Age estimation using medial clavicle by histomorphometry method with artificial intelligence: A review

Abstract

This review research critically assesses the evolving landscape of age estimation methodologies, with a particular focus on the innovative integration of histomorphometry and artificial intelligence (AI) in the analysis of the medial clavicle. The medial clavicle emerges as a crucial skeletal feature for predicting age, offering valuable insights into the morphological changes occurring throughout an individual's lifespan. Through an in-depth exploration of histological complexities, including variations in osteons, trabecular structures, and cortical thickness, this review elucidates their utility as viable indicators for age-related evaluations. This framework is augmented by the incorporation of AI technology, which enables automatic picture identification, feature extraction, and complicated pattern analysis. Our review of previous research highlights the promise of AI in improving prediction models for nuanced age estimates, highlighting the importance of large-scale, diversified datasets and thorough cross-validation. This thorough study, which addresses ethical concerns as well as the influence of population-specific characteristics, moves the debate around age estimate ahead, presenting insights with consequences for forensic anthropology, clinical diagnoses, and future research avenues.

Keywords

Medial clavicle age estimation histomorphometry artificial intelligence

Introduction

In forensic anthropology, age estimation from skeletal remains is crucial for medicolegal investigations, identification, and reconstruction of death-related contexts.¹ Age estimation methods include biological and anthropological approaches. Biological methods analyze dental development, skeletal growth, and cranial suture closure, aiding forensic and anthropological assessments.²

Anthropological methods such as stature estimation, biomechanical analysis, and morphological examinations are used in forensic, archaeological, and medical contexts, benefiting from technological advancements. Diverse methods are integrated for precise age estimation, highlighting the field's interdisciplinary nature.^1,2

The search for accurate techniques for estimating age is an ongoing challenge in the field of anthropological forensics.^2,3 Methods for evaluating mature skeletons differ from those for estimating age at death in fetuses, infants, and young children.⁴ For skeletal age estimation in adults, forensic anthropologists often examine the pubic symphysis, sternal rib end, the auricular surface of the ilium, and cranial sutures.

For adult age estimation, additional skeletal criteria such as the proximal femur, acetabular surface, and clavicle are often used alongside standard techniques.^5,6 Skull analysis offers rich developmental features and techniques for accurate age estimation, but accuracy decreases with age and varies by observer and population.^4,6 Although pubis symphysis shows distinct morphological changes that facilitate accurate estimation, older age groups are likelier to experience subjectivity and inaccuracies.^7,8 Ossification centers give consistent age signals in young people, but there are timing discrepancies and errors after adolescence.^4,9 Age-related changes are well-defined at the sternal end of the fourth rib, but anomalies and interpretation difficulties may compromise accuracy.^10,11 Observable age changes are provided by the auricular surface of the sacroiliac joint, particularly in older adults; however, limitations include population variations and subjective interpretation.^12,13 Compared to other bones, the medial end of the clavicle, with its specific ossification and fusion changes, aids accurate age estimation due to its preservation in skeletal remains, providing insights into age progression in forensic and archaeological contexts.^14,15

Skeletal examinations by forensic anthropologists and osteologists are aided by guidelines for age estimation based on medial clavicle shape and fusion that have been developed via extensive research.^16,17 Research validates the precision and dependability of age assessment using the medial clavicle, emphasizing its importance in forensic and archeological inquiries.^14,15 This review explores age estimation using the medial clavicle, integrating histomorphometry and artificial intelligence (AI) techniques. The medial clavicle reveals its significance in understanding age-related changes.^18,19 As individuals traverse the journey of life, the medial clavicle undergoes dynamic changes in curvature, length, and microstructure.²⁰

Histomorphometry, a method rooted in the microscopic analysis of bone tissue, becomes a powerful lens through which age-related alterations are scrutinized.^19,21 From osteon density variations to trabecular patterns, the histological intricacies of the medial clavicle offer a nuanced narrative of aging.^22,23 Histomorphometry combined with AI advances age estimation, leveraging algorithms trained to automate and enhance precision in analyzing histological patterns.^24–26 This review emphasizes pivotal studies merging histomorphometry and AI, highlighting standardized methodologies and ethical considerations, alongside population-specific variations and external factors affecting age estimation accuracy.^2,18,27 We navigate through the historical evolution of age estimation methodologies, shedding light on the advancements and challenges encountered along with the way.

We envision a future where AI augments histomorphometry for faster, more reliable age estimation, emphasizing the need for multidisciplinary collaboration and awareness of potential pitfalls.^22,27 This review guides researchers in age estimation using the medial clavicle, highlighting histomorphometry and AI as transformative approaches with significant forensic implications.

Anatomy and development of medial clavicle

The medial end refers to the sternal end of the clavicle. It connects to the sternum's manubrium at the sternoclavicular joint.²⁸ The clavicle shaft is long and curved. The middle two-thirds of the body is quite straight.²⁹ The conoid tubercle is a tiny protrusion on the inferior surface. The sternal notch is a depression on the superior side of the manubrium of the sternum where the medial end of the clavicle articulates.^30,31 Several muscles connect to the clavicle, including the sternocleidomastoid.³² The clavicle undergoes intramembranous ossification, with major ossification foci located in the shaft and medial end.³³ During development, these centers merge. This process begins during the 5th to 6th week of fetal development in the case of the clavicle. Ossification centers are areas where ossification begins.³⁴ The medial end of the clavicle may have some ossification variability, resulting in anatomical variances. Congenital changes in clavicle morphology, such as variations in length or curvature, may occur in certain individuals.^34,35 The medial clavicle's shape is tailored to its intended uses, which include contributing to a variety of shoulder motions, supporting the shoulder girdle, and transferring stresses from the upper limb to the axial skeleton.^29,30,32

Analyzing the ossification processes, fusion timeline, and skeletal growth patterns in detail is necessary to comprehend the anatomy and development of the medial clavicle for age estimation.^18,19

Anatomy

Articular Facet: The sternoclavicular joint is formed by the interaction between the unique articular facet of the medial clavicle and the clavicular notch of the manubrium. Over time, wear-related changes to this articulation provide insights into age-related degenerative processes (Figure 1(a)).^28,31

Figure 1.

Medial view of right clavicle: An articular facet is shown in this figure (a). Superior view of right clavicle: The arrow points to the region of the clavicle's sternal end and demonstrates the clavicle's S-shaped curvature (b).

Sternal End: Age estimation is especially interested in the sternal end of the clavicle, where fusion takes place. Examined are the shape and degree of fusion of this area since structural alterations reflect patterns of development and ossification associated with aging (Figure 1(b)).^29,31

S-shaped Curvature: There are individual differences in the S-shaped curvature of the medial clavicle. This curvature can be examined for age-related alterations, such as remodeling patterns or deformities linked to osteoporosis, and it represents the biomechanical stresses made on the clavicle (Figure 1(b)).³¹

Development

Ossification Centers: At its sternal end, an ossification center gives rise to the medial clavicle. Estimating age requires an understanding of the appearance, evolution, and fusion timing of this center. Different groups’ and individuals’ ossification patterns are needed for in-depth investigation.^33–35

Fusion Process: The process of endochondral ossification results in the fusion of the main shaft and the ossification center of the medial clavicle. With a gradual shift from cartilaginous to osseous tissue during adolescence and the early stages of adulthood, the timing and completion of fusion serve as important age markers.^34,35

Epiphyseal Union Lines: On the medial clavicle following fusion, a residual line or scar known as the epiphyseal union line may endure. Clarity, ossification level, and the existence of aging-related secondary alterations can all be assessed in relation to this line, which represents the site of fusion.^33,35

Age estimation via the medial clavicle integrates anatomical observations, developmental patterns, and statistical analyses, leveraging reference samples and standards. Research aims to elucidate biological mechanisms of skeletal growth for enhanced accuracy, utilizing advanced imaging, histology, and computational modeling.

