RESUMO
Skin cancer, encompassing various forms such as melanoma, basal cell carcinoma, and others, remains a significant global health concern, often proving fatal if not diagnosed and treated in its early stages. The challenge of accurately diagnosing skin cancer, particularly melanoma, persists even for experienced dermatologists due to the intricate and unpredictable nature of its symptoms. To address the need for more accurate and efficient skin cancer detection, a novel Golden Hawk Optimization-based Distributed Capsule Neural Network (GHO-DCaNN) is proposed. This novel technique leverages advanced computational methods to improve the reliability and precision of skin cancer diagnosis. An optimized clustering-based segmentation approach is introduced, integrating the innovative Sewer Shad Fly Optimization (SSFO), which combines elements of both mayfly and moth flame optimization. This integration enhances the accuracy of lesion boundary delineation and feature extraction. The core of the innovation lies in the optimized distributed capsule neural network, which is trained using the Hybrid GHO. This optimizer, inspired by the behaviors of the golden eagle and fire hawk, ensures the effectiveness of epidermis lesion detection, pushing the boundaries of skin cancer diagnosis methods. The achievements based on the metrics, like specificity, sensitivity, and accuracy show 97.53%, 99.05%, and 98.83% for 90% of training and 97.83%, 99.50%, and 99.06% for k-fold of 10, respectively.
Assuntos
Ephemeroptera , Melanoma , Neoplasias Cutâneas , Animais , Melanoma/diagnóstico , Reprodutibilidade dos Testes , Neoplasias Cutâneas/diagnóstico , Neoplasias Cutâneas/patologia , Redes Neurais de Computação , Epiderme , Dermoscopia/métodosRESUMO
In India, freshwater aquatic resources are suffering from increasing human population, urbanization and shortage of all kind of natural resources like water. To mitigate this, all the major rivers have been planned for a river-interlinking through an interlinking canal system under a huge scheme; yet, the baseline information on ecological conditions of those tropical rivers and their fish communities is lacking at present. In view of that, the present study was undertaken to assess the ecological condition by comparing the trophic metrics of the fish community, conservation status and water chemistry of the two tropical rivers of the Ganga basin, from October 2007 to November 2009. The analysis of trophic niches of the available fish species indicated dominancy of carnivorous (19 species) in river Ken and omnivorous (23 species) in Betwa. The trophic level score of carnivorous species was recorded similar (33.33%) in both rivers, whereas omnivorous species were mostly found in Betwa (36.51%) than Ken (28.07%). Relatively undisturbed sites of Betwa (B1, B2 and B3) and Ken (K2, K3 and K5) were characterized by diverse fish fauna and high richness of threatened species. The higher mean trophic level scores were recorded at B4 of Betwa and K4 of Ken. The Bray-Curtis index for trophic level identified the carnivorous species (> 0.32) as an indicator species for pollution. Anthropogenic exposure, reflected in water quality as well as in fish community structure, was found higher especially in the lower stretches of both rivers. Our results suggest the importance of trophic metrics on fish community, for ecological conditions evaluation, which enables predictions on the effect of future morphodynamic changes (in the post-interlinking phases), and provide a framework and reference condition to support restoration efforts of relatively altered fish habitats in tropical rivers of India.
Assuntos
Biodiversidade , Peixes/classificação , Rios/química , Poluição da Água/análise , Animais , Espécies em Perigo de Extinção , Monitoramento Ambiental/métodos , Eutrofização , Humanos , ÍndiaRESUMO
In India, freshwater environments are experiencing serious threats to biodiversity, and there is an urgent priority for the search of alternative techniques to promote fish biodiversity conservation and management. With this aim, the present study was undertaken to assess the fish biodiversity within and outside a river protected area, and to evaluate whether the protected river area provides some benefits to riverine fish biodiversity. To assess this, the pattern of freshwater fish diversity was studied in river Gerua, along with some physicochemical conditions, from April 2000 to March 2004. For this, a comparison was made between a 15km stretch of a protected area (Katerniaghat Wildlife Sanctuary), and an unprotected one 85km downstream. In each site some physicochemical conditions were obtained, and fish were caught by normal gears and the diversity per site described. Our results showed that water temperature resulted warmest during the pre-monsoon season (25 degreeC) and low during the winter (14-15 degreeC); turbidity considerably varied by season. In the protected area, a total of 87 species belonging to eight orders, 22 families and 52 genera were collected; while a maximum of 59 species belonging to six orders, 20 families and 42 genera were recorded from the unprotected areas. Cyprinids were found to be the most dominant genera and Salmostoma bacaila was the most numerous species in the sanctuary area. Other numerous species were Eutropiichthys vacha, Notopterus notopterus, Clupisoma garua and Bagarius bagarius. The results indicated more species, greater abundances, larger individuals, and higher number of endangered fishes within the sanctuary area when compared to the unprotected area. Analysis on the mean abundance of endangered and vulnerable species for the evaluated areas in the sanctuary versus unprotected ones indicated significant differences in fish abundance (p<0.05). These results showed that this riverine protected area could be important for conservation and management of fish diversity in the region, especially for resident and threatened species.