Image Feature Analysis Research Articles

Estimation of Abnormal Cell Growth and MCG-Based Discriminative Feature Analysis of Histopathological Breast Images

The accurate prediction of cancer from microscopic biopsy images has always been a major challenge for medical practitioners and pathologists who manually observe the shape and structure of the cells from tissues under a microscope. Mathematical modelling of cell proliferation helps to predict tumour sizes and optimizes the treatment procedure. This paper introduces a cell growth estimation function that uncovers the growth behaviour of benign and malignant cells. To analyse the cellular level information from tissue images, we propose a minimized cellular graph (MCG) development method. The method extracts cells and produces different features that are useful in classifying benign and malignant tissues. The method’s graphical features enable a precise and timely exploration of huge amounts of data and can help in making predictions and informed decisions. This paper introduces an algorithm for constructing a minimized cellular graph which reduces the computational complexity. A comparative study is performed based on the state-of-the-art classifiers, SVM, decision tree, random forest, nearest neighbor, LDA, Naive Bayes, and ANN. The experimental data are obtained from the BreakHis dataset, which contains 2480 benign and 5429 malignant histopathological images. The proposed technique achieves a 97.7% classification accuracy which is 7% higher than that of the other graph feature-based classification methods. A comparative study reveals a performance improvement for breast cancer classification compared to the state-of-the-art techniques.

CT and MRI of Gastrointestinal Stromal Tumors: New Trends and Perspectives.

Gastrointestinal stromal tumors (GISTs) are defined as mesenchymal tumors of the gastrointestinal tract that express positivity for CD117, which is a c-KIT proto-oncogene antigen. Expression of the c-KIT protein, a tyrosine kinase growth factor receptor, allows the distinction between GISTs and other mesenchymal tumors such as leiomyoma, leiomyosarcoma, schwannoma and neurofibroma. GISTs can develop anywhere in the gastrointestinal tract, as well as in the mesentery and omentum. Over the years, the management of GISTs has improved due to a better knowledge of their behaviors and risk or recurrence, the identification of specific mutations and the use of targeted therapies. This has resulted in a better prognosis for patients with GISTs. In parallel, imaging of GISTs has been revolutionized by tremendous progress in the field of detection, characterization, survival prediction and monitoring during therapy. Recently, a particular attention has been given to radiomics for the characterization of GISTs using analysis of quantitative imaging features. In addition, radiomics has currently many applications that are developed in conjunction with artificial intelligence with the aim of better characterizing GISTs and providing a more precise assessment of tumor burden. This article sums up recent advances in computed tomography and magnetic resonance imaging of GISTs in the field of image/data acquisition, tumor detection, tumor characterization, treatment response evaluation, and preoperative planning.

What Matters in Radiological Image Segmentation? Effect of Segmentation Errors on the Diagnostic Related Features.

Segmentation is a crucial step in extracting the medical image features for clinical diagnosis. Though multiple metrics have been proposed to evaluate the segmentation performance, there is no clear study on how or to what extent the segmentation errors will affect the diagnostic related features used in clinical practice. Therefore, we proposed a segmentation robustness plot (SRP) to build the link between segmentation errors and clinical acceptance, where relative area under the curve (R-AUC) was designed to help clinicians to identify the robust diagnostic related image features. In experiments, we first selected representative radiological series from time series (cardiac first-pass perfusion) and spatial series (T2 weighted images on brain tumors) of magnetic resonance images, respectively. Then, dice similarity coefficient (DSC) and Hausdorff distance (HD), as the widely used evaluation metrics, were used to systematically control the degree of the segmentation errors. Finally, the differences between diagnostic related image features extracted from the ground truth and the derived segmentation were analyzed, using the statistical method large sample size T-test to calculate the corresponding p values. The results are denoted in the SRP, where the x-axis indicates the segmentation performance using the aforementioned evaluation metric, and the y-axis shows the severity of the corresponding feature changes, which are expressed in either the p values for a single case or the proportion of patients without significant change. The experimental results in SRP show that when DSC is above 0.95 and HD is below 3mm, the segmentation errors will not change the features significantly in most cases. However, when segmentation gets worse, additional metrics are required for further analysis. In this way, the proposed SRP indicates the impact of the segmentation errors on the severity of the corresponding feature changes. By using SRP, one could easily define the acceptable segmentation errors in a challenge. Additionally, the R-AUC calculated from SRP provides an objective reference to help the selection of reliable features in image analysis.

