26 August 2023 in Biomedical Signal Processing and Control
25 October 2022 in Biomedical Signal Processing and Control
28 September 2023 in Physica Medica
8 December 2023 in IEEE Sensors Letters
Semantic segmentation is an important task in which the class label of each pixel is predicted. Thus, it is quite tough compared with classification and classical segmentation. Recently, deforestation has been a serious environmental issue, as it causes numerous environmental concerns: climate change and ecological loss. Hence, it is essential to recognize deforestation to save the environment. In this work, an efficient convolution neural network (CNN) model is proposed to identify deforestation in the Amazon Rainforest more precisely. At first, two modified SegNet methods are presented to make the semantic segmentation more effective. More importantly, an efficient semantic segmentation framework is proposed by integrating the merits of ResNet18-based modified SegNet, ShuffleNet-based modified SegNet, and UNet to yield more effective segmentation. Moreover, the employment of computationally faster ResNet18 or ShuffleNet in modified SegNet leads to improvement of computational efficiency and semantic segmentation performance. Thus, the proposed framework also retains advantages such as residual learning, skip connection, pointwise group convolution, and channel shuffling, which are responsible for making the optimization easier and the network efficient and faster. In addition, a Laplacian-of-Gaussian-based modified high boosting filter (LoGMHF) is employed for deblurring, edge enhancement, and denoising. The experimental analysis also shows that the proposed framework outperforms others.
21 December 2023 in IEEE Sensors Letters
Automated and early diagnosis of blood cancer is an essential task. Currently, deep learning-based systems are widely used for medical diagnosis. However, the major drawbacks of such systems are the requirement of huge labeled databases and slow training. Thus, an efficient sparse-based deep cascade model is proposed to make the diagnosis of Acute Lymphoblastic Leukemia (ALL) and Acute Myeloid Leukemia (AML) more accurate. The major contributions are: 1) A novel sparse-based deep cascade blood cancer detection network (SBCDNet) is proposed to yield outstanding performance in small databases. 2) An efficient single-level deep cascade model is suggested for selecting more important features and making the system faster and more accurate. 3) Softmax vector coding of errors is used to achieve nonlinear transformation in layer-wise learning. 4) Modified Fast Iterative Shrinkage, and Thresholding Algorithm (MFISTA) is employed to solve inverse linear problems. Experimental results demonstrate the proposed SBCDNet outperforms its competitors with the best performances, including 98.15% and 97.50% accuracies for ALL and AML detection, respectively.
21 February 2024 in Measurement