Developing countries have a high rate of cervical cancer due to the detection of cancer at the stage of regional spread of the disease rather than detection at the early stage. Early detection of the disease by a periodical pap smear test is the most effective way of reducing the rate of cervical cancer. This research paper shares a novel model for the automated classification of cervical cancer with weighted majority voting. Upsampling and downsampling are done to overcome the inefficiency due to an imbalance in the database. Size, statistical, intensity, and texture features are extracted from the modified fuzzy c-means segmented image. Principal Component Analysis (PCA) is done on the extracted features to rank and identify the most significant features. Supported Vector Machine (SVM), K Nearest Neighbour (KNN), and Naïve Bayes (NB) are tuned to the kernel function and a number of cross validations and are given as the input to the weighted majority voting system. The performance of the proposed method with up sampling, down sampling, and no sampling are compared, and it was found that the weighted majority voting with upsampling has the highest classification accuracy of 99%.

In medicine, early detection of lung cancer is essential for successful treatment regimens. Despite the small sample size, cancer databases often include gene expression levels as attributes. Therefore, in order to increase the convergence speed of the classification algorithm, redundant features should be eliminated. In this work, we present a unique hybrid method for cancer classification called artificial bee colony with support vector machine selection. This approach for early search of a subset of genes for cancer prognosis optimizes feature selection by combining cuckoo search algorithm (CSA) and Spider-Monkey optimization (SMO) algorithm In order with increased accuracy of artificial bee colony method, we perform Minimum Redundancy Maximum Relevance (mRMR).) method, we reduce the redundancy in gene expression data from the cancer dataset and then these selected gene subsets are efficiently classified into different cancer clusters or groups using deep learning (DL) techniques. We evaluate the effectiveness of our proposed method against eight reference microarray gene expression data sets of various cancers. Classification performance is evaluated using measures such as confusion matrix, F1-score, recall, and accuracy. Our findings show that, when applied to large gene expression datasets linked to cancer, the suggested gene selection strategy in conjunction with DL provides improved classification accuracy compared to other existing DL and machine learning methods.

Robots are programmable machines built to mimic human actions. One such action is locomotion of robots which is the recent area of research. Two-wheeled robots which diligently stabilize it may contribute to the locomotion of robots in upcoming decades. In this paper, the PID controller and Arduino are utilized to design the self-balancing robot. This paper focuses primarily on developing a controller that will aid the robot and test against several parameters such as position, balance along vertical axis and signals for controlling. The accelerometer and gyroscope sensor values are used to determine the precise position of the robot in 3 dimensional space. The sensor values are sent to the controller which controls the rotation of wheels thus aiding in balancing the robot. The two-wheeled robot turns precisely while navigating through different obstacles as against four-wheeled robots.

With the rapid development of automation technology, the manufacturing industry is facing the transformation from traditional manufacturing to intelligent manufacturing. Manufacturing product design inspection is an important process to ensure product quality. With the development of computer technology, especially the continuous improvement and application of computer vision and deep learning algorithm theory, more and more companies apply computer vision to product design inspection to improve the automation of equipment, reduce the cost of employing people, and improve the accuracy and efficiency of product design inspection. In order to meet specific design requirements, this paper designs and implements a metal product design system based on machine learning design. The system is not only able to meet the needs of manual design, but also can use mature computer vision-related algorithms to complete the initial automatic design and manual modification, thus greatly improving the design efficiency. It is worth mentioning that the system also has good scalability, which lays a good foundation for the preparation of image datasets. On this basis, the detailed design and implementation inside each module is presented, using class diagrams and data tables to show the implementation of methods in the classes. Immediately afterwards, each function of the designed system is tested, describing the expected and actual results. Finally, a summary of the completed work is presented, and the next steps related to the image design system are envisioned.

The death rate due to brain tumor is increasing rapidly. The accumulation of cells in an uncontrolled manner leads to brain tumor. The life taking tumor can be cured if it is detected in the early stage. This paper deals with analyzing the performance of the different kernels functions on SVM classification algorithm. The performance of the proposed algorithm is being measured by parameters like accuracy, sensitivity, specificity and precision. The kernels like Laplace RBF kernel, Gaussian kernel, polynomial kernel, Hyperbolic tangent kernel, and Sigmoid kernel are applied on SVM. Among the various kernel functions, Gaussian kernel SVM gave good results. Otsu segmentation is applied on the abnormal images to high light the tumor region in the brain MRI images.