Imaging features based on CT and MRI for predicting prognosis of patients with intrahepatic cholangiocarcinoma: a single-center study and meta-analysis

BackgroundTo evaluate the prognostic role of imaging features based on CT and MRI in intrahepatic cholangiocarcinoma (ICC).MethodsTwo hundred and four patients from a single-center database who underwent radical ICC surgery from 2010 to 2019 were enrolled in the study. Cox proportional hazard model was used for survival analysis of imaging features. A meta-analysis was performed to determine imaging features that predict overall survival (OS) and event-free survival (EFS) in ICC.ResultsIn the CT group of the retrospective cohort, tumor multiplicity, infiltrative tumor margin, lymph node metastasis, enhancement pattern in hepatic arterial phase and tumor necrosis correlated with poorer EFS and OS; moreover, enhancing capsules, high carcinoembryonic antigen levels contributed to poor OS. In the MRI group, tumor multiplicity and enhancement pattern were prognostic factors for OS; tumor multiplicity and enhancement pattern resulted in poor EFS. A total of 13 articles containing 1822 patients with ICC were enrolled in the adjusted hazard ratios meta-analysis. The results showed that enhancement pattern and infiltrative tumor margin were predictors of OS and EFS, whereas bile duct invasion was a predictor of OS.ConclusionsArterial enhancement patterns and tumor margin status were associated with both OS and EFS of ICC patients following resection.

Radiomic applications in upper gastrointestinal cancer surgery.

Cross-sectional imaging plays an integral role in the management of upper gastrointestinal (UGI) cancer, from initial diagnosis and staging to determining appropriate treatment strategies. Subjective imaging interpretation has known limitations. The field of radiomics has evolved to extract quantitative data from medical imaging and relate these to biological processes. The key concept behind radiomics is that the high-throughput analysis of quantitative imaging features can provide predictive or prognostic information, with the goal of providing individualised care. Radiomic studies have shown promising utility in upper gastrointestinal oncology, highlighting a potential role in determining stage of disease and degree of tumour differentiation and predicting recurrence-free survival. This narrative review aims to provide an insight into the concepts underpinning radiomics, as well as its potential applications for guiding treatment and surgical decision-making in upper gastrointestinal malignancy. Outcomes from studies to date have been promising; however, further standardisation and collaboration are required. Large prospective studies with external validation and evaluation of radiomic integration into clinical pathways are needed. Future research should now focus on translating the promising utility of radiomics into meaningful patient outcomes.

Discordant and Converting Receptor Expressions in Brain Metastases from Breast Cancer: MRI-Based Non-Invasive Receptor Status Tracking.

Discordance and conversion of receptor expressions in metastatic lesions and primary tumors is often observed in patients with brain metastases from breast cancer. Therefore, personalized therapy requires continuous monitoring of receptor expressions and dynamic adaptation of applied targeted treatment options. Radiological in vivo techniques may allow receptor status tracking at high frequencies at low risk and cost. The present study aims to investigate the potential of receptor status prediction through machine-learning-based analysis of radiomic MR image features. The analysis is based on 412 brain metastases samples from 106 patients acquired between 09/2007 and 09/2021. Inclusion criteria were as follows: diagnosed cerebral metastases from breast cancer; histopathology reports on progesterone (PR), estrogen (ER), and human epidermal growth factor 2 (HER2) receptor status; and availability of MR imaging data. In total, 3367 quantitative features of T1 contrast-enhanced, T1 non-enhanced, and FLAIR images and corresponding patient age were evaluated utilizing random forest algorithms. Feature importance was assessed using Gini impurity measures. Predictive performance was tested using 10 permuted 5-fold cross-validation sets employing the 30 most important features of each training set. Receiver operating characteristic areas under the curves of the validation sets were 0.82 (95% confidence interval [0.78; 0.85]) for ER+, 0.73 [0.69; 0.77] for PR+, and 0.74 [0.70; 0.78] for HER2+. Observations indicate that MR image features employed in a machine learning classifier could provide high discriminatory accuracy in predicting the receptor status of brain metastases from breast cancer.