Histology of medial clavicle

The clavicle is a long bone composed of dense, compact bone tissue. Compact bone comprises the clavicle's outer layer, giving it strength and stability.³⁶ The periosteum is a fibrous membrane that covers the outside of the clavicle. Blood vessels and nerves can be found in the periosteum, which aids in the bone's innervation and nutrition.³⁷ The clavicle's exterior layer is made up of compact bone, which gives it strength and protection. It is made up of densely clustered osteons, which are cylindrical structures with concentric layers of bone matrix inside of them.³⁸ The trabecular bone is located on the inside of the clavicle. The bone struts known as trabeculae aid in stress resistance and structural stability. Mature bone cells included in the bone matrix are called osteocytes.^36,39 They contribute to bone remodeling and preserving bone tissue. During the process of bone development, osteoblasts oversee the synthesis and secretion of bone matrix.^36,39,40 During remodeling, osteoclasts tear down bone tissue as part of the process known as bone resorption.^36,37,40 Hyaline cartilage may be present on the articular surfaces of the clavicle, especially the sternal end. At the sternoclavicular joint, this cartilage facilitates easy articulation of the sternum.^36,40

Compact bone and trabecular bone of the medial clavicle

The Haversian system or osteon is the fundamental functional unit of compact bone (Figure 2).²¹ Cylindrical structures called osteons are arranged along with the long axis of the bone. The Haversian Canal contains blood arteries, nerves, and lymphatics within the osteon.⁴¹ Concentric lamellae encircle the Haversian canal in layers of bone matrix. The mineralized lamellae sustain the structure. Between undamaged osteons are found interstitial lamellae fill in the crevices between osteons to strengthen the bone.^42,43 They are remains of earlier osteons that have largely resorbed during bone remodeling.^44,45 Circumferential Lamellae that are located on the compact bone's exterior and interior surfaces. Lacunae are tiny gaps in the bone matrix that contain mature bone cells known as osteocytes.^21,45,46 Canaliculi are tiny channels that link lacunae. The periosteum protects the compact bone's exterior surface.^46,47 Blood arteries, nerves, and osteoprogenitor cells are found in the dense connective tissue layer known as the periosteum. Compact bone has the strength to withstand mechanical stress and is dense.^41–43 Compact bone can offer structural support and respond to mechanical demands according to the arrangement of its osteons and lamellae.²¹ Osteoblasts are the cells that generate new bone matrix during the remodeling process of bone. Osteoclasts play a role in bone resorption, which is the breakdown and removal of aging or diseased bone tissue.^46,47

Figure 2.

Cross-section of compact bone representing the Haversian system or osteon is the fundamental functional unit of compact bone (the area enclosed dash circle is one of the various Haversian systems), the arrow points out osteocytes are in lacunae, and HC stands for Haversian canal.

Trabecular bones are interconnecting bony struts or plates that are thin. Create a lattice-like network for structural support. Bone marrow fills the gaps between trabeculae. Bone health is maintained via the continual rebuilding process.^43,47 Osteoclasts resorb bone, whereas osteoblasts build new bone matrix.²¹ Trabecular bone adjusts to mechanical loading changes. Stress can cause remodeling to improve strength and support.^21,48 Changes in trabecular bone density and structure are important in diseases such as osteoporosis.^38,48 Reduced trabecular bone density may increase the risk of fracture. Trabecular and compact bone work together to give the clavicle strength, flexibility, and shock absorption.⁴⁸

Age-related histological alterations of medial clavicle

The medial clavicle represents an important skeletal component for age estimation considering its patterns of growth and development across an individual's life.¹⁸ Several properties of the medial clavicle have been shown to vary with aging.¹⁹ The medial clavicle is related to the estimated age.²⁰ Over time, the gross morphology and shape of the medial clavicle might alter. The clavicle may change in curvature, length, and overall morphology, offering further clues for age estimation.^49,50 The thickness of the cortical bone in the medial clavicle can be measured to indicate age-related changes and cortical thinning or variations across the cortical structure of bone may be correlated with aging.⁵¹

The number and degree of ossification centers in the medial clavicle can be used to provide information on the development of the skeleton and the fusion and appearance of these centers can be used to estimate an individual's age.⁵² The medial end of the clavicle features an epiphyseal plate that undergoes fusion as an individual age. The timing of this fusion might indicate various age groups and adolescents normally have unfused epiphyseal plates, but aged individuals display varied degrees of fusion.^53,54 Sexual dimorphisms in the medial clavicle are also examined because of variations between males and females based on size, shape, and patterns of development.^51,55,56

The histomorphometry method of the medial clavicle can show aging-related changes in bone microstructure and bone density, trabecular patterns, and the presence of characteristics such as osteons or resorption lines can all be indicators of age.^33,51 When estimating the age of using bone, it is critical to consider population-specific variances as well as the possible impacts of factors such as health and diet.¹ Combining histomorphometry, bone microstructure changes, epiphyseal fusion, and clavicle morphology enhances age estimation accuracy. Validation across diverse populations is crucial to ensure reliability in medial clavicle age estimation methods.^51,53,54 Furthermore, research on age estimation at death may establish a way to estimate age by utilizing technology to create a new approach such as AI that is simple to use, accurate, and produces rapid estimates.^57,58 Previous study has shown that AI may be applied to biological identification in forensic anthropology, with excellent accuracy.^57–62

The histology of the medial clavicle is used to estimate age by examining the microscopic structure of the bone tissue and evaluating different characteristics that vary with age.⁵¹ Osteons are compact bone tissue's structural components. The osteon density gradually increases with age. Changes associated with aging may be shown by the number of osteons per unit area.^22,23,51 As people age, cortical bone thickness often decreases. This results in a thinner cortex in part from the resorption of bone tissue.^63–65 Age can be inferred from alterations in the microarchitecture of the trabecular bone. Changes in trabecular connection and spacing, for example, might be seen. Aging may be linked to increased cortical porosity. This is a reference to the cortical bone's pores or holes, which may compromise the structural integrity of the bone.^66,67 Bone remodeling may be indicated by characteristics including reversal lines, cement lines, and Howship's lacunae.^21,43 As people age, these traits might change in frequency and distribution. Bone tissue may eventually develop microdamage, such as microcracks. It is possible to determine the existence and degree of microdamage histologically.^63,66,68 Osteocytes contribute to bone maintenance by being embedded in the bone matrix. Aging may be linked to changes in osteocyte shape and density.^66,68 It is significant to remember that a variety of variables, such as environmental and genetic effects, can change bone histology.⁶⁹ Furthermore, age estimation accuracy is higher in groups with established reference standards, despite individual variability. Ongoing research may lead to new approaches or improvements in current methods, contingent on factors such as sample quality, preservation, and analyst expertise.^69–71

