Search

Inkjet printing technology can make tiles with very rich and realistic patterns, so it is widely adopted in the ceramic industry. However, the frequent nozzle blockage and inconsistent inkjet volume by inkjet printing devices, usually leads to defects such as stayguy and color blocks in the tile surface. Especially, the stayguy in complex pattern is difficult to identify by naked eyes due to it is easily covered by complex patterns and becomes invisible, this brings great challenge to tile quality inspection. Nowadays, the machine learning is employed to address the issues. The existing machine learning methods based on hand‐crafted features are capable of stayguy detection of the tiles with a simple pattern, but not applicable for complex patterns due to the interference of pattern in feature extraction. The emerging deep‐learning‐based methods have the potential to be applied for stayguy detection with complex patterns, but cannot achieve real‐time detection due to high complexity. In this paper, a lightweight hand‐crafted feature enhanced convolutional neural network (named HFENet) is proposed for rapid defect detection of tile surface. First, we perform data enhancement on the original image by global histogram equalization and image addition. Second, for the special shape of stayguy which is usually vertical, we embed the extended vertical edge detection operator (Prewitt) as convolution kernel into HFENet to extract the hand‐crafted vertical edge features of the test image and eliminate the interference of complex pattern in the feature extraction. Third, the 5 × 1 asymmetric convolution kernel with a dilation rate of 2 is used to improve the utilization of convolution kernel and reduce the complexity of the model. Fourth, to reach the real‐time requirements, a memory access cost‐aware design is proposed, which can orchestrate the number of shallow convolution layers and deep convolution layers in feature extraction. The experiments were performed on the ceramic tile image data set captured by high‐resolution industrial cameras in ceramic tile production line. Experimental results show that the HFENet outperforms the state‐of‐the‐art semantic segmentation networks (i.e., UNet, FCN‐8s, SegNet, DeepLabV3+, etc.) and lightweight networks (i.e., ShuffleNet, MobileNet, and SqueezeNet). All the code and data are available at a GitHub repository (https://github.com/RobotvisionLab/HFENet). [ABSTRACT FROM AUTHOR]

Alpha-1 antitrypsin deficiency associated liver disease (AATD-LD) is a rare genetic disorder and not well-recognized. Predicting the clinical outcomes of AATD-LD and defining patients more likely to progress to advanced liver disease are crucial for better understanding AATD-LD progression and promoting timely medical intervention. We aimed to develop a tailored machine learning (ML) model to predict the disease progression of AATD-LD. This analysis was conducted through a stacking ensemble learning model by combining five different ML algorithms with 58 predictor variables using nested five-fold cross-validation with repetitions based on the UK Biobank data. Performance of the model was assessed through prediction accuracy, area under the receiver operating characteristic (AUROC), and area under the precision-recall curve (AUPRC). The importance of predictor contributions was evaluated through a feature importance permutation method. The proposed stacking ensemble ML model showed clinically meaningful accuracy and appeared superior to any single ML algorithms in the ensemble, e.g., the AUROC for AATD-LD was 68.1%, 75.9%, 91.2%, and 67.7% for all-cause mortality, liver-related death, liver transplant, and all-cause mortality or liver transplant, respectively. This work supports the use of ML to address the unanswered clinical questions with clinically meaningful accuracy using real-world data. [ABSTRACT FROM AUTHOR]

Defects in the textile manufacturing process lead to a great waste of resources and further affect the quality of textile products. Automated quality guarantee of textile fabric materials is one of the most important and demanding computer vision tasks in textile smart manufacturing. This survey presents a thorough overview of algorithms for fabric defect detection. First, this review briefly introduces the importance and inevitability of fabric defect detection towards the era of manufacturing of artificial intelligence. Second, defect detection methods are categorized into traditional algorithms and learning-based algorithms, and traditional algorithms are further categorized into statistical, structural, spectral, and model-based algorithms. The learning-based algorithms are further divided into conventional machine learning algorithms and deep learning algorithms which are very popular recently. A systematic literature review on these methods is present. Thirdly, the deployments of fabric defect detection algorithms are discussed in this study. This paper provides a reference for researchers and engineers on fabric defect detection in textile manufacturing. [ABSTRACT FROM AUTHOR]