Value of High-Frequency Ultrasonography in the Qualitative and Semi-Quantitative Assessment of Immunoglobulin G4-Related Submandibular Sialadenitis.

To assess the value of high-frequency ultrasonography in the evaluation of immunoglobulin G4-related submandibular sialadenitis (IgG4-SS). Thirty-four submandibular glands in 17 patients with IgG4-SS were retrospectively enrolled, as well as 34 submandibular glands in 17 healthy control subjects. Qualitative ultrasonic features including submandibular gland size, border, echogenicity, and vascularity were reviewed. Two different scoring systems (0-16 and 0-48, respectively) were used for semi-quantitative analysis of imaging features. Comparison of both qualitative and semi-quantitative ultrasonic analysis were made between patients with IgG4-SS and healthy controls. Spearman correlation was used to explore relationships between variables. The submandibular glands with IgG4-SS presented with enlarged size, rough border, increased vascularity, and abnormal echogenicity (All P < .05). The most common echogenicity pattern for IgG-SS was diffuse hypoechoic foci pattern (44.1%), followed by superficial hypoechoic pattern (20.6%), tumor-like pattern (14.7%), and diffuse hypoechogenicity pattern (11.8%). Most IgG4-SS glands presented linear hyperechogenicity in parenchyma (91.2%). Based on both scoring system, scores of IgG4-SS were significantly higher than those of the controls (All P < .05). Association analysis of both scoring systems showed positive correlation of scores with vascularity in the gland parenchyma (All P < .05). The ultrasonic features of IgG4-SS comprise enlarged gland, rough border, increased vascularity, and abnormal echogenicity, which correlate with its pathological characteristics. The most common echogenicity pattern for IgG4-SS was diffuse hypoechoic foci pattern. Semi-quantitative analysis systems could be useful in the assessment of IgG4-SS. Ultrasound is a potential, valuable, and non-invasive tool for the diagnosis and evaluation of IgG4-SS.

Detection and infected area segmentation of apple fire blight using image processing and deep transfer learning for site-specific management

Advanced sensing technologies and deep learning models are needed for automatic recognition of pathogens to protect trees in orchards. This study developed a fire blight disease detection and infected area segmentation system using image processing and deep learning approaches to automate the detection process in a complex apple orchard environment for site-specific management. Two types of images were acquired: multispectral images from an unmanned aerial vehicle (UAV) using a multispectral camera and red–greenblue (RGB) images from the ground using two different cameras. Multispectral images were preprocessed and used for image feature analysis by calculating vegetation indices, including excessive blue (ExB), normalized difference vegetation index (NDVI), green normalized difference vegetation index (GNDVI), red-edge normalized difference vegetation index (RENDVI), modified ratio vegetation index (RVI), and triangular blueness index (TBI). Vegetation indices were calculated from a total of 60 multispectral images (30 heathy and 30 fire blight infected). Results showed that RVI was most sensitive to fire blight infection among the six indices. A support vector machine model was used to classify unhealthy tree canopies. A Mask Region-Convolutional Neural Network (Mask R-CNN) based deep learning model was developed from RGB infected images. A total of 880 images were used for training, and 220 images were used for validation. Another 110 images were used for testing the trained Mask R-CNN model. A precision of 92.8 % and recall of 91.2 % were obtained by detecting the infected canopies using a ResNet-101 backbone and intersection over union (IoU) threshold of 0.7. The high precision demonstrates the effectiveness of Mask R-CNN for the identification and segmentation of fire blight infection in images taken in complex orchard conditions. These results prove the potential of this non-invasive sensing method in detecting disease in commercial fruit production for site-specific infected canopies removing.

