Assessing tuberculosis in the skeleton with the use of decision tree analysis.

Diagnosis of specific infectious diseases in the skeleton is often difficult and relies on expert opinion. Statistics is not often used as a tool to assist in such diagnoses, and therefore this study aimed at employing data mining and machine learning in the form of decision tree analysis to aid in recognizing tuberculosis (TB) in skeletal remains and find patterns of skeletal involvement. The sample included 387 modern South African individuals (n = 207 individuals known to have died of TB and n = 180 as a control group) which were scored for the presence or absence of 21 skeletal lesions documented to be associated with TB. A pruned decision tree classification analysis was done to detect significant patterns and associations between variables which produced a model with a moderate classification rate based on four of the variables. As expected, vertebral changes were selected first, followed by rib, acetabular and lastly cranial changes. As a proof of concept, it was shown that machine learning was able to identify patterns of changes in TB skeletons versus a control group. However, further investigation into the use of machine learning in assessing skeletal changes associated with specific diseases is needed.

The association between skeletal lesions and tuberculosis in a pre-antibiotic South African sample.

OBJECTIVE: This study assessed the sensitivity and specificity of skeletal lesions to accurately diagnose TB in a pre-antibiotic South African skeletal sample. MATERIALS: A total of 435 skeletons of individuals who died before 1950 from the Raymond A. Dart Collection of Human Skeletons. 176 died of TB, 109 died of other pulmonary diseases, and 150 died of other causes. METHODS: The presence / absence of 23 skeletal lesions were assessed for differences in frequency between groups. Sensitivities and specificities were calculated and compared to Dangvard Pedersen et al. (2019). RESULTS: Lesions on the ventral surface of thoracic and lumbar vertebral bodies were observed significantly more often in TB and pulmonary cases than in other cause of death group and yielded a 55% probability of a true TB diagnosis, if observed. An association between skeletal lesions and TB was found for rib and vertebral lesions. CONCLUSIONS: The results suggest that even when not documented to have died of TB, TB-related changes are observed in many individuals in a South African skeletal sample, indicating that they may have been infected with the disease. SIGNIFICANCE: The study provides information that can assist palaeopathologists in making inferences about the prevalence of TB in past populations. LIMITATIONS: Sample sizes were small, and the inclusion of a pulmonary disease group may have confounded the results. SUGGESTIONS FOR FURTHER RESEARCH: The selection of a control group without any possible contact with TB may improve the results and should be investigated.