The known beetle fauna of Príncipe, São Tomé, and Annobón amounts to 403 species and subspecies, of which 190 (47%) are endemic. The most diverse families of beetles are the Cerambycidae (61 species), the Tenebrionidae (57 species), the Carabidae (45 species), the Scarabaeidae (34 species), and the Coccinellidae (31 species). Most records come from São Tomé, with 297 species. In comparison, Príncipe, with 151 recorded species, and especially Annobón, with 16 recorded species, still require extensive faunistic investigations. The families Staphylinidae and Curculionidae probably hold numerous undescribed species and should be the focus of future research. Most of the endemic species live in forests. Therefore, the continued conservation of large forest areas on the islands is key to the long-term survival of their unique beetle fauna. As elsewhere, the beetle fauna will likely suffer from the effects of climatic change, and high-altitude species are likely to be the most severely affected.

No abstract

Image Guided Radiation Therapy (IGRT) is one of the most efficient ways to treat prostate cancer. It is well‐known that Electronic Portal Images (EPI) acquired prior to the radiation therapy are of poor quality due to the low radiation dose. Several image parameters need to be manually adjusted in order to be able to visualize and localize the fiducial markers (FMs) seeds at the beginning of each day of treatment. This localization process is essential for the proper alignment of the patient. In this paper, we propose a novel technique for automatic image enhancement. There are many challenges involved in the automation of the image enhancement for FM seed detection. An automated enhancement algorithm is image and noise dependant. The level of noise in an input image and global image properties such as average contrast and intensity vary significantly between different patients and even for the same patient between different image acquisition sessions. Moreover, the small size of the targets (the FM seeds) makes it very difficult to distinguish them from noise, and hence to enhance them. Our approach addresses these challenges by using a contrast enhancement scheme based on properties of the human visual system (HVS). The rationale behind considering this approach lies in the ability of the HVS to adapt to detect objects of small size under low contrast and noise conditions. Our main contribution lies in the automatic set‐up of the parameters based upon a maxima search over a contrast metric.

No abstract