A hierarchical three-step superpixels and deep learning framework for skin lesion classification.

Eff2Net: An efficient channel attention-based convolutional neural network for skin disease classification

Skin covers the entire body and is the largest organ. Skin cancer is one of the most dreadful cancers that is primarily triggered by sensitivity to ultraviolet rays from the sun. However, the riskiest is melanoma, although it starts in a few different ways. The patient is extremely unaware of recognizing skin malignant growth at the initial stage. Literature is evident that various handcrafted and automatic deep learning features are employed to diagnose skin cancer using the traditional machine and deep learning techniques. The current research presents a comparison of skin cancer diagnosis techniques using handcrafted and non-handcrafted features. Additionally, clinical features such as Menzies method, seven-point detection, asymmetry, border color and diameter, visual textures (GRC), local binary patterns, Gabor filters, random fields of Markov, fractal dimension, and an oriental histography are also explored in the process of skin cancer detection. Several parameters, such as jacquard index, accuracy, dice efficiency, preciseness, sensitivity, and specificity, are compared on benchmark data sets to assess reported techniques. Finally, publicly available skin cancer data sets are described and the remaining issues are highlighted.

Computer vision for microscopic skin cancer diagnosis using handcrafted and non‐handcrafted features

A hierarchical three-step superpixels and deep learning framework for skin lesion classification

Skin diseases are a major and worrying problem in societies due to their physical and psychological effects on patients. Detecting skin diseases at an early stage has an important role in treatment. The process of diagnosing and treating skin injury is related to the skill and experience of the specialist doctor. The diagnostic process must be accurate and timely. Recently, artificial intelligence science has been used in the field of diagnosing skin diseases through the use of machine learning algorithms and the exploitation of the vast amount of data available in health centers and hospitals. In this paper, quite many previous studies related to methods of classification of skin diseases based on the principle of machine learning were collected. In a group of previous studies, the researchers used some systems, mechanisms, and algorithms. Several systems have been successful in classifying skin diseases and achieving varying diagnostic accuracy. Various systems have relied on methods of image processing and feature extraction that help predict and detect disease type. There are other systems designed to identify specific types of skin disease through clinical features and features obtained from tissue analyzes after a skin biopsy of the affected area. This survey shows that the diagnostic accuracy in image processing methods was relatively uneven, ranged between (50% to 100%). As for the methods of treating tissue features, the accuracy was of an excellent level of 94% or more. The results provide an overview of the actual relevant studies found in the literature and highlight most of which research gaps have emerged.

Skin Disease Classification System Based on Machine Learning Technique: A Survey

Skin disease detection and segmentation using dynamic graph cut algorithm and classification through Naive Bayes classifier

In India, tea is the most consumed and popular beverage, all over the world too. It is one of the popular and widely cultivated perennial plantation crop in our state. Their productions are heavily affected and destroyed by different diseases. Deep learning is a machine learning technique that teaches a computer with the help of pre-existing data and enables the system to do what comes naturally to humans. The main challenge faced in deep learning is that it needs a large amount of training data. In this model, we are going to propose a unique idea to detect and classify diseases in tea leaves by incorporating deep learning techniques. The critical processes in this tea leaf disease classification are health monitoring and disease detection and are essential for sustainable agriculture. Manually observing the tea leaf diseases is a tedious and time taking processthat requires skilled workers, extra time, and manpower with knowledge about tea leaf diseases. Hence, image processing models were widely utilized for these kinds of disease detections. In this proposed work, we have applied the Convolutional Neural Network(CNN) model with 1 input layer, 4 convolution layers, and 2 fully connected layers. The image is passed to the input layer. The convolution layers mainly extract features from the input image in the dataset and the output layer classifies the given image to 8 classes such as the normal leaf, Algal leaf spot, Gray blight, White spot, Brown blight, Red scab, Bud blight, and Grey blight.

Automatic Detection of Tea Leaf Diseases using Deep Convolution Neural Network

The mobile users have acquired the benefits of cloud computing with the help of Mobile Edge Computing (MEC) technology in order to satisfy the increasing data demands. The efficiency of the system is highly limited by the bandwidth limitations and limitations associated with the mobile devices despite the rapid development of MEC as well as the cloud computing technology. Our aim is to provide an optimal method to optimize the energy consumption in the mobile edge computing. In this regard, the research paper proposed a Green Cloud based Queue Management system for 5G networks that helps in addressing the issues related to latency and energy consumption. While serving the users, the proposed methodology results in less amount of energy being wasted and hence the reduced latency. By means of alleviating the congestion and implementing the virtual list, this issue can be resolved greatly. Simulation is done with the help of NS2 green cloud simulator and the results are obtained by comparing the proposed model to conventional cloud model and cloudlet based on throughput, latency, energy consumption and normalized overhead as these are the evaluation measures. The results show that there has been considerable enhancement in the energy consumption. As the throughput increases, the quality of the service also increases.

An Enhanced Green Cloud Based Queue Management (GCQM) System to Optimize Energy Consumption in Mobile Edge Computing

The Internet of Medical Things (IoMT) is an online device that senses and transmits medical data from users to physicians within a time interval. In, recent years, IoMT has rapidly grown in the medical field to provide healthcare services without physical appearance. With the use of sensors, IoMT applications are used in healthcare management. In such applications, one of the most important factors is data security, given that its transmission over the network may cause obtrusion. For data security in IoMT systems, blockchain is used due to its numerous blocks for secure data storage. In this study, Blockchain-assisted secure data management framework (BSDMF) and Proof of Activity (PoA) protocol using malicious code detection algorithm is used in the proposed data security for the healthcare system. The main aim is to enhance the data security over the networks. The PoA protocol enhances high security of data from the literature review. By replacing the malicious node from the block, the PoA can provide high security for medical data in the blockchain. Comparison with existing systems shows that the proposed simulation with BSD-Malicious code detection algorithm achieves higher accuracy ratio, precision ratio, security, and efficiency and less response time for Blockchain-enabled healthcare systems. 

Proof of Activity Protocol for IoMT Data Security

Real-time live detection of fruits and vegetables is the most important task to know the availability of the current stock of fruits and vegetables that the customers looking for in the vegetable market. For this problem, a new technique based on Deep learning and IoT is required. The proposed work has applied the YOLO model for identifying different types of vegetables and fruits available in the vegetable market and hence the customer can know the updates the livestock of fruits and vegetables in that shop. The images of commonly used fruits and vegetables in India are collected from Google and also from Kaggle. The collected images are labeled using the Roboflow framework. To implement the proposed model, YOLOv4 -tiny model is chosen which is a super-fast object detector with better accuracy and is also most suitable for embedded devices. Further, this tiny model performs fast prediction in real-time video processing scenarios. The result of the proposed system is evaluated using the metric mean Average Precision (mAP). A high-value of mAP indicates that the model performs better is in its detection. Our YOLOv4-tiny model produced an mAP value of 51% with an inference time of 18 milliseconds.

R. Rajadevi

Papers

Skin disease detection and segmentation using dynamic graph cut algorithm and classification through Naive Bayes classifier

Automatic Detection of Tea Leaf Diseases using Deep Convolution Neural Network

An Enhanced Green Cloud Based Queue Management (GCQM) System to Optimize Energy Consumption in Mobile Edge Computing

Proof of Activity Protocol for IoMT Data Security

Fruits and Vegetables Recognition using YOLO