Advancing Bag-of-visual-words Representations For Lesion Classification In Retinal Images.

Diabetic Retinopathy (DR) is a complication of diabetes that can lead to blindness if not readily discovered. Automated screening algorithms have the potential to improve identification of patients who need further medical attention. However, the identification of lesions must be accurate to be useful for clinical application. The bag-of-visual-words (BoVW) algorithm employs a maximum-margin classifier in a flexible framework that is able to detect the most common DR-related lesions such as microaneurysms, cotton-wool spots and hard exudates. BoVW allows to bypass the need for pre- and post-processing of the retinographic images, as well as the need of specific ad hoc techniques for identification of each type of lesion. An extensive evaluation of the BoVW model, using three large retinograph datasets (DR1, DR2 and Messidor) with different resolution and collected by different healthcare personnel, was performed. The results demonstrate that the BoVW classification approach can identify different lesions within an image without having to utilize different algorithms for each lesion reducing processing time and providing a more flexible diagnostic system. Our BoVW scheme is based on sparse low-level feature detection with a Speeded-Up Robust Features (SURF) local descriptor, and mid-level features based on semi-soft coding with max pooling. The best BoVW representation for retinal image classification was an area under the receiver operating characteristic curve (AUC-ROC) of 97.8% (exudates) and 93.5% (red lesions), applying a cross-dataset validation protocol. To assess the accuracy for detecting cases that require referral within one year, the sparse extraction technique associated with semi-soft coding and max pooling obtained an AUC of 94.2 ± 2.0%, outperforming current methods. Those results indicate that, for retinal image classification tasks in clinical practice, BoVW is equal and, in some instances, surpasses results obtained using dense detection (widely believed to be the best choice in many vision problems) for the low-level descriptors.

Yellowing And Bleaching Of Grey Hair Caused By Photo And Thermal Degradation.

Yellowing is an undesirable phenomenon that is common in people with white and grey hair. Because white hair has no melanin, the pigment responsible for hair colour, the effects of photodegradation are more visible in this type of hair. The origin of yellowing and its relation to photodegradation processes are not properly established, and many questions remain open in this field. In this work, the photodegradation of grey hair was investigated as a function of the wavelength of incident radiation, and its ultrastructure was determined, always comparing the results obtained for the white and black fibres present in grey hair with the results of white wool. The results presented herein indicate that the photobehaviour of grey hair irradiated with a mercury lamp or with solar radiation is dependent on the wavelength range of the incident radiation and on the initial shade of yellow in the sample. Two types of grey hair were used: (1) blended grey hair (more yellow) and (2) grey hair from a single-donor (less yellow). After exposure to a full-spectrum mercury lamp for 200 h, the blended white hair turned less yellow (the yellow-blue difference, Db(*) becomes negative, Db(*)=-6), whereas the white hair from the single-donor turned slightly yellower (Db(*)=2). In contrast, VIS+IR irradiation resulted in bleaching in both types of hair, whereas a thermal treatment (at 81 °C) caused yellowing of both types of hair, resulting in a Db(*)=3 for blended white hair and Db(*)=9 for single-donor hair. The identity of the yellow chromophores was investigated by UV-Vis spectroscopy. The results obtained with this technique were contradictory, however, and it was not possible to obtain a simple correlation between the sample shade of yellow and the absorption spectra. In addition, the results are discussed in terms of the morphology differences between the pigmented and non-pigmented parts of grey hair, the yellowing and bleaching effects of grey hair, and the occurrence of dark-follow reactions.