Imaging the WHO 2021 Brain Tumor Classification: Fully Automated Analysis of Imaging Features of Newly Diagnosed Gliomas.

The fifth version of the World Health Organization (WHO) classification of tumors of the central nervous system (CNS) in 2021 brought substantial changes. Driven by the enhanced implementation of molecular characterization, some diagnoses were adapted while others were newly introduced. How these changes are reflected in imaging features remains scarcely investigated. We retrospectively analyzed 226 treatment-naive primary brain tumor patients from our institution who received extensive molecular characterization by epigenome-wide methylation microarray and were diagnosed according to the 2021 WHO brain tumor classification. From multimodal preoperative 3T MRI scans, we extracted imaging metrics via a fully automated, AI-based image segmentation and processing pipeline. Subsequently, we examined differences in imaging features between the three main glioma entities (glioblastoma, astrocytoma, and oligodendroglioma) and particularly investigated new entities such as astrocytoma, WHO grade 4. Our results confirm prior studies that found significantly higher median CBV (p = 0.00003, ANOVA) and lower median ADC in contrast-enhancing areas of glioblastomas, compared to astrocytomas and oligodendrogliomas (p = 0.41333, ANOVA). Interestingly, molecularly defined glioblastoma, which usually does not contain contrast-enhancing areas, also shows significantly higher CBV values in the non-enhancing tumor than common glioblastoma and astrocytoma grade 4 (p = 0.01309, ANOVA). This work provides extensive insights into the imaging features of gliomas in light of the new 2021 WHO CNS tumor classification. Advanced imaging shows promise in visualizing tumor biology and improving the diagnosis of brain tumor patients.

Prediction of Hole Expansion Ratio for Advanced High-Strength Steel with Image Feature Analysis of Sheared Edge

The punching process of AHSS induces edge cracks in successive process, limiting the application of AHSS for vehicle bodies. Controlling and predicting edge quality is substantially difficult due to the large variation in edge quality, die wear induced by high strength, and the complex effect of phase distribution. To overcome this challenge, a quality prediction model that considers the variation of the entire edge should be developed. In this study, the image of the entire edge was analyzed to provide a comprehensive evaluation of its quality. Statistical features were extracted from the edge images to represent the edge quality of DP780, DP980, and MART1500 steels. Combined with punching monitoring signals, a prediction model for hole expansion ratio was developed under punch conditions of varying clearance, punch angle, and punch edge radius. It was found that the features of grayscale variation are affected by the punching conditions and are related to the double burnish and uneven burr, which degrades the edge quality. Prediction of HER was possible based on only edge image and monitoring signals, with the same performance as the prediction based solely on punching parameters and material properties. The prediction performance improved when using all the features.

A comparative analysis of demographic, clinical and imaging features of myelin oligodendrocyte glycoprotein antibody positive, aquaporin 4 antibody positive, and double seronegative demyelinating disorders - An Indian tertiary care center prospective study.

