No abstract

The diagnosis of tuberculosis (TB) in osteoarcheological series relies on the identification of osseous lesions caused by the disease. The study of identified skeletal collections provides the opportunity to investigate the distribution of skeletal lesions in relation to this disease. The aim of this study was to examine the skeletal evidence for TB in late adolescent and adult individuals from the identified human collection of the Certosa cemetery of Bologna (Italy, 19th-20th c.). The sample group consists of 244 individuals (138 males, 106 females) ranging from 17 to 88 years of age. The sample was divided into three groups on the basis of the recorded cause of death: TB (N = 64), pulmonary non-TB (N = 29), and other diseases (N = 151). Skeletal lesions reported to be related to TB were analyzed. The vertebral lesions were classified into three types: enlarged foramina (EnF, vascular foramina with diameter of 3-5 mm), erosions (ER), and other foramina (OtF, cavities of various shapes > 3 mm). A CT scan analysis was also performed on vertebral bodies. Some lesions were seldom present in our sample (e.g., tuberculous arthritis). OtF (23.7%) and subperiosteal new bone formation on ribs (54.2%) are significantly more frequent in the TB group with respect to the other groups. The CT scan analysis showed that the vertebrae of individuals who have died of TB may have internal cavities in the absence of external lesions. These traits represent useful elements in the paleopathological diagnosis of TB.

In this paper we investigate the feasibility of using an SVM (support vector machine) classifier in our automatic system for the detection of clustered microcalcifications in digital mammograms. SVM is a technique for pattern recognition which relies on the statistical learning theory. It minimizes a function of two terms: the number of misclassified vectors of the training set and a term regarding the generalization classifier capability. We compare the SVM classifier with an MLP (multi-layer perceptron) in the false-positive reduction phase of our detection scheme: a detected signal is considered either microcalcification or false signal, according to the value of a set of its features. The SVM classifier gets slightly better results than the MLP one (Az value of 0.963 against 0.958) in the presence of a high number of training data; the improvement becomes much more evident (Az value of 0.952 against 0.918) in training sets of reduced size. Finally, the setting of the SVM classifier is much easier than the MLP one.

No abstract