Age estimation from clavicle by histomorphometry method

Bone remodeling

Bone remodeling shapes adult skeleton histomorphology through continuous turnover, driven by osteoclasts and osteoblasts in a coordinated process known as the basic multicellular unit.^21,72,73 Activation, resorption, and formation are the three discrete phases that may be helpfully used to explain the process. The osteon or Haversian system in cortical bone, or hemi-osteon in cancellous and endosteal bone, is referred to as a basic structural unit micromorphologically.^22,37 Human histological aging may be estimated because lifelong remodeling results in distinct, measurable, and quantifiable microscopic characteristics like osteons.^72,74 All animals experience bone remodeling: however, due to species variations in non-age-related aspects including locomotion, life duration, and endocrine functioning, the formation of osteons in cortical bone is not universal and when it does occur, it frequently has a very weak age correlation.^74,75 All bone envelopes, including the trabeculae and endosteum, undergo bone remodeling. The secondary osteon in cortical bone provides the easiest histomorphology evidence for remodeling.^75,76 Remodeling is linked to a sinus, also called a bone remodeling compartment, that is bordered on its osseous side by the bone surface and on its marrow side by flattened cells in the case of cancellous and endosteal bone surfaces.⁷⁶ In cortical bone, the osteon, a basic multicellular unit, is distinctly visible in cross-sections, forming the basis for most histomorphometry techniques in anthropological studies.^75–77

Bone histomorphometry

Histomorphometry is extensively used by researchers to analyze bones from ancient and modern times.^75,78 Histomorphometry analyzes microstructures and properties of skeletal tissue, providing unique quantitative data such as microarchitecture and bone remodeling not available through other methods.^21–23,79 Anthropologists have made extensive use of age-at-death estimations and the identification of humans from nonhuman bones.⁸⁰ The amount of unremodeled lamellar bone, the mean osteon size, the observation of age-dependent changes in the bone microstructure, and most often the indication of bone remodeling activity serves as the basis for histological age assessment.⁸¹ The quantification of bone turnover, microarchitecture, and static and dynamic cell activity is performed by measuring and counting structures to characterize changes in bone histomorphology.⁸² Anthropology, as a discipline, is a relative in the quantification of bone structure and organization.⁸³ It is critical to review the methods and follow the required magnification, sample area, and other specified processes as given for each method to guarantee the proper application of any method employing histomorphometry techniques.^21,22 Evaluating histological structures requires precise microscope adjustments for optimal resolution using both polarized and nonpolarized light. Although automated methods can save time, direct microscope viewing remains essential due to current technological limitations.^21–23,83

Estimating age from the clavicle using the histomorphometry method: A brief history

Estimating age from skeletal remains is an important component of forensic anthropology, and the clavicle is one of the bones that are frequently investigated for this reason.²¹ Histomorphometry, or the quantitative measurement of bone microstructure, has been used to analyze numerous skeletal features for age estimates. While histomorphometry can give vital information, it is only one of numerous approaches used in forensic anthropology, and multiple parameters are examined when estimating age from skeletal remains.^22,23,27 Gross morphological traits were the focus of early forensic anthropological investigations to estimate age.^84,85 With the development of technology, scientists started investigating microscopic methods to gain a deeper comprehension of the microstructure of bones.^24,25,57,58 Under a microscope, histomorphometry is the quantitative examination of bone tissue to determine factors including porosity, bone density, and structural characteristics.^22,27 Histomorphometry became well-known as a method for examining the age-related changes in bone microstructure in the late twentieth century.²¹ Because of its susceptibility to age-related changes and relatively excellent preservation in skeletal remains, the clavicle has been recognized as a potential bone for age estimate.^51,52,63 To determine which clavicle histomorphometry features are correlated with age, researchers performed investigations. In these investigations, age-related alterations in clavicular microstructure were frequently determined by examining samples from individuals with known ages.^21,52,63 Technological developments in imaging,^57,58 including high-resolution micro-CT scanning, have given researchers access to more accurate representations of the microstructure of bones,^18,53 facilitating histomorphometry study.⁵¹ Furthermore, the use of histomorphometry to estimate age from the clavicle necessitates improving the statistical models that are employed as well as verifying the technique on a variety of populations.^22,27

In their 1992 study, Stout and Paine concentrated on creating an age prediction approach for histological age assessment utilizing clavicles and ribs. Forty individuals with known demographics had to have their bones sampled for the study. Histomorphometry measures, including complete and fragmented osteon densities, were made with special emphasis on the clavicle. The clavicle formula yielded reasonably accurate age estimations. The study highlights the clavicle's distinctive contribution to age prediction, especially when paired with rib data. The combination of rib and clavicle in the formula is advised for improved accuracy. The method's application to instances missing lengthy bones, as well as its promise in forensic and demographic investigations, is highlighted, providing a noninvasive technique with practical implications.⁸⁶

An age-predicting equation based on clavicle histomorphometry was the goal of Lee and Jung's 2014 investigation. Variables including mean osteon area, osteon population density (OPD), and relative cortical area (RCA) were assessed in samples taken from 46 Korean cadavers. When compared to RCA in the age estimation equation, OPD had the strongest association with age. The clavicle's distinct qualities that its intact nature, low sensitivity to lifestyle changes, and its minor impact on other anthropological assessments that highlight its appropriateness for precise age estimation. In forensic anthropology, clavicular histomorphometry is a useful method for estimating age. This research highlights its dependability and highlights its potential applicability in many circumstances involving skeletal remains.⁵¹

Sobol et al. investigate age estimation by examining cortical bone histology within the human clavicle. Analyzing 39 males and 25 females, the research identifies the shaft of the clavicle as the optimal site for sampling. Various histological parameters, including osteon characteristics, demonstrate significant age-related changes. Notably, the number of osteons with larger diameters increases with age. Univariate linear regression highlights key predictors, such as Haversian canal diameter, for accurate age estimation. The study emphasizes the efficiency of histological section preparation in determining the age of unidentified remains.⁷¹

In their 2020 study, Kranioti et al. investigated the use of clavicular histomorphometry to estimate the age of a contemporary Albanian Balkan sample. The study looks at 33 individuals whose age and reason of death are known, and it concentrates on microanatomical characteristics such as cortical area and OPD. Both intra- and interobserver errors fell within permissible bounds. The Albanian sample's age was estimated using current histology techniques inaccurately, with variations of 8–11 years from the known age. The most accurate regression formula's cross-validation revealed comparable errors. There were notable distinctions between the Albanian and European-American communities. With demographic influences, food, and health state considered, the research proposes the use of clavicular histology in age estimates. It is advised that more study and verification be done.⁶³ From past studies, there is research related to age estimation from clavicle using histomorphometry methods described in Table 1.

Table 1.

Previous research related to age estimation from clavicle using histomorphometry methods.

