32 Hartmann, Dinah Maria Das, Marco 532010-2 Publikationsserver der RWTH Aachen University Aachen German 103 S. : Ill., graph. Darst. Computer tomography is currently the most sensitive method for detection of intrapulmonary nodules. Accordingly, the multi-slice computed tomography (MSCT) is used as screening method for lung cancer and clarification of intrapulmonary metastases. This aspect implies, that intrapulmonary nodules can be detected by MSCT with a very high sensitivity. Due to the increasing frequency of studies with MSCT, findings of an intrapulmonary nodule are no longer a rarity, but the most common pathological abnormality of the lung. Any abnormalities should be assessed and judged on dignity, size, degree of infiltration and form. In recent years different software systems (so-called Computer-aided diagnosis (CAD)) have been developed to assist the radiologist in diagnosis and determination of the correct classification of intrapulmonary nodules. By using the CAD-Software intrapulmonary abnormalities should automatically be detected. Also CAD should support the radiologist as a second opinion. In this study the sensitivity of Computer-aided diagnosis was researched. In particular, the sensitivity has been evaluated in relation to the size of intrapulmonary nodules, which is a significant parameter for the further diagnostic procedures and selection of an appropriate therapeutic approach. The individual nodules were classified according to their size in four different size categories (< 4 mm, < 4-6 mm, < 6-8 mm and > 8 mm). Furthermore, the nodules were distinguished by their localization in central, peripheral, or pleural nodules. In comparison to other studies the CAD software, used in this study, showed a higher sensitivity for detection of intrapulmonary nodules. Especially in the smallest size category (< 4 mm) a high sensitivity of 86% was achieved. The automatic volumetry by the CAD-Software is another possibility to quantify intrapulmonary nodules. The Volumetric is an important aspect in assessing the follow-up to exclude an increase or detect a decrease of size. In the present study automatic volumetry was conducted by using the CAD at each intrapulmonary nodule. This was performed not only to evaluate the total volume and the anatomical structures, but also to get a closer look at the three-dimensional level to show the surface morphology of each nodule. Based on these surface states, characteristics of malignant nodules can already be differentiated from characteristics of benign nodules. The Computer-aided diagnosis and its different software prototypes have to aim a very high sensitivity and a minimal rate of false-positive results to ensure the use in routine diagnosis in the future. This assumes that the field of CAD will be further investigated. Also a variety of independent studies has to show a high sensitivity in detection of intrapulmonary nodules. Due to the constant progress in radiological diagnostics, not only concerning computed tomography and applied algorithms, an advancement and extension of diagnostic possibilities will be expected in the near future. Aachen, Techn. Hochsch., Diss., 2011 ; PUB:(DE-HGF)11, ; 2, ; Mehrzeilendetektorcomputertomographie Spiral-CT Sensitivität Kleinzelliges Bronchialkarzinom Nicht-kleinzelliges Bronchialkarzinom Dissertation / PhD Thesis 2011 RWTH-CONV-125314 2011 https://publications.rwth-aachen.de/record/63915