29 August 2022 in Computational Intelligence and Neuroscience
The use of an automatic histopathological image identification system is essential for expediting diagnoses and lowering mistake rates. Although it is of enormous clinical importance, computerized breast cancer multiclassification using histological pictures has rarely been investigated. A deep learning-based classification strategy is suggested to solve the challenge of automated categorization of breast cancer pathology pictures. The attention model that acts on the feature channel is the channel refinement model. The learned channel weight may be used to reduce superfluous features when implementing the feature channel. To increase classification accuracy, calibration is necessary. To increase the accuracy of channel recalibration findings, a multiscale channel recalibration model is provided, and the msSE-ResNet convolutional neural network is built. The multiscale properties flow through the network’s highest pooling layer. The channel weights obtained at different scales are delivered into line fusion and used as input to the next channel recalibration model, which may improve the results of channel recalibration. The experimental findings reveal that the spatial recalibration model fares poorly on the job of classifying breast cancer pathology pictures when applied to the semantic segmentation of brain MRI images. The public BreakHis dataset is used to conduct the experiment. The network performs benign/malignant breast pathology picture classification collected at various magnifications with a classification accuracy of 88.87 percent, according to experimental data. The diseased images are also more resilient. Experiments on pathological pictures at various magnifications show that msSE-ResNet34 is capable of performing well when used to classify pathological images at various magnifications.
21 July 2022 in BioMed Research International
An algorithm framework based on CycleGAN and an upgraded dual‐path network (DPN) is suggested to address the difficulties of uneven staining in pathological pictures and difficulty of discriminating benign from malignant cells. CycleGAN is used for color normalization in pathological pictures to tackle the problem of uneven staining. However, the resultant detection model is ineffective. By overlapping the images, the DPN uses the addition of small convolution, deconvolution, and attention mechanisms to enhance the model’s ability to classify the texture features of pathological images on the BreaKHis dataset. The parameters that are taken into consideration for measuring the accuracy of the proposed model are false‐positive rate, false‐negative rate, recall, precision, and F1 score. Several experiments are carried out over the selected parameters, such as making comparisons between benign and malignant classification accuracy under different normalization methods, comparison of accuracy of image level and patient level using different CNN models, correlating the correctness of DPN68‐A network with different deep learning models and other classification algorithms at all magnifications. The results thus obtained have proved that the proposed model DPN68‐A network can effectively classify the benign and malignant breast cancer pathological images at various magnifications. The proposed model also is able to better assist the pathologists in diagnosing the patients by synthesizing the images of different magnifications in the clinical stage.
22 June 2023 in Revista de Gestão Social e Ambiental
Objectives: The Internet of Things (IoT) framework is crucial for improving monitoring applications for smart cities and controlling municipal operations in real time. The most significant issue with applications to smart cities has been the handling of solid waste, which may have negative consequences on the health and well-being of people. Waste management has become a problem that developing and developed nations must face. The management of solid waste is a significant and exciting issue that affects habitats all around the world. Thus, it is necessary to create an efficient method to eliminate these issues or, at the very least, reduce them to a manageable level. Theoretical framework: This work proposed an Improved Particle Swarm Optimization with Deep Learning-based Municipal Solid Waste Management (IPSODL-MSWM) in smart cities. Methods: The IPSODL-MSWM approach aims to identify various types of solid waste materials and enable sustainable waste management. A Single Shot Detection (SSD) model enables efficient object detection in the IPSODL-MSWM paradigm. Then, feature vectors were generated using the MobileNetV2 model based on a deep Convolutional Neural Network (CNN). IPSO has been obtained by using a hybrid Genetic Algorithm (GA) and PSO algorithm. Results and Conclusion: The IPSODL method has been employed for automatic hyperparameter tuning since manual trial-and-error hyperparameter tuning is time-consuming. Implications of the research: The IPSODL-MSWM approach uses Support Vector Machine (SVM) for accurate municipal excess categorization in this work. This implies sustainable waste management model for better smart city development. Originality/value: With an optimal accuracy of 99.45%, many simulations show the IPSODL-MSWM model's enhanced capability for classification.
29 April 2022 in Journal of Food Quality
Medical science in recent times has witnessed the large implications of AI-based IoT approaches that made the clinical process easier than before. However, effective IoT technologies can connect as well as exchange necessary clinical data with other healthcare systems and devices conducted across the vast Internet facilities. With the help of IoT-enabled big data processing technologies, physicians can measure accurate weight, blood pressure, and daily symptoms related to spreading breast cancer cases across the globe. Utilizing IoT is essential for providing proper medical assistance, treatment, and detection at the initial stages within the healthcare environment regulated by the facilities of the Internet of Things. The implementation of IoT-based big data processes food products for supporting the detection and prevention of breast cancer. The study supports in making a critical analysis on the role of IoT in the big data mainly in cancer detection and increasing the quality of food products. The study’s main scope is to employ IoT-enabled big data processing to aid in the identification of breast cancer. However, the research is mainly focused on studying the assistance offered to healthcare professionals and others in identifying the disease effectively. The overall research study is going to investigate the role of IoT in the early detection of breast cancer symptoms. A total of 20 women were studied and certain factors have been identified which are the early symptoms of breast cancer and can potentially cause breast cancer. These include age, family history, breast density, and breast temperature (independent variables). A dependent variable has been selected: probability of breast cancer occurrence. After that, linear regression analysis has been carried out to understand how the independent variables impact the dependent variable. Findings showed that age, family history of cancer, breast density, and breast temperature are some measurable factors for breast cancer detection. The work contributes to a critical investigation of the function of IoT in big data, specifically in cancer detection and improving food product quality. Age acceleration increases the risk of breast cancer development; breast temperature increases slightly during cancer formation, and breast density has a positive impact on cancer development. Lastly, this study has provided a future scope of using IoT in cancer detection and prevention.
24 October 2017 in Cluster Computing
1 April 2023 in Indonesian Journal of Electrical Engineering and Computer Science
The Coronavirus disease has spread throughout the world and its fear has made people to be more cautious in public places. Since precautionary measures are the only reliable protocol to defend ourselves, social distancing is the only best approach to defend against the pandemic situation. The reproduction number i.e. R0 factor of COVID-19, can be slowed down only through the physical distancing norms. This research proposes a deep learning approach for maintaining the social distance by tracking and detecting the people present indoor and outdoor scenarios. Surveillance video is taken as the input and applied into you only look once (YOLO) V3 algorithm. The persons in the video are identified based on the segmentation algorithm present within the framework and then using Euclidean distance the image is evaluated. The bounding box algorithm helps to segregate the humans based on the minimum distance threshold. The proposed method is evaluated for images with peoples in the market, availing essential commodities and students entry inside a campus. Our proposed region-based convolutional neural network (RCNN) algorithm gives a better accuracy over the traditional models and hence the service can be implemented in general for places where social distancing is mandatory.