The aim of the study was to study the demographical, clinical, radiological features, and outcome of anti-myelin oligodendrocyte glycoprotein (MOG) antibody spectrum disorder and compare these features with patients negative for anti-MOG antibody. MOG antibody-associated disease (MOGAD) and aquaporin-4 (AQP4) antibody-related diseases are immunologically distinct pathologies. Our aim was to compare the clinical and radiological features of MOG antibody-related diseases with AQP4 antibody-related diseases and seronegative demyelinating diseases (Non-multiple sclerosis). This was a prospective and cohort study conducted at an apex tertiary care institute in the northern part of India from Jan 2019 to May 2021. We compared clinical, laboratory, and radiological findings of patients with MOGAD, AQP4 antibody-related diseases, and seronegative demyelinating disease. There were a total of 103 patients - 41 patients of MOGAD, 37 patients of AQP4 antibody-related diseases and 25 seronegative demyelinating disease. Bilateral optic neuritis was the most frequent phenotype in patients with MOGAD (18/41) whereas myelitis was the most common phenotype in the AQP4 (30/37) and seronegative groups (13/25). Cortical, juxtacortical lesions, anterior segment optic neuritis, optic sheath enhancement, and conus involvement in myelitis were radiological findings that separated MOGAD from AQP4 related diseases. Nadir Expanded Disability Status Scale (EDSS) and visual acuity were similar across the groups. Last follow-up EDSS was significantly better in the MOG antibody group as compared to AQP4 antibody group (1 [0-8] vs. 3.5 [0-8]; P = 0.03). Encephalitis, myelitis, and seizures were more common in the younger population (<18 vs. >18 years) in MOGAD (9 vs. 2, P = 0.001; 9 vs. 7, P = 0.03; 6 vs. 0, P = 0.001). We identified several clinical and radiological features that can help physicians to distinguish MOGAD from AQP4-immunoglobulin G+neuromyelitis optica spectrum disorder. Differentiation is vital as treatment response might vary among both groups.

FEATURES OF HANDWRITING IMAGE PROCESSING AND ANALYSIS

FEATURES OF HANDWRITING IMAGE PROCESSING AND ANALYSIS

Research on welding penetration status monitoring based on Residual-Group convolution model

Welding penetration cannot be directly visible in non-penetrating laser lap welding, making it difficult to monitor the penetration status online. Aiming at this issue, scale-invariant feature transform (SIFT) and image thresholding segmentation are adopted to calculate and analyze the steady evolution of image feature information of welding images in three different statuses of the side-axis image acquisition system. According to the regularity evolution, a Residual-Group convolution model (Res-GCM) is proposed for feature extraction and analysis of welding images to predict the penetration status of non-penetrating laser welding. Compared with the VGG, ResNet, and WresNet models, Res-GCM has the best test accuracy of 90.53%. In the ablation experiments, the introduction of the Residual-Group convolution module can improve test accuracy by 6% compared with the residual module and 8% compared with the traditional convolutional module. Gradient-weighted class activation mapping (Grad-CAM) visualization results show that the train weights in each model layer are concentrated in the plasma and molten pool regions.

Pre-operative MRI features predict early post-operative recurrence of hepatocellular carcinoma with different degrees of pathological differentiation

PurposeTo investigate the value of pre-operative gadoxetate disodium (Gd-EOB-DTPA) enhanced MRI predicting early post-operative recurrence (< 2 years) of hepatocellular carcinoma (HCC) with different degrees of pathological differentiation.MethodsRetrospective analysis of pre-operative MR imaging features of 177 patients diagnosed as suffering from HCC and that underwent radical resection. Multivariate logistic regression assessment was adopted to assess predictors for HCC recurrence with different degrees of pathological differentiation. The area under the curve (AUC) of receiver operating characteristics (ROC) was utilized to assess the diagnostic efficacy of the predictors.ResultsAmong the 177 patients, 155 (87.5%) were males, 22 (12.5%) were females; the mean age was 49.97 ± 10.71 years. Among the predictors of early post-operative recurrence of highly-differentiated HCC were an unsmooth tumor margin and an incomplete/without tumor capsule (p = 0.037 and 0.033, respectively) whereas those of early post-operative recurrence of moderately-differentiated HCC were incomplete/without tumor capsule, peritumoral enhancement along with peritumoral hypointensity (p = 0.006, 0.046 and 0.004, respectively). The predictors of early post-operative recurrence of poorly-differentiated HCC were peritumoral enhancement, peritumoral hypointensity, and tumor thrombosis (p = 0.033, 0.006 and 0.021, respectively). The AUCs of the multi-predictor diagnosis of early post-operative recurrence of highly-, moderately-, and poorly-differentiated HCC were 0.841, 0.873, and 0.875, respectively. The AUCs of the multi-predictor diagnosis were each higher than for those predicted separately.ConclusionsThe imaging parameters for predicting early post-operative recurrence of HCC with different degrees of pathological differentiation were different and combining these predictors can improve the diagnostic efficacy of early post-operative HCC recurrence.

Remote Sensing Image Scene Classification: Advances and Open Challenges

Deep learning approaches are gaining popularity in image feature analysis and in attaining state-of-the-art performances in scene classification of remote sensing imagery. This article presents a comprehensive review of the developments of various computer vision methods in remote sensing. There is currently an increase of remote sensing datasets with diverse scene semantics; this renders computer vision methods challenging to characterize the scene images for accurate scene classification effectively. This paper presents technology breakthroughs in deep learning and discusses their artificial intelligence open-source software implementation framework capabilities. Further, this paper discusses the open gaps/opportunities that need to be addressed by remote sensing communities.

A Survey on Evolutionary Computation for Computer Vision and Image Analysis: Past, Present, and Future Trends

Computer vision (CV) is a big and important field in artificial intelligence covering a wide range of applications. Image analysis is a major task in CV aiming to extract, analyse and understand the visual content of images. However, image-related tasks are very challenging due to many factors, e.g., high variations across images, high dimensionality, domain expertise requirement, and image distortions. Evolutionary computation (EC) approaches have been widely used for image analysis with significant achievement. However, there is no comprehensive survey of existing EC approaches to image analysis. To fill this gap, this paper provides a comprehensive survey covering all essential EC approaches to important image analysis tasks including edge detection, image segmentation, image feature analysis, image classification, object detection, and others. This survey aims to provide a better understanding of evolutionary computer vision (ECV) by discussing the contributions of different approaches and exploring how and why EC is used for CV and image analysis. The applications, challenges, issues, and trends associated to this research field are also discussed and summarised to provide further guidelines and opportunities for future research.

Nipple-areolar complex (NAC) or skin flap ischemia necrosis post nipple-sparing mastectomy (NSM)—analysis of clinicopathologic factors and breast magnetic resonance imaging (MRI) features

BackgroundThe purpose of this study is to identify clinicopathologic factors and/or preoperative MRI vascular patterns in the prediction of ischemia necrosis of the nipple-areola complex (NAC) or skin flap post nipple-sparing mastectomy (NSM).MethodsWe performed a retrospective analysis of 441 NSM procedures from January 2011 to September 2021 from the breast cancer database at our institution. The ischemia necrosis of NAC or skin flap was evaluated in correlation with clinicopathologic factors and types of skin incision. Patients who received NSM with preoperative MRI evaluation were further evaluated for the relationship between vascular pattern and the impact on ischemia necrosis of NAC or skin flap.ResultsA total of 441 cases with NSM were enrolled in the current study, and the mean age of the cases was 49.1 ± 9.8 years old. A total of 41 (9.3%) NSM procedures were found to have NAC ischemia/necrosis. Risk factors were evaluated of which old age, large mastectomy specimen weight (> 450 g), and peri-areola incision were identified as predictors of NAC necrosis. Two-hundred seventy NSM procedures also received preoperative MRI, and the blood supply pattern was 18% single-vessel type and 82% double-vessel pattern. There were no correlations between MRI blood supply patterns or types of skin flap incisions with ischemia necrosis of NAC. There were also no correlations between blood loss and the pattern or size of the blood vessel.ConclusionFactors such as the type of skin incision, age, and size of mastectomy weight played an important role in determining ischemia necrosis of NAC; however, MRI vascular (single or dual vessel supply) pattern was not a significant predictive factor.

Pretreatment CT differential diagnosis of tuberculous peritonitis from peritoneal carcinomatosis of advanced epithelial ovarian cancer

Delayed diagnosis of female genital tuberculosis (FGTB) can lead to inappropriate treatment and unnecessary surgical procedures rather than standard anti-TB medication. We tried to evaluate the use of computed tomography (CT) imaging to differentiate TB peritonitis from peritoneal carcinomatosis of advanced epithelial ovarian cancer (AEOC). We investigated women who underwent CT to distinguish between TB peritonitis and peritoneal carcinomatosis of AEOC. We evaluated various CT imaging features to identify differences between the two diseases. In addition, we performed univariate and multivariate logistic regression analyses to identify the independent imaging parameters associated with TB peritonitis and evaluated the diagnostic performance of the combined imaging parameters. We also performed the histopathological analysis of the available salpinx specimens of TB peritonitis. We included 25 women with TB peritonitis and 34 women with peritoneal carcinomatosis of AEOC. A multivariate analysis of the discriminant CT imaging features between the two diseases revealed that changes in fallopian tubes and peritoneal micronodules were independent parameters associated with TB peritonitis (p ≤ 0.012). Combining the two imaging parameters showed an area under the receiver operating characteristic curve of 0.855, a sensitivity of 88.0%, and a specificity of 67.7% for differentiating TB peritonitis from peritoneal carcinomatosis. Furthermore, changes in fallopian tubes were correlated with histopathological abnormalities in salpinx specimens. Pretreatment CT evaluation with useful imaging features could help differentiate TB peritonitis from peritoneal carcinomatosis of AEOC.

Image segmentation and flow prediction of digital rock with U-net network

Computed tomography (CT) images of sandstone contain rich reservoir information. Analyzing digital rock images is important for geological research and the flow in the subsurface. This paper presents a workflow for assessing digital rock petrophysical properties based on machine learning techniques, including 1) automatic segmentation of sandstone rock images using U-net networks, 2) permeability prediction using machine learning, and 3) flow simulation by deep learning. First, using the U-net network, the rock images are binary-segmented into matrix and pore, and multisegmented into the matrix, pore, and mineral. The accuracy and intersection over union (IOU) are used to evaluate the performance of image segmentation. The accuracy and IOU of binary segmentation results are 99.87% and 0.9986, and the results for multi-segmentation are 96.77% and 0.7281, respectively. Then, the key features of CT images influencing sandstone permeability are extracted, and the analysis of image features reveals that the hydraulic radius is the most important parameter for permeability prediction. After that, the sandstone permeability is predicted by long short-term memory (LSTM) and random forest (RF) and then compared with the permeability calculated by the lattice Boltzmann (LBM) method. The mean square error (MSE), mean absolute error (MAE), and root mean square error (RMSE) are used to quantitatively evaluate the error of permeability prediction. The studies show that the precision of RF in permeability prediction is higher than that of LSTM, and when all the feature parameters are used as input, the accuracy of permeability prediction is a little higher than that when only the hydraulic radius is used as input. Finally, this paper refines a new U-net model to predict the flow velocity field from CT images, and this new U-net model can reduce the computation time by 98.59% compared with the LBM method. This study will be significant for applying deep learning in simulate the flow in digital rock.

Masked Face Emotion Recognition Based on Facial Landmarks and Deep Learning Approaches for Visually Impaired People.

Current artificial intelligence systems for determining a person's emotions rely heavily on lip and mouth movement and other facial features such as eyebrows, eyes, and the forehead. Furthermore, low-light images are typically classified incorrectly because of the dark region around the eyes and eyebrows. In this work, we propose a facial emotion recognition method for masked facial images using low-light image enhancement and feature analysis of the upper features of the face with a convolutional neural network. The proposed approach employs the AffectNet image dataset, which includes eight types of facial expressions and 420,299 images. Initially, the facial input image's lower parts are covered behind a synthetic mask. Boundary and regional representation methods are used to indicate the head and upper features of the face. Secondly, we effectively adopt a facial landmark detection method-based feature extraction strategy using the partially covered masked face's features. Finally, the features, the coordinates of the landmarks that have been identified, and the histograms of the oriented gradients are then incorporated into the classification procedure using a convolutional neural network. An experimental evaluation shows that the proposed method surpasses others by achieving an accuracy of 69.3% on the AffectNet dataset.