Computer-aided diagnosis in chest radiography for detection of childhood pneumonia

Texture-Based Statistical Detection and Discrimination of Some Respiratory Diseases Using Chest Radiograph

“…This study investigate the same problem but for three diseases (simultaneously). The results from this study suggest that the proposed statistical discrimination procedure can be used to detect either of PNEU, PTB, and LC when the comparison is made with normals yielding results that are comparable with similar studies (Oliveira et al 2007;Katsuragawa and Doi 2007;Arzhaeva et al 2009;Homma et al 2009). …”

“…Related studies tend to address the problem of detecting and comparing a particular disease with normals. For example, Oliveira et al (2007) studied the problem of pneumonia present and pneumonia absent using chest radiograph in detecting childhood pneumonia and van Ginneken et al (2002) studied the problem of detecting pulmonary tuberculosis from mass TB screening program.…”

“…A lot of work has been done on chest CAD algorithm for detection of lung nodules, interstitial opacities, cardiomegaly, vertebral fractures, interval changes in chest radiograph, classification of benign and malignant nodules, and the differential diagnosis of interstitial lung diseases (Schillham et al 2006;Oliveira et al 2007;van Ginneken et al 2002;Hara et al 2007).…”

“…The proposed statistical-based CAD algorithm does not depend on establishing precise boundaries. It is also not required to minimize any cost function associated with a given segmentation algorithm (Oliveira et al 2007;van Ginneken et al 2002;Hara et al 2007;Katsuragawa and Doi 2007).…”

A discrimination method for the detection of pneumonia using chest radiograph

“…The second order statistics used in this method is derived from the co-occurrence matrix and run-length matrix and the features obtained from these matrices attempts to quantify microscopic variation of density. Recently, Oliveira et al [15] has developed a computer aided diagnosis method using Haar wavelet to extract features from chest radiographs and distance-dependent weighting for image classification in detection of childhood pneumonia with high sensitivity and specificity. They have designed a prototype system, named Pneumo-CAD that is able to detect pneumonia present against pneumonia absent in chest X-ray images to aid pneumonia diagnosis in children.…”