Human age at death estimation from cortical bone: An automated system

The skeleton of human changes with increasing age due to change in bone density. Bone tissues have the capability to store the information of these variations, which can be used to predict age from human remains. This research study focused on the process and factors that affect human skeleton with increasing age. The techniques to carry out human age estimation are employed, which shows that there is a strong co-relation between increasing age and bone micro structures. Microscopic analysis of haversian canal was analyzed in these techniques to estimate age.  The objectives of this study is to propose automatic human age estimation method based on image processing techniques, which will attempt to detect and analyze bone microstructure from microscopic image, and will estimate human age from the bone specimen.

This study had developed algorithms using image processing and pattern recognition techniques to estimate human age at death automatically. Since manual techniques of age estimation shows that age estimation factor in bone cells varies in different populations, hence this project considers Malaysian population bone samples, and develops a novel method that will detect age from bone microstructures automatically.

The microscopic examination of bone tissue to estimate age is based upon age-associated changes in histomorphological features that occur due to the life-long metabolic process in bone called remodeling. Modeling and remodeling are the two processes in human body that control the shape and structure of bones with increasing age. Previous studies showed that there is a strong co-relation of bone micro structures such as osteon number, osteon size and haversian canal with increasing age. These micro structures can be analyzed and measured using microscopic image processing techniques to estimate human age after death, which can help in many cases, especially in forensic field, reconstruction of population demographics and in individual analysis of human remains for identification purposes. However, the manual techniques often produce subjective results, and requires diligent concentration from a highly trained operator. Also manual interpretation of microscopic bone images is error prone because of statistical, structural and temporal variations of objects in a raw bone images. Hence, there is a need of introducing an automatic method to extract micro features from microscopic bone image that do not rely on the expertise and experience of the investigator.

First phase involves the study and determination of bone microstructures parameters that shows increasing or decreasing correlation with age. Second phase involve development and application of algorithms to microscopic image in order to extract bone microstructures and estimate age at death automatically. Several image processing steps e.g. Image preprocessing, structural and textural feature extraction algorithms, and classification algorithms are developed and applied to the images to get the most possible accurate results. The ability to digitize the human age estimation methods displayed several benefits over manual methods e.g. automatic digital age estimation methods can reduce the introduction of subjective bias or human error. It is easier to gather large number of data, and increases the accuracy of the measurements of extracted bone microstructures. This study is hoped to assist the forensic anthropologists, scientists and law enforcement officials to solve cases of human skeletal remains for identification purposes based on age estimation, and to direct further investigation by the police, particularly in murder cases.

Source: Pathology Dept, Faculty of Medicine
More info about this research: faridah.nor@ukm.edu.my