Please use this identifier to cite or link to this item: https://idr.l2.nitk.ac.in/jspui/handle/123456789/7800
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSam, M.-
dc.contributor.authorAreeckal, A.S.-
dc.contributor.authorSumam, David S.-
dc.date.accessioned2020-03-30T10:02:49Z-
dc.date.available2020-03-30T10:02:49Z-
dc.date.issued2018-
dc.identifier.citationProceedings - 13th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2017, 2018, Vol.2018-January, , pp.173-178en_US
dc.identifier.urihttps://idr.nitk.ac.in/jspui/handle/123456789/7800-
dc.description.abstractOsteoporosis is a condition of fragile bone with an increased susceptibility to fracture. Since the gold standard method used for the diagnosis of osteoporosis, Dual X-ray Absorptiometry (DXA), is expensive and not widely available in low economies, there is a need for low cost approaches to detect bone loss in people. A new automated radiogrammetric method for early diagnosis of osteoporosis from a single hand radiograph is proposed. In this technique, the third metacarpal bone is segmented from hand X-ray images using Active Appearance Models (AAM). Points of interest acquired from the segmented bone are used to take radiogrammetric measurements, from which bone indices are calculated. Data used in this work was acquired from 138 subjects in two hospitals in India. Significant radiogrammetric features were selected using statistical analysis. The bone indices are observed to be significantly correlated with Bone Mineral Density (BMD) of the lumbar spine measured using DXA. Different classification models were trained using the significant features. The results obtained are promising and can be used as a cost effective diagnostic tool for early detection of osteoporosis. � 2017 IEEE.en_US
dc.titleEarly diagnosis of osteoporosis using active appearance model and metacarpal radiogrammetryen_US
dc.typeBook chapteren_US
Appears in Collections:2. Conference Papers

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.