Simple Quantification of Surface Uptake in F-18 Florapronol PET/CT Imaging for the Validation of Alzheimer’s Disease

We developed a novel quantification method named shape feature using F-18 florapronol positron emission tomography–computed tomography (PET/CT) and evaluated its sensitivity and specificity for discriminating between patients with Alzheimer’s disease (AD) and patients with mild cognitive impairment or other precursors dementia (non-AD). We calculated the cerebral amyloid smoothing score (CASS) and brain atrophy index (BAI) using the surface area and volume of the region of interest in PET images. We calculated gray and white matter from trained CT data, prepared using U-net. Shape feature was calculated by multiplying CASS with BAI scores. We measured region-based standard uptake values (SUVr) and performed receiver operating characteristic (ROC) analysis to compare SUVr, shape feature, CASS, and BAI score. We investigated the relationship between shape feature and neuropsychological tests. Fifty subjects (23 with AD and 27 with non-AD) were evaluated. SUVr, shape feature, CASS, and BAI score were significantly higher in patients with AD than in those with non-AD. There was no statistically significant difference between shape feature and SUVr in ROC analysis. Shape feature correlated well with mini-mental state examination scores. Shape feature can effectively quantify beta-amyloid deposition and atrophic changes in the brain. These results suggest that shape feature is useful in the diagnosis of AD.

Failure in Stock Price Prediction: A Comparison bettwen the Curve-Shape-Feature and Non-Curve-Shape-Feature Modes of Existing Machine Learning Algorithms

In the era of artificial intelligence, machine learning methods are successfully used in various fields. Machine learning has attracted extensive attention from investors in the financial market, especially in stock price prediction. However, one argument for the machine learning methods used in stock price prediction is that they are black-box models which are difficult to interpret. In this paper, we focus on the future stock price prediction with the historical stock price by machine learning and deep learning methods, such as support vector machine (SVM), random forest (RF), Bayesian classifier (BC), decision tree (DT), multilayer perceptron (MLP), convolutional neural network (CNN), bi-directional long-short term memory (BiLSTM), the embedded CNN, and the embedded BiLSTM. Firstly, we manually design several financial time series where the future price correlates with the historical stock prices in pre-designed modes, namely the curve-shape-feature (CSF) and the non-curve-shape-feature (NCSF) modes. In the CSF mode, the future prices can be extracted from the curve shapes of the historical stock prices. Conversely, in the NCSF mode, they can’t. Secondly, we apply various algorithms to those pre-designed and real financial time series. We find that the existing machine learning and deep learning algorithms fail in stock price prediction because in the real financial time series, less information of future prices is contained in the CSF mode, and perhaps more information is contained in the NCSF. Various machine learning and deep learning algorithms are good at handling the CSF in historical data, which are successfully applied in image recognition and natural language processing. However, they are inappropriate for stock price prediction on account of the NCSF. Therefore, accurate stock price prediction is the key to successful investment, and new machine learning algorithms handling the NCSF series are needed.

An Improved Automatic Shape Feature Extraction Method Based on Template Matching

Image shape extraction is an important step in the image analysis, AI electronic industry and automation, as well as a significant part of content-based image retrieval(CBIR), which cannot be separated from contour extraction. However, traditional approach of the border following algorithm is susceptible to noise interference, thus the shape extracted is always complex in real images and cannot express feature of the target image well. Therefore, an improved shape feature extraction method is proposed, which converts color space into HSV model when preprocessing, filters contour by area size, merges adjacent contours by drawing convex hull and filters with template shapes. Lastly, this method is tested on UAV123 and YCB_Video dataset, which showed that the proportion of valid contour improved from less than 10% to 87.7% based on border following algorithm. In the experiment of OPenCV open source library in Visual Studio environment, we hope to improve the extraction efficiency of shape features.

Shape Feature Extraction Techniques for Fruits: A Review

Fruits sorting, recognizing, and classifying are essential post-harvest operations, as they contribute to the quality of food industry, thereby increasing the exported quantity of food. Today, an automated system for fruit classification and recognition is very important, especially when exporting to markets where quality of fruit must be high. In this study, the advantages and disadvantages of the various shape-based feature extraction algorithms and technologies that are used in sorting, classifying, and grading of fruits, as well as fruits quality estimation, are discussed in order to provide a good understanding of the use of shape-based feature extraction techniques.

Combination of Texture and Shape Features Using Machine and Deep Learning Algorithms for Breast Cancer Diagnosis

Worldwide, breast cancer is a commonly occurring disease in women. Automatic diagnosis of the lesions based on mammographic images is playing an essential role to assist experts. A novel Computer-Aided Diagnosis (CADx) scheme of breast lesion classification is proposed in this paper based on an optimized combination of texture and shape features using machine and deep learning algorithms for mass classification as benign-malignant namely C(M-ZMs)*. The main advantage of using Zernike moments for shape feature extraction is their scale, translation, and rotation invariance property, this allows omitting some of the preprocessing stages in our case. We implemented for texture feature extraction the Monogenic-Local Binary Pattern taking the advantage of lower time and space complexity because monogenic signal analysis needs fewer convolutions and generates more compact feature vectors. Therefore, we used Zernike moments for shape feature extraction due to their scale, translation, and rotation invariance property, this allows omitting some of the preprocessing stages in our proposed system. The proposed system proves its performance on some challenging breast cancer cases where the lesions exist in dense breast tissues. Validation has been undertaken on 520 mammograms from the Digital Database for Screening Mammography Database (DDSM), yielding an accuracy rate of 99.5\%.

Investigation of Tactile Perception Evoked by Ridged Texture Using ERP and Non-linear Methods

The triangular ridged surface can improve the grip reliability of products, but the sharp edge of triangular ridge induces sharp and uncomfortable feeling. To study the effect of edge shape (sharp, round, and flat shape) of triangular ridges on brain activity during touching, electroencephalograph (EEG) signals during tactile perception were evaluated using event-related potentials (ERP) and non-linear analysis methods. The results showed that the early component of P100 and P200, and the late component of P300 were successfully induced during perceiving the ridged texture. The edge shape features affect the electrical activity of brain during the tactile perceptions. The sharp shape feature evoked fast P100 latency and high P100 amplitude. The flat texture with complex (sharp and flat) shape feature evoked fast P200 latency, high P200 amplitude and RQA parameters. Both of the sharp shape and complex shape feature tended to evoke high peak amplitude of P300. The large-scale structures of recurrence plots (RPs) and recurrence quantification analysis (RQA) parameters can visually and quantitatively characterize the evolution regulation of the dynamic behavior of EEG system along with the tactile process. This study proved that RPs and RQA were protential methods for the feature extraction and state recognition of EEG during tactile perception of textured surface. This research contributes to optimize surface tactile characteristics on products, especially effective surface textures design for good grip.