Alternative Thresholding Technique for Image Segmentation Based on Cuckoo Search and Generalized Gaussians

Object segmentation is a widely studied topic in digital image processing, as to it can be used for countless applications in several fields. This process is traditionally achieved by computing an optimal threshold from the image intensity histogram. Several algorithms have been proposed to find this threshold based on different statistical principles. However, the results generated via these algorithms contradict one another due to the many variables that can disturb an image. An accepted strategy to achieve the optimal histogram threshold, to distinguish between the object and the background, is to estimate two data distributions and find their intersection. This work proposes a strategy based on the Cuckoo Search Algorithm (CSA) and the Generalized Gaussian (GG) distribution to assess the optimal threshold. To test this methodology, we carried out several experiments in synthetic and practical scenarios and compared our results against other well-known algorithms from the literature. These practical cases comprise a medical image database and our own generated database. The results in a simulated environment show an evident advantage of the proposed strategy against other algorithms. In a real environment, this ranks among the best algorithms, making it a reliable alternative.

Neural network pattern recognition of ultrasound image gray scale intensity histogram of breast lesions to differentiate between benign and malignant lesions (Preprint)

UNSTRUCTURED The aim of this study is to analyze the effectiveness of grayscale intensity histogram to differentiate benign and malignant lesions using a convolutional neural network. Data (200 USG images, 100-malignant, 100-benign) was downloaded from an online access repository. The images were despeckled using ImageJ software and the grayscale intensity histogram values were extracted. In-built neural network pattern recognition application in Matlab R2019b was used to classify the images, which is a two-layer feed-forward network, with sigmoid hidden and softmax output neurons. The positive predictive value of the CNN was 95%. The best performance of 0.078264 was achieved at 36 epochs in the validation set. This study suggests that the grayscale intensity histogram of a USG image is an easy and feasible method to identify malignant lesions through an artificial neural network.

Assessment of engine combustion network recommendations for measurement of ignition and lift-off length

The engine combustion network recommends two different imaging-based diagnostics for the measurement of diesel spray ignition delay and lift-off length, respectively. To measure ignition delay, high-speed imaging of broadband luminosity, spectrally filtered to limit collected wavelengths below 600 nm, is recommended. This diagnostic is often referred to as broadband natural luminosity. For lift-off length measurements, the engine combustion network recommends imaging of OH* chemiluminescence. This diagnostic requires using an image-intensified camera to detect narrowly filtered light around 310 nm. Alternatively, it has been shown that the lift-off length can be measured using broadband natural luminosity, avoiding the need for an intensifier and ultraviolet-transmitting optics. However, care is needed in the collection and processing of this diagnostic to accurately isolate the chemiluminescence signal. Particularly, standard intensity thresholding techniques are not sufficient for isolating the chemiluminescence signal in broadband natural luminosity images. Thus, an intensity-histogram-based thresholding method is introduced. This article assesses the feasibility and practicality of measuring lift-off length using broadband natural luminosity using a detailed comparison to OH* chemiluminescence measurements. It is shown that lift-off length measurements using broadband natural luminosity are prone to user bias error in the optical setup and data processing, especially under moderate- to high-sooting conditions. We conclude that while OH* imaging provides the most reliable and accurate measurement of lift-off length at a wide range of ambient conditions, an intensity-histogram analysis can help discriminate the high-temperature chemiluminescence signal from others in a broadband natural luminosity image at higher-sooting operating conditions than demonstrated in current literature.

The impact of the variation of imaging factors on the robustness of Computed Tomography Radiomic Features: A review

ABSTRACTThe field of radiomics is at the forefront of personalized medicine. However, there are concerns regarding the robustness of its features against multiple medical imaging parameters and the performance of the predictive models built upon them. Therefore, our review aims to identify image perturbation factors (IPF) that most influence the robustness of radiomic features in biomedical research. We also provide insights into the validity and discrepancy of different methodologies applied to investigate the robustness of radiomic features. We selected 527 papers based on the primary criterion that the papers had imaging parameters that affected the reproducibility of radiomic features extracted from computed tomography (CT) images. We compared the reported performance of these parameters along with IPF in the eligible studies. We then proceeded to divide our studies into three groups based on the type of their IPF: (i) scanner parameters, (ii) acquisition parameters and (iii) reconstruction parameters. Our review highlighted that the reconstruction algorithm was the most reproducible factor and shape along with intensity histogram (IH) were the most robust radiomic features against variation in imaging parameters. This review identified substantial inconsistencies related to the methodology and the reporting style of the reviewed studies such as type of study performed, the metrics used for robustness, the feature extraction techniques, the image perturbation factors, the reporting style and their outcome inclusion. Finally, we hope the IPFs and the methodology inconsistencies identified will aid the scientific community in conducting research in a way that is more reproducible and avoids the pitfalls of previous analyses.

Neural network pattern recognition of ultrasound image gray scale intensity histogram of breast lesions to differentiate between benign and malignant lesions

ABSTRACTThe aim of this study is to analyze the effectiveness of grayscale intensity histogram to differentiate benign and malignant lesions using a convolutional neural network. Data (200 USG images, 100-malignant, 100-benign) was downloaded from an online access repository. The images were despeckled using ImageJ software and the grayscale intensity histogram values were extracted. In-built neural network pattern recognition application in Matlab R2019b was used to classify the images, which is a two-layer feed-forward network, with sigmoid hidden and softmax output neurons. The positive predictive value of the CNN was 95%. The best performance of 0.078264 was achieved at 36 epochs in the validation set. This study suggests that the grayscale intensity histogram of a USG image is an easy and feasible method to identify malignant lesions through an artificial neural network.