Authors	Part of clavicle	Numbers	Age Range (years)	Parameter	Key finding
Stout and Paine (1992)⁸⁶	Midshaft of clavicle	N = 40	13–62 (Young to elderly)	Cortical area Intact osteon and fragmentary osteon density Total visible osteon density	Regarding the clavicle indicate that when combined with rib data, the formula provides enhanced accuracy and reliability.
		Male = 32
		Female = 7
Lee and Jung (2014)⁵¹	Medial end of clavicle	N = 46	20–70 (adult to elderly)	Ratio of the relative cortical area (RCA) Osteon population density (OPD) Mean osteon area	OPD exhibits the strongest correlation with age, making it a significant parameter for age estimation. The age-predicting equation, incorporating OPD and RCA, demonstrates a high correlation (R² = 0.628), indicating its reliability for clavicular age estimation in forensic scenarios.
		Male = 28
		Female = 18
Sobol et al. (2015)⁷¹	Shaft of clavicle	N = 64	22–86 (adult to elderly)	Number of osteons, Average diameter of the Haversian canals Percent area of interstitial lamellae Average thickness of outer circumferential lamellae Percent of osteons	Univariate linear regression highlights key predictors, such as Haversian canal diameter, for accurate age estimation. The study emphasizes the efficiency of histological section preparation in determining the age of unidentified remains.
		Male = 39
		Female = 25
Kranioti et al. (2020)⁶³	Midshaft of clavicle	N = 33	12–59 (Young to elderly)	Intact secondary and fragmented osteon number OPD, OPD intact and fragment Total number of osteons RCA	The study reveals limitations in current histology methodologies and underlines the importance of additional validation and consideration of population-specific characteristics.
		Male = 21
		Female = 12

Forensic anthropologists use clavicular histomorphometry for age estimation, aiming to improve precision and reliability, accounting for population-specific differences. Estimating age is complex, involving genetic, environmental factors, and ongoing research seeks to enhance existing techniques or develop new ones in forensic anthropology.

The limitation of using the medial clavicle to estimate age by histomorphometry method

The medial clavicle can be used to determine age at death by a technique called histomorphometry, which examines bone tissue under a microscope.⁵¹ However, before bone samples can be utilized for histomorphometry analysis, they must first be decalcified, sectioned, and stained.²⁷ Certain tissue structures may be destroyed because of these operations, which might have an impact on how accurately ages are estimated.

There may be restrictions on the clavicle's size, especially in the medial region, which means that there will be a small sample size for histological examination.^22,27 The robustness of age estimation models may be limited by small sample numbers, which also raises the possibility of sampling bias.⁸⁶ Since the clavicle develops unevenly, a sample taken from one area may not be indicative of the complete bone.⁵⁴ It is important to make sure that the sample selected really represents the changes in the medial clavicle that occur with aging.⁸⁶

The interpretation of microscopic characteristics is necessary for histomorphometry analysis, and there may be interobserver variability in the detection and assessment of these features.⁵¹ This may impact the dependability of age estimations. The histomorphometry characteristics of the clavicle can be influenced by pathological disorders that impact bone metabolism, such as illnesses or nutritional deficiencies.^27,51 These circumstances could make estimating an individual's age more difficult and add more sources of variation.^1,64

Standardizing histomorphometry techniques across different laboratories and researchers may be challenging.^51,63,71,86 Variations in the staining methods, instruments, and measurement standards employed might cause discrepancies in age estimates.^51,63,71 Histomorphometry features can differ between populations and ethnic groupings.^51,63,71,86 Research relevant to a demographic is necessary since models created for one may not be immediately applicable to another. Studies on histomorphometry frequently concentrate on certain age groups, and the method's generalizability may be constrained.^71,86 Age estimations may become less accurate when extrapolated beyond the investigated range, particularly in poorly preserved or destroyed clavicles in forensic or archaeological contexts. Limited samples for histomorphometry examination further challenge accuracy.⁸⁶

In forensic contexts, ethical and legal considerations are paramount when collecting histology samples from human remains. Despite these challenges, histomorphometry remains valuable for studying age-related changes in bone structure.^63,71,86 Researchers continually refine methodologies to enhance accuracy in age assessment using medial clavicle histomorphometry, emphasizing cross-validation with other methods and ongoing research for reliability.^63,71,78

Application using AI in forensic to age estimation of medial clavicle

Artificial intelligence can improve and automate several elements of age estimation in skeletal investigations, including bone histomorphometry.^87,88

Artificial intelligence can estimate age by analyzing histology images of bones, detecting age-related traits. It recognizes patterns in bone microstructure, including the clavicle, to determine age groupings.^57,89,90 Artificial intelligence can automate feature extraction from histology images, such as osteon density and trabecular thickness, improving the efficiency and consistency of age estimates.^51,58,91 Artificial intelligence algorithms can identify complex patterns in large histomorphometry datasets, revealing subtle age-related changes in bone that may be difficult for human observers to detect.^51,57,92 Artificial intelligence can integrate histomorphometry, clinical, and demographic data to improve the accuracy of age estimation techniques. This comprehensive approach enhances overall age prediction.^86,91,92 Artificial intelligence uses regression analysis and machine learning on histomorphometry data to create prediction models, offering precise and individualized age estimates.^51,91,92 Artificial intelligence can enhance quality control in age estimation by detecting outliers in histology images and standardizing measurements across different populations and studies.^86,87,92 Artificial intelligence models adapt to population-specific bone histomorphometry differences by training on diverse datasets, crucial for generalizing age estimation across ethnic groups and regions. Effective AI application requires robust training data, validation studies, and ethical considerations regarding privacy.^{51,58,86,87,92} Furthermore, AI needs to be viewed as an adjunct to human knowledge, improving the effectiveness and precision of age estimate procedures in conjunction with subject matter specialists.^87,88

The method of estimating the age of the medial clavicle may be made more accurate, reliable, and efficient by utilizing AI such as Automated Analysis of Images: Histological images or radiography of the medial clavicle can be analyzed by AI algorithms that have been trained and specific characteristics associated with aging-related changes, such as epiphyseal fusion, may be automatically detected and measured.^{51,58,86,87,92} Extracting Features: Relevant information on age estimation, such as the degree of cortical thickness, the existence of certain bone structures, or the degree of epiphyseal fusion, may be automatically extracted from images using AI and the accuracy and efficiency of AI can be beneficial for feature extraction, which can be difficult to accomplish manually.^86,87,92 Recognition of Patterns: Complex patterns within the anatomy of the medial clavicle linked with distinct age groups can be recognized by machine learning approaches. This provides for a more detailed understanding of age-related changes that may be difficult to detect using standard approaches.^51,58,92 Development of Predictive Models: AI models, including regression models or deep learning networks, can be developed to predict age based on features extracted from medial clavicle images. These models can consider multiple variables simultaneously, improving the accuracy of age estimation.^86,88,92 Population-Specific Adaptation: By training on varied datasets, AI systems can adjust to population-specific changes in medial clavicle shape. This versatility is critical for generalizing age estimation models across ethnic groups and geographical areas.^51,87,88 Integration with combining clinical data: To improve the overall accuracy of age prediction models for the medial clavicle, AI may include information from clinical data, such as medical history or demographic data.^51,88,92 Standardization and Control of Quality: AI can help with quality control by spotting abnormalities or discrepancies in medial clavicle images. Artificial intelligence algorithms can help standardize measurements among various populations and investigations.^51,58,88,89 In addition to the abovementioned, AI can analyze medial clavicle images in real time for rapid and accurate age estimation in clinical and forensic settings. Collaboration among anthropology, histology, and machine learning specialists, along with access to high-quality datasets, is crucial for successful AI application. Ethical considerations and transparency in AI model development and deployment are also essential.

