Abstract :

Osteoporosis Is A Skeletal Condition That Is Hard To Diagnose Before Symptoms Appear. Due To Cost And Safety Concerns, Current Skeletal Disease Screening Techniques, Such Dual-energy X-ray Absorptiometry, Are Only Employed In Certain Situations After Symptoms Appear. In Terms Of Early Treatment And Cost, Early Identification Of Osteopenia And Osteoporosis Employing Alternative Methods For Comparatively Frequent Tests Is Beneficial. Deep Learning-based Osteoporosis Detection Techniques For A Variety Of Modalities Have Been Proposed In Numerous Recent Papers, With Excellent Results. However, Because These Investigations Include Laborious Procedures Like Physically Cropping A Region Of Interest Or Diagnosing Osteoporosis Rather Than Osteopenia, Their Clinical Applicability Is Limited. In This Work, We Describe A Classification Task That Uses Computed Tomography (CT) To Diagnose Osteopenia And Osteoporosis. We Also Suggest A Multi-view CT Network (MVCTNet) That Uses Two Images From The Original CT Scan To Automatically Classify Osteopenia And Osteoporosis. The MVCTNet Extracts Different Features From The Pictures Produced By Our Multi-view Settings, In Contrast To Earlier Approaches That Use A Single CT Image As Input. The MVCTNet Consists Of Three Task Layers And Two Feature Extractors. Using The Photos As Distinct Inputs, Two Feature Extractors Utilize Dissimilarity Loss To Discover Distinct Features. The Two Feature Extractors Features Are Used By The Target Layers To Learn The Target Task, Which They Then Aggregate. We Employ A Dataset With 2,883 Patients CT Images Classified As Normal, Osteopenia, And Osteoporosis For The Tests. Based On The Quantitative And Qualitative Assessments, We Also Note That The Suggested Approach Enhances The Performance Of Every Experiment

Published:

14-8-2026

Issue:

Vol. 26 No. 8 (2026)

Page Nos:

877-883

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite