Acute lymphoblastic leukemia(ALL)is characterized by overgrowth of immature lymphoid cells in the bone marrow at the expense of normal hematopoiesis.One of the most prioritized tasks is the early and correct diagnosis...Acute lymphoblastic leukemia(ALL)is characterized by overgrowth of immature lymphoid cells in the bone marrow at the expense of normal hematopoiesis.One of the most prioritized tasks is the early and correct diagnosis of this malignancy;however,manual observation of the blood smear is very time-consuming and requires labor and expertise.Transfer learning in deep neural networks is of growing importance to intricate medical tasks such as medical imaging.Our work proposes an application of a novel ensemble architecture that puts together Vision Transformer and EfficientNetV2.This approach fuses deep and spatial features to optimize discriminative power by selecting features accurately,reducing redundancy,and promoting sparsity.Besides the architecture of the ensemble,the advanced feature selection is performed by the Frog-Snake Prey-Predation Relationship Optimization(FSRO)algorithm.FSRO prioritizes the most relevant features while dynamically reducing redundant and noisy data,hence improving the efficiency and accuracy of the classification model.We have compared our method for feature selection against state-of-the-art techniques and recorded an accuracy of 94.88%,a recall of 94.38%,a precision of 96.18%,and an F1-score of 95.63%.These figures are therefore better than the classical methods for deep learning.Though our dataset,collected from four different hospitals,is non-standard and heterogeneous,making the analysis more challenging,although computationally expensive,our approach proves diagnostically superior in cancer detection.Source codes and datasets are available on GitHub.展开更多
Vertebrates have particular habitat needs as a function of life cycle and reproductive stage. This paper uses four species as examples to illustrate a paradigm of environmental assessment that includes physical, biolo...Vertebrates have particular habitat needs as a function of life cycle and reproductive stage. This paper uses four species as examples to illustrate a paradigm of environmental assessment that includes physical, biological, toxicological and human dimensions. Species used include Chinook salmon (Oncorhynchus tshawytscha), northern leopard frog (Rana pipiens), northern pine snake (Pituophis m. melanoleucus), and red knot (Calidris canutus rufa, a sandpiper). The life cycles of these species include reliance on habitats that are aquatic, terrestrial, aerial, or combinations of these. Two species (frog, snake) are sedentary and two (salmon, sandpiper) are long-distance migrants. While some measurement endpoints are similar for all species (reproductive success, longevity, contaminant loads), others vary depending upon life cycle and habitat. Salmon have a restricted breeding habitat requiring coarse sand, moderate current, and high oxygen levels for adequate egg incubation. Leopard frogs require still water of appropriate temperature for development of eggs. Pine snakes require sand compaction sufficient to sustain a nest burrow without collapsing, and full sun penetration to the sand to allow their eggs in underground nests to incubate and hatch. Red knots migrate to high Arctic tundra, but incubate their own eggs, so temperature is less of a constraint, but feedinging habitat is. These habitat differences suggest the measurement endpoints that are essential to assess habitat suitability and to manage habitats to achieve stable and sustainable populations. Habitat use and population stability have implications for human activities for some, but not all species. Salmon are important economically, recreationally, and as part of Native American culture and diet. Red knots are of interest to people mainly because of their long, intercontinental migrations and declining populations. Other measurement endpoints for these four species illustrate the differences and similarities in metrics necessary to assess habitat needs. The implications of these differences are discussed.展开更多
文摘Acute lymphoblastic leukemia(ALL)is characterized by overgrowth of immature lymphoid cells in the bone marrow at the expense of normal hematopoiesis.One of the most prioritized tasks is the early and correct diagnosis of this malignancy;however,manual observation of the blood smear is very time-consuming and requires labor and expertise.Transfer learning in deep neural networks is of growing importance to intricate medical tasks such as medical imaging.Our work proposes an application of a novel ensemble architecture that puts together Vision Transformer and EfficientNetV2.This approach fuses deep and spatial features to optimize discriminative power by selecting features accurately,reducing redundancy,and promoting sparsity.Besides the architecture of the ensemble,the advanced feature selection is performed by the Frog-Snake Prey-Predation Relationship Optimization(FSRO)algorithm.FSRO prioritizes the most relevant features while dynamically reducing redundant and noisy data,hence improving the efficiency and accuracy of the classification model.We have compared our method for feature selection against state-of-the-art techniques and recorded an accuracy of 94.88%,a recall of 94.38%,a precision of 96.18%,and an F1-score of 95.63%.These figures are therefore better than the classical methods for deep learning.Though our dataset,collected from four different hospitals,is non-standard and heterogeneous,making the analysis more challenging,although computationally expensive,our approach proves diagnostically superior in cancer detection.Source codes and datasets are available on GitHub.
文摘Vertebrates have particular habitat needs as a function of life cycle and reproductive stage. This paper uses four species as examples to illustrate a paradigm of environmental assessment that includes physical, biological, toxicological and human dimensions. Species used include Chinook salmon (Oncorhynchus tshawytscha), northern leopard frog (Rana pipiens), northern pine snake (Pituophis m. melanoleucus), and red knot (Calidris canutus rufa, a sandpiper). The life cycles of these species include reliance on habitats that are aquatic, terrestrial, aerial, or combinations of these. Two species (frog, snake) are sedentary and two (salmon, sandpiper) are long-distance migrants. While some measurement endpoints are similar for all species (reproductive success, longevity, contaminant loads), others vary depending upon life cycle and habitat. Salmon have a restricted breeding habitat requiring coarse sand, moderate current, and high oxygen levels for adequate egg incubation. Leopard frogs require still water of appropriate temperature for development of eggs. Pine snakes require sand compaction sufficient to sustain a nest burrow without collapsing, and full sun penetration to the sand to allow their eggs in underground nests to incubate and hatch. Red knots migrate to high Arctic tundra, but incubate their own eggs, so temperature is less of a constraint, but feedinging habitat is. These habitat differences suggest the measurement endpoints that are essential to assess habitat suitability and to manage habitats to achieve stable and sustainable populations. Habitat use and population stability have implications for human activities for some, but not all species. Salmon are important economically, recreationally, and as part of Native American culture and diet. Red knots are of interest to people mainly because of their long, intercontinental migrations and declining populations. Other measurement endpoints for these four species illustrate the differences and similarities in metrics necessary to assess habitat needs. The implications of these differences are discussed.
