTRx8vV_VWY3sW3?usp=sharing

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.