Cervical cancer is having the second-highest mortality rate next to breast cancer among women in developing countries. Early detection of the abnormality is the only way to prevent morbidity. As the decision about the abnormality of the cell is made manually by the traditional Pap smear test – the clinical test conducted for the detection of cervical cancer is more prone to false-negative and false-positive cases. This paper presents a novel approach for the automatic detection of cervical cancer using modified fuzzy C-means, extracting the geometrical and texture features, Principal Component Analysis (PCA), and classification. Modified fuzzy C-means show promising results in segmenting the input image into meaningful regions even when there is uncertainty. PCA is being performed to reduce the dimensionality of the data set by maintaining only the uncorrelated features thereby reducing the processing time of the algorithm. The classification of the pap smear images into normal and abnormal cells is being done by K Nearest Neighbour (KNN) classification with k-fold cross-validation and the result obtained in the proposed method is being compared with Fine Gaussian SVM, Ensemble Bagged trees, and Linear Discriminant. The efficiency of the proposed method is measured by calculating minimum accuracy, maximum accuracy, average accuracy, sensitivity, specificity, F1-score, and precision. The experimental results of the proposed method show impressive results with minimum accuracy 94.15%, maximum accuracy 96.28%, average accuracy 94.86%, sensitivity 97.96%, specificity 83.65%, F1-score 96.87%, and precision 96.31% for threefold cross-validation.

Medical imaging provides the visual representation of internal organs of the body which facilitate in diagnosis, monitoring health etc. Previously, the automated diagnosis procedure was done using edge detection and tedious mathematical computations. With the advancement in artificial intelligence, medical imaging is now supporting the diagnosis of cancer, diabetic retinopathy, Detection of Alzheimer’s and Parkinson’s disease, brain injury etc. Diagnosis through medical imaging has reduced the mortality rate drastically especially in the field of cancer. In order to decrease the probability of human error machine learning came into existence. The commonly used machine learning algorithms are K-Nearest Neighbors, Supported Vector Machine (SVM), and Decision Trees etc. But machine learning method has its own limitation of high dependency to the features extracted which depends on many factors. In order to improve the efficiency by removing the dependency on feature deep learning method came into existence. This paper has made a detailed survey on the application of deep learning method in health care service.

Brain tumor identifications are the most common issues for recent scenario of health care community. The accurate discovery of various brain abnormalities is highly essential for treatment planning that can minimize the fatal results. Performance can be measured only through soft computing techniques. Besides being accurate, these systems must touch rapidly in order to apply themfor day-to-dayapplications. Now a day’s manycomputerizedtechniques are available for this desirableperformance measures, but no clear discrimination between these techniques about the aptness forrelevant applications. Lot of reports insists its work to be greater but a detailed analysis is missing in these works. In thispaper, awidespreadrelative analysis is focused to illustrate the qualities and limitations of various existing methods. The main goal of this work is to emphasizethe variety of automated methods which can ultimatelyserveto developing novel ideas forsolving the health care issues of the current society.

Cervical Cancer is an abnormal growth in the cervix - the lower part of the uterus which joins to the vagina. Human Papilloma Virus (HPV) is identified as the main cause for cervical cancer. Cervical cancer is the second most deadly disease among women next to breast cancer in developing countries. However, it is considered to be the most preventable female cancer if identified at an early stage. The cancerous cells may spread to other parts of the body if not identified at an early stage. Pap Smear test and acetic acid test are usually done for cancer screening. In pap test cells are taken from the vagina and cervix and are examined under a microscope for the presence of an abnormal cell. In acetic acid test, the change in features after and before the application of acetic acid is analyzed to find the existence of abnormal cell. Automated screening is becoming most common than manual screening because the latter is erroneous. This paper surveys the different automated methods available for screening the abnormal cells in pap images.

Palmprint has proved to be one of themost unique andstable biometric characteristics.Image alignment is an important step in various biometric authentication systems. Most of the existing palmprint recognition algorithmsmakes use of competitive valley detection algorithm to find some key points between fingers to establish the local coordinate system for extracting the region of interest (ROI). The ROI is consequently used for feature extraction and matching. Such alignment methods usually yield a coarse alignment of the palmprint images, while many missed and false matches are actually caused by inaccurate image alignments. To improve the palmprint verification accuracy, in this paper,we present an alignment method which involves moving and rotating the palmprints to locate at their correct position with the same direction and this helps easier feature extraction and matching along with developing the remaining portions of a palm given the partial palm images.We propose a new palmprint recognition strategy that combines the line features to refine the image alignment and then using the local features for matching. The experimental resultsshow that the proposed method greatly improves the palmprint recognition accuracy and it works in real time.