Class-specific cost regulation extreme learning machine (CCR-ELM) can effectively deal with the class imbalance problems. However, its key parameters, including the number of hidden nodes, the input weights, the biases and the tradeoff factors are normally generated randomly or preset by human. Moreover, the number of input weights and biases depend on the size of hidden layer. Inappropriate quantity of hidden nodes may lead to the useless or redundant neuron nodes, and make the whole structure complex, even cause the worse generalization and unstable classification performances. Based on this, an adaptive CCR-ELM with variable-length brain storm optimization algorithm is proposed for the class imbalance learning. Each individual consists of all above parameters of CCR-ELM and its length varies with the number of hidden nodes. A novel mergence operator is presented to incorporate two parent individuals with different length and generate a new individual. The experimental results for nine imbalance datasets show that variable-length brain storm optimization algorithm can find better parameters of CCR-ELM, resulting in the better classification accuracy than other evolutionary optimization algorithms, such as GA, PSO, and VPSO. In addition, the classification performance of the proposed adaptive algorithm is relatively stable under varied imbalance ratios. Applying the proposed algorithm in the fault diagnosis of conveyor belt also proves that ACCR-ELM with VLen-BSO has the better classification performances. [ABSTRACT FROM AUTHOR]

Slit-lamp images play an essential role for diagnosis of pediatric cataracts. We present a computer vision-based framework for the automatic localization and diagnosis of slit-lamp images by identifying the lens region of interest (ROI) and employing a deep learning convolutional neural network (CNN). First, three grading degrees for slit-lamp images are proposed in conjunction with three leading ophthalmologists. The lens ROI is located in an automated manner in the original image using two successive applications of Candy detection and the Hough transform, which are cropped, resized to a fixed size and used to form pediatric cataract datasets. These datasets are fed into the CNN to extract high-level features and implement automatic classification and grading. To demonstrate the performance and effectiveness of the deep features extracted in the CNN, we investigate the features combined with support vector machine (SVM) and softmax classifier and compare these with the traditional representative methods. The qualitative and quantitative experimental results demonstrate that our proposed method offers exceptional mean accuracy, sensitivity and specificity: classification (97.07%, 97.28%, and 96.83%) and a three-degree grading area (89.02%, 86.63%, and 90.75%), density (92.68%, 91.05%, and 93.94%) and location (89.28%, 82.70%, and 93.08%). Finally, we developed and deployed a potential automatic diagnostic software for ophthalmologists and patients in clinical applications to implement the validated model. [ABSTRACT FROM AUTHOR]

Background: Immune checkpoint inhibitor (ICI) therapy is a promising treatment for cancer. However, the response rate to ICI therapy in hepatocellular carcinoma (HCC) patients is low (approximately 30%). Thus, an approach to predict whether a patient will benefit from ICI therapy is required. This study aimed to design a classifier based on circulating indicators to identify patients suitable for ICI therapy. Methods: This retrospective study included HCC patients who received immune checkpoint inhibitor therapy between March 2017 and September 2023 at Nanjing Drum Tower Hospital and Jinling Hospital. The levels of the 17 serum biomarkers and baseline patients' characters were assessed to discern meaningful circulating indicators related with survival benefits using random forest. A prognostic model was then constructed to predict survival of patients after treatment. Results: A total of 369 patients (mean age 56, median follow-up duration 373 days,) were enrolled in this study. Among the 17 circulating biomarkers, 11 were carefully selected to construct a classifier. Receiver operating characteristic (ROC) analysis yielded an area under the curve (AUC) of 0.724. Notably, patients classified into the low-risk group exhibited a more positive prognosis (P = 0.0079; HR, 0.43; 95% CI 0.21– 0.87). To enhance efficacy, we incorporated 11 clinical features. The extended model incorporated 12 circulating indicators and 5 clinical features. The AUC of the refined classifier improved to 0.752. Patients in the low-risk group demonstrated superior overall survival compared with those in the high-risk group (P = 0.026; HR 0.39; 95% CI 0.11– 1.37). Conclusion: Circulating biomarkers are useful in predicting therapeutic outcomes and can help in making clinical decisions regarding the use of ICI therapy. [ABSTRACT FROM AUTHOR]

