Deep Learning and Its Applications in Biomedicine.

Deep image mining for diabetic retinopathy screening.

Deep learning based computer-aided diagnosis systems for diabetic retinopathy: A survey.

Convolutional neural networks (CNNs) show impressive performance for image classification and detection, extending heavily to the medical image domain. Nevertheless, medical experts are skeptical in these predictions as the nonlinear multilayer structure resulting in a classification outcome is not directly graspable. Recently, approaches have been shown which help the user to understand the discriminative regions within an image which are decisive for the CNN to conclude to a certain class. Although these approaches could help to build trust in the CNNs predictions, they are only slightly shown to work with medical image data which often poses a challenge as the decision for a class relies on different lesion areas scattered around the entire image. Using the DiaretDB1 dataset, we show that on retina images different lesion areas fundamental for diabetic retinopathy are detected on an image level with high accuracy, comparable or exceeding supervised methods. On lesion level, we achieve few false positives with high sensitivity, though, the network is solely trained on image-level labels which do not include information about existing lesions. Classifying between diseased and healthy images, we achieve an AUC of 0.954 on the DiaretDB1.

Weakly-supervised localization of diabetic retinopathy lesions in retinal fundus images

Application of deep learning for retinal image analysis: A review

Rapid Internet use growth and applications of diverse military have managed researchers to develop smart systems to help applications and users achieve the facilities through the provision of required service quality in networks. Any smart technologies offer protection in interactions in dispersed locations such as, e-commerce, mobile networking, telecommunications and management of network. Furthermore, this article proposed on intelligent feature selection methods and intrusion detection (ISTID) organization in webs based on neuron-genetic algorithms, intelligent software agents, genetic algorithms, particulate swarm intelligence and neural networks, rough-set. These techniques were useful to identify and prevent network intrusion to provide Internet safety and improve service value and accuracy, performance and efficiency. Furthermore, new algorithms of intelligent rules-based attributes collection algorithm for efficient function and rules-based improved vector support computer, were proposed in this article, along with a survey into the current smart techniques for intrusion detection systems.

https://www.mdpi.com/1999-4893/14/8/224/pdf

Intelligent Network Intrusion Prevention Feature Collection and Classification Algorithms

This paper presents the retina-based disease diagnosis through deep learning-based feature extraction method. This process helps in developing automated screening system, which is capable of diagnosing retina for diseases such as age-related molecular degeneration, diabetic retinopathy, macular bunker, retinoblastoma, retinal detachment, and retinitis pigmentosa. Some of these diseases share a common characteristic, which makes the classification difficult. In order to overcome the above-mentioned problem, deep learning feature extraction and a multi-class SVM classifier are used. The main contribution of this work is the reducing the dimension of the features required to classify the retinal disease, which enhances the process of reducing the system requirement as well as good performance.

Multi-retinal disease classification by reduced deep learning features

Cryptography is a method for secure communication by hiding information with secret keys so that only authorised users can read and process it. Efficient random sequence generators provide robust cipher design for cryptographic applications; further, these sequences are used for data encryption. In this paper, the highly chaotic nature of hybrid chaos maps and neural network is combined to build a random number generator for cryptographic applications. A custom neural network with a user-defined layer transfer function is built to increase the generator’s randomness. In this work, the two-hybrid chaotic map’s control parameters and iteration value are designed as a layer transfer function to obtain high randomness. Colour image encryption is performed with the extracted sequences and deoxyribonucleic acid encoding technique. Various tests like NIST, attractor test and correlation are applied to the generator to show the degree of randomness. Simulation analysis such as keyspace, key sensitivity, statistical, differential analysis, and chosen-plaintext attack shows the encryption algorithm’s strength.

Colour image encryption based on customized neural network and DNA encoding

Life time and its quality can be enhanced with early diagnosis of osteoporosis. Osteoporosis is a state of having brittle and fragile bone which arises due to vitamin deficiency, tissue loss, hormonal changes, etc. Diagnosis of osteoporosis is very important and mandatory in current scenario. Now a days Computer Aided Diagnostic (CAD) is used to detect osteoporosis more efficiently. This osteoporosis can be efficiently detected by calculating various features like Bone mineral density (BMD), statistical features from various trabecular region such as hip, toe, elbow, etc. Analysis of trabecular bounding on digital calcaneus radiographic (DCR) images will help in efficient identification of low BMD. Bone Mineral density measurement can be achieved by various segmentation methods such as K-means, Fuzzy segmentation etc. In this paper we proposed a Regression Artificial Neural Network (ANN) classifier-based computer-aided diagnosis (CAD) system for accurate osteoporotic risk detection using digital calcaneus radiographic images. The evaluation of diagnostic capability of the proposed method is to spot areas of Low Bone Mineral Density at the calcaneus region. The proposed system has achieved the highest classification accuracy of 90%, sensitivity, specificity and positive predictive value.

Efficient and early detection of osteoporosis using trabecular region

In this Era, the signal processing system is widely used. In the digital signal processing system, adder is the basic building block of all the major circuits. Today the requirements for minimizing the delay, area, and power of adder circuit improve the efficiency of whole system which drives the technology to the next level. Even though the Carry Select Adder (CSLA) occupies more area, it is been used instead of ripple carry adder to avoid propagation delay. In other models Binary to Excess-I Converter (BEC) based Carry Select Adder was also used which uses less number of logic resources than conventional CSLA. But these CSLAs are not more efficient because it rejects one sum after the calculation. So the delay was not more effectively reduced. In order to overcome this problem the reduced logic CSLA is used. But by using Gate Diffusion Input (GDI) Technique can give less delay than this recently proposed reduced logic CSLA. The proposed technique provides low power consumption, less propagation delay. By using this GDI based CSLA the number of transistors required for the circuit also minimized. So an efficient adder design can be achieved through this technique.

R. Arunkumar

Papers

Intelligent Network Intrusion Prevention Feature Collection and Classification Algorithms

Multi-retinal disease classification by reduced deep learning features

Colour image encryption based on customized neural network and DNA encoding

Efficient and early detection of osteoporosis using trabecular region

GDI based area delay power efficient carry select adder