Artificial intelligence–powered histomorphometry of medial clavicle samples enhances age estimation accuracy in forensic anthropology and medical diagnostics, particularly beneficial when skeletal remains are incomplete or deteriorated.^18,93 Artificial intelligence learns age-related bone changes by training on known-age data, enhancing forensic age prediction via medial clavicle histomorphometry. In clinical settings, this method aids in diagnosing disorders, assessing skeletal maturation, and monitoring age-related bone changes.^18,93,94 Artificial intelligence algorithms analyzing medial clavicle histomorphometry in pediatric patients can estimate bone age, aid anomaly detection, and study skeletal biology across populations, detecting nuanced bone structure variations linked to age, environment, and ethnicity.^94,95 Artificial intelligence–based medial clavicle histomorphometry enhances understanding of skeletal variation. It provides unbiased age estimation in legal contexts, improving reproducibility and reliability in age assessments, reducing subjectivity and biases.^96,97 With standardized methods, AI in medial clavicle histomorphometry promises accurate age estimation in forensic and medical contexts, improving diagnostics and anthropology.^94,97

Artificial intelligence can potentially improve age estimations using the medial clavicle histomorphometry method.^51,97

Automated Measurement Extraction: Artificial intelligence automates histomorphometry measurements from medial clavicle images such as cortical thickness, trabecular density, and vascular canals, cutting down manual labor in forensic anthropology.^94,97

Age Regression Modeling: Artificial intelligence uses known-age datasets to correlate medial clavicle histomorphometry with age, creating regression models for age prediction from histological data.^18,98

Feature Identification and Classification: Artificial intelligence detects nuanced histomorphometry features in the medial clavicle linked to age, such as ossification patterns and epiphyseal fusion, aiding precise age classification.^18,99

Integration with Imaging Technologies: Artificial intelligence integrates with micro-CT scans and digital histology images to enhance medial clavicle analysis, improving bone morphology assessments for accurate age estimation.^19,97

Cross-Population Analysis: Artificial intelligence analyzes histomorphometry data across populations to detect specific aging patterns in the medial clavicle, enhancing model generalizability across diverse demographics.^20,100

Quality Control and Standardization: Artificial intelligence ensures measurement reliability by detecting errors in histomorphometry, enhancing age estimation consistency. It also standardizes protocols across labs, reducing variability in forensic age estimation practices.^51,97

Real-Time Decision Support: Artificial intelligence software enhances forensic age estimation with real-time decision support, aiding anthropologists in interpreting histomorphometry data instantly. Integrating AI into forensic software improves accuracy and efficiency in age determination processes.^18,101

Artificial intelligence enhances medial clavicle histomorphometry for efficient, accurate, and standardized age estimations in forensic anthropology and medicolegal investigations.^51,102

Specific application of AI for age estimation of the medial clavicle in the Thai population for instance in the research conducted by Kengkard et al., used digital radiographs for training and testing a CNN, specifically GoogLeNet. The model learned age-related patterns with data augmentation to prevent overfitting. Fine-tuning optimized the CNN's parameters for accurate age estimation from clavicular images. Validation and testing on separate datasets ensured reliability, showcasing AI's potential in forensic anthropology and human identification. This research represents a significant advancement with promising implications for age estimation efficiency in Thai forensic investigations and human identification efforts.¹⁰³

Artificial intelligence and histomorphometry based age estimation

Histomorphometry is a technique that analyzes the microscopic structure of tissues quantitatively.^51,63,86 There are several potential advantages and benefits when applying AI to estimate age from the medial clavicle, including Microscopic Precision: Compared to macroscopic techniques, the accuracy of age estimations obtained from medial clavicle examinations is improved by AI's microscopic histomorphometry analysis.^88,89 Quantitative Analysis: Histomorphometry quantitatively measures tissue features, processed efficiently by AI algorithms for consistent, unbiased analysis.^36,79 Finding subtle changes: AI detects subtle age-related changes in medial clavicle histomorphometry, revealing patterns challenging for human analysis to discern accurately.^98,100 Performance and Automation: Forensic and large-scale dataset studies require accurate age estimation, which AI accelerates and improves with the automation of histomorphometry analysis.^51,99 Observers’ Consistency: In contrast to human observers, AI improves age consistency in histomorphometry by minimizing interobserver variability through consistent application of criteria.^86,101 Objectivity and Bias Reduction: AI models have been designed to be objective, and they can aid in the reduction of any biases in the age estimate process.^51,60,101 This impartiality is critical for the scientific rigor of histomorphometry investigations.^51,101 Improved Accuracy: The integration of AI and histomorphometry can improve age estimation accuracy. Artificial intelligence systems may detect complicated patterns and correlations in histomorphometry data that traditional approaches might disregard.^{51,101,102,104} Methodology Standardization: AI in histomorphometry helps standardize age estimation procedures, ensuring consistency and comparability across research and applications through a uniform approach.^86,101,102 Pattern Recognition at an Advanced Level: AI's pattern recognition enhances identification of detailed histomorphometry patterns in the medial clavicle, refining age estimation models. This scientific progress aids in understanding age-related bone structure changes, benefiting broader bone biology studies beyond age estimation.^51,101,102 Collaboration among histomorphometry experts, AI researchers, and domain specialists is also essential for refining and developing these AI models. Analyze the usage of AI in age estimates against not utilizing AI in age estimation.^101,104

Using AI to Estimate Age from the Medial Clavicle by Histomorphometry that advantageous compared to not using AI in age assessment, when comparing the use of AI to estimate age from the medial clavicle using histomorphometry versus not utilizing AI, several criteria must be considered, including accuracy, efficiency, consistency, objectivity, and practical implications, as shown in Table 2.

Table 2.

Advantages of using AI for age estimation from medial clavicle by histomorphometry and compared with not using AI (manual analysis).

	Pros (using AI)	Cons (not using AI)
Accuracy	AI can assess complicated histomorphometry traits with accuracy, perhaps finding minor patterns linked with age-related changes more accurately than conventional approaches.	Manual evaluation may be susceptible to interobserver variability, resulting in less reliable age estimates.
Efficiency	AI offers automated and efficient processing of massive databases, decreasing the time required for histomorphometry analysis dramatically.	Manual analysis takes time and may not be practical for huge datasets or scenarios requiring speedy analyses.
Observer Consistency	AI ensures that analytic criteria are used consistently, reducing the influence of interobserver variability.	Individual perception and experience may impact manual procedures, resulting in variations in age estimations.
Objectivity and Bias Reduction	AI algorithms are created to be objective, decreasing the possibility of bias in the age estimation procedure.	Subjective factors may impact manual analysis, injecting bias into the results.
Advanced Training Pattern Recognition	AI excels at detecting detailed patterns in histomorphometry data that are not immediately evident through manual examination.	Manual approaches may rely on simpler pattern recognition, which may overlook complicated correlations in the data.

The restriction of utilizing the medial clavicle to estimate age at death using AI and histomorphometry. Certain difficulties continue when using histomorphometry of the medial clavicle with AI for age estimation, integrating issues inherent in both classical histomorphometry and AI applications. Consider. Considering the following: (1) Training Data Quality: The performance of AI models heavily relies on the quality and diversity of training data. Homogeneous datasets in terms of age, sex, population, or pathologies can limit the model's ability to generalize across different scenarios.^87,98 (2) Histomorphometry Feature Interpretability: AI's ability to learn patterns is clear, yet interpreting the biological relevance of its findings remains a challenge. For accurate age estimation, comprehending the meaning behind AI-detected traits is essential.^87,101,102 (3) Generalization of Population: Population-specific changes in histomorphometry characteristics of the medial clavicle may provide a barrier. When applied to a different group with differing bone formation patterns, an AI model trained on one population may not perform as well.^51,87,101 (4) Ethnically and Geographical Bias: If the training dataset is skewed toward a certain ethnic or geographic group, the AI model may not perform well for people from other backgrounds. It is critical to provide variety in training data to minimize age estimation biases.^86,97,101 (5) Data Preprocessing Difficulties: Preprocessing histomorphometry data for AI analysis can be a challenge. Variations in staining procedures, slice thickness, and picture quality can all have an influence on the AI model's performance. It is critical to standardize preprocessing procedures.^87,100 (6) Sample Size Restrictions: A small training dataset may limit the AI model's ability to capture diverse histomorphometry changes across ages, potentially hindering its generalization to broader populations.^51,86,102 (7) Context of clinical and forensic: In real-world applications, AI models must consider postmortem changes, preservation conditions, and pathological states, which pose challenges not fully addressed in training.^51,86,98 (8) Bias from Algorithm: AI models can exhibit bias if the training data include skewed representations of sex, age, or other demographics, potentially leading to inaccurate age estimates, especially for underrepresented groups.^99,101,104 (9) Insufficient Base Fact: In the training dataset, the absence of a precise “ground truth” for age might be a constraint. While histomorphometry traits are utilized as age proxies, they may not exactly coincide with chronological age, resulting in inconsistencies.^51,90,102 (10) Ongoing model validation: AI models necessitate continual validation on independent datasets to ensure sustained accuracy in age estimation, reflecting the dynamic nature of bone growth and aging.^99,104

Addressing these limits requires multidisciplinary collaboration among specialists in anthropology, histology, AI, and other related domains. When incorporating AI into age prediction approaches based on the medial clavicle and histomorphometry, rigorous validation, and openness in the development of the model.

Future direction and research suggestion

Future research in AI-integrated histomorphometry for medial clavicle age estimation can focus on: Selection of discriminative histomorphometry features critical for accuracy. Advanced image analysis techniques, like CNNs tailored for histology images, to detect subtle patterns. Integration of multimodal data to enhance method robustness. Creation of large-scale histology datasets for model strength and demographic generalizability. Development of interpretable AI models to explain age estimation. Rigorous cross-validation and external validation for model reliability. Real-time histomorphometry systems for rapid clinical application. Automation of histology image annotation for scalability. Collaboration among anatomists, anthropologists, AI experts, and clinicians to meet medical community standards. Emphasis on multidisciplinary collaboration, validation studies, and practical clinical application for effective AI-based histomorphometry in medial clavicle age estimation.

Conclusion

In conclusion, the synergy of histomorphometry and AI for age estimation via the medial clavicle shows great potential. Age-related changes in microstructure, trabecular patterns, and cortical thickness make the clavicle invaluable for these methods. Artificial intelligence integration automates analyses, enhances precision, and could revolutionize the field. Promising advancements in image recognition, feature extraction, and predictive modeling improve accuracy, though challenges remain, including standardization and ethical concerns. Multidisciplinary collaboration among anthropologists, histologists, and AI experts is crucial. Future efforts should refine AI models, expand datasets, and incorporate diverse demographics, aiming for rapid, reliable, and ethically sound age estimation in forensic anthropology and beyond.

Footnotes

Acknowledgements

The authors are appreciative of the Excellence of the Osteology Research and Training Center (ORTC) for their assistance, which was partly upheld by Chiang Mai University.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Faculty of Medicine, Chiang Mai University, Excellence of the Osteology Research and Training Center (ORTC), Chiang Mai University.

ORCID iD

Pasuk Mahakkanukrauh

References

Ubelaker

Khosrowshahi

. Estimation of age in forensic anthropology: historical perspective and recent methodological advances. Forensic Sci Res 2019; 4: 1–9.

Konigsberg

Herrmann

Wescott

, et al. Estimation and evidence in forensic anthropology: age-at-death. J Forensic Sci 2008; 53: 541–557.

Traithepchanapai

Mahakkanukrauh

Kranioti

. History, research and practice of forensic anthropology in Thailand. J Forensic Sci Int 2016; 261: 167.e1–6.

Priya

. Methods of skeletal age estimation used by forensic anthropologists in adults: a review. Forensic Res Criminol Int J 2017; 4: 41–51.

Guo

Olze

Ottow

, et al. Dental age estimation in living individuals using 3.0 T MRI of lower third molars. Int J Legal Med 2015; 129: 1265–1270.

Gonsior

Ramsthaler

Gehl

, et al. Morphology as a cause for different classification of the ossification stage of the medial clavicular epiphysis by ultrasound, computed tomography, and macroscopy. Int J Legal Med 2013; 127: 1013–1021.

Todd

. Age changes in the pubic bone. I. The male white pubis. Am J Phys Anthropol 1920; 3: 285–334.

Brooks

Suchey

. Skeletal age determination based on the os pubis: a comparison of the Acsádi-Nemeskéri and Suchey-Brooks methods. Hum Evol 1990; 5: 227–238.

Franklin

. Forensic age estimation in human skeletal remains: current concepts and future directions. Legal Med 2010; 12: 1–7.

10.

McCormick

Stewart

. Age related changes in the human plastron: a roentgenographic and morphologic study. J Forensic Sci 1988; 33: 100–120.

11.

Yaşar

Loth

Wright

. Age estimation from the rib by phase analysis: white males. J Forensic Sci 1984; 29: 1094–1104.

12.

Lovejoy

Meindl

Pryzbeck

, et al. Chronological metamorphosis of the auricular surface of the ilium: a new method for the determination of adult skeletal age at death. Am J Phys Anthropol 1985; 68: 15–28.

13.

Saunders

Fitzgerald

Rogers

, et al. A test of several methods of skeletal age estimation using a documented archaeological sample. Can Soc Forensic Sci 1992; 25: 97–118.

14.

Langley-Shirley

Jantz

. A Bayesian approach to age estimation in modern Americans from the clavicle. J Forensic Sci 2010; 55: 571–583.

15.

Walker

Lovejoy

. Radiographic changes in the clavicle and proximal femur and their use in the determination of skeletal age at death. Am J Phys Anthropol 1985; 68: 67–78.

16.

Todd

Lyon

. Endocranial suture closure. It’s progress and age relationship. Part I.—adult males of white stock. Am J Phys Anthropol 1924; 7: 325–384.

17.

Key

Aiello

Molleson

. Cranial suture closure and its implications for age estimation. Int J Osteoarchaeol. 1994; 4: 193–207.

18.

Buckley

Clark

. Forensic age estimation using the medial clavicular epiphysis: a study review. Radiol Technol 2017; 88: 482–498.

19.

Milenkovic

Djukic

Djonic

, et al. Skeletal age estimation based on medial clavicle—a test of the method reliability. Int J Legal Med 2013; 127: 667–676.

20.

Falys

Prangle

. Estimating age of mature adults from the degeneration of the sternal end of the clavicle. Am J Phys Anthropol 2015; 156: 203–214.

21.

Crowder

Stout

. Bone histology: An anthropological perspective. London: CRC Press, 2011.

22.

Chompoophuen

Mahakkanukrauh

Mekjaidee

. Histomorphometry based age estimation at death using femur bone: a narrative review. Int Medical J 2018; 25: 254–258.

23.

Inthasan

Mahakkanukrauh

. Age estimation from clavicle by histomorphometry method: a review. Med & Health 2017; 12: 4–17.

24.

Lui

Guo

Leung

. Accuracy of artificial intelligence on histology prediction and detection of colorectal polyps: a systematic review and meta-analysis. Gastrointest Endosc 2020; 92: 11–22.

25.

Eekelen

Litjens

Hebeda

. Artificial intelligence in bone marrow histological diagnostics: potential applications and challenges. Pathobiology 2023; 15: 1–10.

26.

Abdelsamea

Zidan

Senousy

, et al. A survey on artificial intelligence in histopathology image analysis. Wiley Interdiscip Rev Data Min Knowl Discov 2022; 12: 1–44.

27.

Gumpangseth

Mahakkanukrauh

. Age estimation in the combined long bones and ribs by histomorphometry: past, present, and future. Med Sci Law 2024; 64: 52–71.

28.

White

Folkens

. The human bone manual. Burlington: Elsevier, 2005.

29.

White

Black

Folkens

. Human osteology. 3rd ed. Burlington: Academic Press, 2011.

30.

Mahakkanukrauh

. Anatomy and forensic anthropology of the human bone. Chiang Mai: Siampimnana, 2012.

31.

Javed

Maldonado

Ashmyan

. Anatomy, shoulder and upper limb, muscles. In: StatPearls. Treasure Island, FL: StatPearls Publishing, 2024, https://www.ncbi.nlm.nih.gov/books/NBK482410/.

32.

Mahakkanukrauh

. Locomotion system. Chiang Mai: Faculty of Medicine, Chiang Mai University, 2014.

33.

Calixto

Penagos

Jaramillo

, et al. A histological study of postnatal development of clavicle articular ends. Uni Sci 2015; 20: 361–368.

34.

Moore

Persaud

Torchia

. The developing human-e-book: Clinically oriented embryology. Amsterdam, The Netherlands: Elsevier Health Sciences, 2018.

35.

Schmidt

Ottow

Pfeiffer

, et al. Magnetic resonance imaging-based evaluation of ossification of the medial clavicular epiphysis in forensic age assessment. Int J Legal Med 2017; 131: 1665–1673.

36.

Cui

Daley

Fratkin

, et al. Atlas of histology: With functional and clinical correlations. Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins, 2011.

37.

Abdullah

Jamil

MMA

Ambar

, et al. Bone histology: a key for human sex determination after death. In: Journal of physics: Conference series. Malaysia: IOP Publishing, 2018, pp.012010.

38.

Young

Woodford

O’Dowd

. Wheater’s functional histology E-Book: A text and colour atlas. Philadelphia: Elsevier Health Sciences, 2013.

39.

Gartner

. Textbook of histology e-book. Philadelphia: Elsevier Health Sciences, 2020.

40.

Mescher

. Junqueira’s basic histology text & atlas. New York: McGraw Hill, 2018.

41.

Mitchell

Peel

. Histology E-Book: An illustrated colour text. United Kingdom: Elsevier Health Sciences, 2009.

42.

Garg

Bahl

Kaul

. A textbook of histology: A colour atlas and text. New Delhi: CBS Publishers & Distributors, 1999.

43.

Dominguez

Mavroudas

. Bone histology for skeletal age-at-death estimation. In: Age estimation. London: Elsevier, 2019, pp.145–159.

44.

Frost

. Bone remodeling dynamics. Arthritis Rheumatol 1963; 7: 545–545.

45.

Andronowski

Crowder

Soto Martinez

. Recent advancements in the analysis of bone microstructure: new dimensions in forensic anthropology. Forensic Sci Res 2018; 3: 294–309.

46.

Frost

. Tetracycline-based histological analysis of bone remodeling. Calcif Tissue Res 1969; 3: 211–237.

47.

Kiebzak

. Age-related bone changes. Exp Gerontol 1991; 26: 171–187.

48.

Crane

Kato

Patel

, et al. Histovariability in human clavicular cortical bone microstructure and its mechanical implications. J Anat 2019; 235: 873–882.

49.

Mays

Steele

Ford

. Directional asymmetry in the human clavicle. Int J Osteoarchaeol 1999; 9: 18–28.

50.

Melillo

Gunz

Coqueugniot

, et al. Structural effects of variation in the human clavicle. Am J Phys Anthropol 2019; 168: 687–704.

51.

Lee

U-Y

Jung

G-U

Choi

S-G

, et al. Anthropological age estimation with bone histomorphometry from the human clavicle. Anthropologist 2014; 17: 929–936.

52.

Kreitner

Schweden

Riepert

, et al. Bone age determination based on the study of the medial extremity of the clavicle. Eur Radiol 1998; 8: 1116–1122.

53.

Pattamapaspong

Madla

Mekjaidee

, et al. Age estimation of a Thai population based on maturation of the medial clavicular epiphysis using computed tomography. Forensic Sci Int 2015; 246: 123.e1–5.

54.

Chantharawetchakun

Vachirawongsakorn

. Age estimation in the Thai male population using epiphyseal union of the medial clavicle. Chiang Mai Med J 2021; 60: 149–155.

55.

Trithepchanapai

. Age, Sex, and Stature Estimations from Clavicles by Morphology, Osteometry and Radiological Image Analysis in a Thai Population . PhD Thesis, Chiang Mai University, Thailand, 2020.

56.

Ishwarkumar

Pillay

Haffajee

, et al. Sex determination using morphometric and morphological dimensions of the clavicle within the KwaZulu-Natal population. Int J Morphol 2016; 34: 244–251.

57.

Nguyen

Hermann

Ventre

, et al. High performance for bone age estimation with an artificial intelligence solution. Diagn Interv Imaging 2023; 104: 330–336.

58.

Lee

. Automated bone age assessment using artificial intelligence: the future of bone age assessment. Korean J Radiol 2021; 22: 792–800.

59.

Bewes

Low

Morphett

, et al. Artificial intelligence for sex determination of skeletal remains: application of a deep learning artificial neural network to human skulls. J Forensic Leg Med 2019; 62: 40–43.

60.

Malatong

Intasuwan

Palee

, et al. Deep learning and morphometric approach for sex determination of the lumbar vertebrae in a Thai population. Med Sci Law 2023; 63: 14–21.

61.

Intasuwan

Taranop

Mahakkanukrauh

, et al. A comparative study of visual assessment between dry bone, 2-dimensional photograph, and deep learning methods in sex classification on the auricular area of the Os Coxae in a Thai population. Int J Morphol 2022; 40: 107–114.

62.

Pichetpan

Singsuwan

Intasuwan

, et al. Sex determination using the clavicle by deep learning in a Thai population. Med Sci Law 2024; 64: 8–14.

63.

Kranioti

Michopoulou

Tsiminikaki

, et al. Bone histomorphometry of the clavicle in a forensic sample from Albania. Forensic Sci Int 2020; 313: 110335.

64.

Burr

Ruff

Thompson

. Patterns of skeletal histologic change through time: comparison of an archaic native American population with modern populations. Anat Rec 1990; 226: 307–313.

65.

Robling

. Histomorphometry of human cortical bone: applications to age estimation, p.187–213. 2000.

66.

Carando

Portigliatti-Barbos

Ascenzi

, et al. Macroscopic shape of, and lamellar distribution within, the upper limb shafts, allowing inferences about mechanical properties. Bone 1991; 12: 265–269.

67.

McGivern

Greenwood

Márquez-Grant

, et al.

Age-related trends in the trabecular micro-architecture of the medial clavicle: is it of use in forensic science?

Front Bioeng Biotechnol 2020; 7: 1–8.

68.

Lagacé

Verna

Adalian

, et al. Testing the accuracy of a new histomorphometric method for age-at-death estimation. Forensic Sci Int 2019; 296: 48–52.

69.

García-Donas

Dyke

Paine

, et al. Accuracy and sampling error of two age estimation techniques using rib histomorphometry on a modern sample. J Forensic Leg Med 2016; 38: 28–35.

70.

García-Donas

Bonicelli

Scholl

, et al. Rib histomorphometry: a reliability and validation study with a critical review of histological techniques for forensic age estimation. Leg Med 2021; 49: 101827.

71.

Sobol

Ptaszyńska-Sarosiek

Charuta

, et al. Estimation of age at death: examination of variation in cortical bone histology within the human clavicle. Folia Morphol 2015; 74: 378–388.

72.

Khan

Jamil

Nor

. Evaluation and reliability of bone histological age estimation methods. J Fundam Appl Sci 2017; 9: 663–680.

73.

Ruddle

. An investigation of bone histology as a potential age indicator in Roe Deer. United Kingdom: University of London, University College London, 1997.

74.

Aiello

Molleson

. Are microscopic ageing techniques more accurate than macroscopic ageing techniques? J Archaeol Sci Rep 1993; 20: 689–704.

75.

Valencia-V

Esquivias

Botella

, et al. Age determination in subadults using histological analysis of bones. Salud UIS 2020; 52: 366–370.

76.

Chan

Crowder

Rogers

. Variation in cortical bone histology within the human femur and its impact on estimating age at death. Am J Phys Anthrop 2007; 132: 80–88.

77.

Pfeiffer

Crowder

Harrington

, et al. Secondary osteon and Haversian canal dimensions as behavioral indicators. Am J Phys Anthropol 2006; 131: 460–468.

78.

Rogers

. A study of histological aging of the human clavicle. 1996.

79.

Wolf

Streit

Dokládal

, et al. Determining human age at death using cremated bone microstructure. J Sci Tech Res 2017; 1: 785–791.

80.

Botha

Bhagwandin

Lynnerup

, et al. The use of stereological methods in the histomorphometric assessment of bone for age-at-death estimation. Forensic Sci Int 2018; 290: 353.e1–353.e7.

81.

Castillo

Ubelaker

Djorojevic

. Age estimation through histological study of trabecular volume and cortical bone width of the iliac crest. Sci Justice 2012; 52: 177–180.

82.

Crowder

Heinrich

Stout

. Rib histomorphometry for adult age estimation. Methods Mol Biol 2012;915: 109–127.

83.

Crowder

Pinto

Andronowski

, et al. Forensic Anthropology: Theoretical Framework and Scientific Basis. 1st ed. NJ: Wiley and Sons Ltd, 2018.

84.

Walters

Solomons

Roche

. A morphometric study of the clavicle. SA Orthop J 2010; 9: 47–52.

85.

Voisin

. Clavicle, a neglected bone: morphology and relation to arm movements and shoulder architecture in primates. The anatomical record part A: discoveries in molecular, cellular, and evolutionary biology. Am Assoc Anat 2006; 288: 944–953.

86.

Stout

Paine

. Histological age estimation using rib and clavicle. Am J Phys Anthropol 1992; 87: 111–115.

87.

Russell

Peter

Ernest

. Artificial intelligence: A modern approach. 3rd ed. Upper Saddle River: Prentice Hall, 2010.

88.

Shrestha

Mahmood

. Review of deep learning algorithms and architectures. IEEE Access 2019; 7: 53040–53065.

89.

LeCun

Bengio

Hinton

. Deep learning. Nature 2015; 521: 436–444.

90.

Cai

Gao

Zhao

. A review of the application of deep learning in medical image classification and segmentation. An Tran Med 2020; 8: 713.

91.

Litjens

Kooi

Bejnordi

, et al. A survey on deep learning in medical image analysis. Med Image Anal 2017; 42: 60–88.

92.

Ker

Wang

Rao

, et al. Deep learning applications in medical image analysis. IEEE Access 2017; 6: 9375–9389.

93.

Garvin

Passalacqua

Uhl

, et al. Developments in forensic anthropology: Age-at-death estimation. Blackwell Publishing Ltd, 2012.

94.

Alzubaidi

Zhang

Humaidi

, et al. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. J Big Data 2021; 8: 1–74.

95.

Baccino

Schmitt

. Forensic anthropology and medicine: Complementary sciences from recovery to cause of death. Humana Press Inc., 2006.

96.

Baccino

Ubelkar

Hayek

LAC

, et al. Evaluation of seven methods of estimating age at death from mature human skeletal remains. J Forensic Sci 1999; 44: 931–936.

97.

Murphy

. Machine learning: A probabilistic perspective. The MIT Press, 2012.

98.

Wang

Cidan

, et al. Performance of an artificial intelligence system for bone age assessment in Tibet. Br J Radiol 2021; 94: 20201119.

99.

Dallora

Anderberg

Kvist

, et al. Bone age assessment with various machine learning techniques: a systematic literature review and meta-analysis. PLoS One 2019; 14: e0220242.

100.

Booz

Yel

Wichmann

, et al. Artificial intelligence in bone age assessment: accuracy and efficiency of a novel fully automated algorithm compared to the Greulich-Pyle method. Eur Radiol Ex 2020; 4: 1–8.

101.

Nadeem

Goh

Ali

, et al. Bone age assessment empowered with deep learning: a survey, open research challenges and future directions. Diagnostics 2020; 10: 781.

102.

Lee

Tajmir

Lee

, et al. Fully automated deep learning system for bone age assessment. J Digit Imaging 2017; 30: 427–441.

103.

Kengkard

Choovuthayakorn

Mahakkanukrauh

, et al. Convolutional neural network of age-related trends digital radiographs of medial clavicle in a Thai population: a preliminary study. Anat Cell Biol 2023; 56: 86–93.

104.

McKern

Stewart

. Skeletal age changes in young American males: Analyzed from the standpoint of age identification. Natick, MA: Headquarters, Quartermaster Research & Development Command, 1957.