Detecting license plates has never been easy, particularly with the proliferation of sophisticated radars on highways and roads. By 2021, the gendarmerie and National Security Road control agents will have access to more than 1 billion smart traffic radars worldwide. This research presents a revolutionary technique for detecting and recognizing Arabic and Latin license plates. After assembling the gathered images to create a novel dataset, we utilized YOLO v7 to locate and identify the number plate in the image as the first step of the suggested procedure. Before the dataset was fed to the detection system, it was manually labeled. Afterward, we improved the recognized license plate using machine learning methods. To do this, we used kernel methods as well as thresholding to get rid of the extra vertical lines on the plate. After that, we employed Arabic OCR along with Easy OCR methods to decipher the Latin and Arabic characters on the number plate. Eventually, the proposed method achieved an F1 score of 98%,with a precision and recall of 97% and 98%, respectively. We also obtained an accuracy of 99% for image segmentation. The segmentation and detection results from the suggested strategy have shown satisfactory results. [ABSTRACT FROM AUTHOR]

Background: To develop and validate an MRI texture-based machine learning model for the noninvasive assessment of renal function. Methods: A retrospective study of 174 diabetic patients (training cohort, n = 123; validation cohort, n = 51) who underwent renal MRI scans was included. They were assigned to normal function (n = 71), mild or moderate impairment (n = 69), and severe impairment groups (n = 34) according to renal function. Four methods of kidney segmentation on T2-weighted images (T2WI) were compared, including regions of interest covering all coronal slices (All-K), the largest coronal slices (LC-K), and subregions of the largest coronal slices (TLCO-K and PIZZA-K). The speeded-up robust features (SURF) and support vector machine (SVM) algorithms were used for texture feature extraction and model construction, respectively. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of models. Results: The models based on LC-K and All-K achieved the nonsignificantly highest accuracy in the classification of renal function (all p values > 0.05). The optimal model yielded high performance in classifying the normal function, mild or moderate impairment, and severe impairment, with an area under the curve of 0.938 (95% confidence interval [CI] 0.935–0.940), 0.919 (95%CI 0.916–0.922), and 0.959 (95%CI 0.956–0.962) in the training cohorts, respectively, as well as 0.802 (95%CI 0.800–0.807), 0.852 (95%CI 0.846–0.857), and 0.863 (95%CI 0.857–0.887) in the validation cohorts, respectively. Conclusion: We developed and internally validated an MRI-based machine-learning model that can accurately evaluate renal function. Once externally validated, this model has the potential to facilitate the monitoring of patients with impaired renal function. Key points: Texture analysis based on coronal T2-weighted MR images could evaluate the renal function in patients with diabetes. The All-K and LC-K outperformed other segmentation methods in the evaluation of renal function impairment. The segmentation methods could affect the results of renal function evaluation and the integrity of the coronal slices was crucial for renal imaging texture analysis. [ABSTRACT FROM AUTHOR]

Objectives: To evaluate the predictive value of intratumoral and peritumoral radiomics and radiomics nomogram for preoperative lymphovascular invasion (LVI) status and overall survival (OS) in patients with non-small cell lung cancer (NSCLC). Methods: In total, 240 NSCLC patients from our institution were randomly divided into the training cohort (n = 145) and internal validation cohort (n = 95) with a ratio of 6:4, and 65 patients from the Cancer Imaging Archive were enrolled as the external validation cohort. We extracted 1217 CT-based radiomics features from the gross tumor volume (GTV) and gross tumor volume incorporating peritumoral 3, 6, and 9 mm regions (GPTV3, GPTV6, GPTV9). A radiomics nomogram based on clinical independent predictors and radiomics score (Radscore) of the best radiomics model was constructed. The correlation between factors and OS was evaluated with the Kaplan-Meier survival analysis and Cox proportional hazards regression analysis. Results: Compared with GTV, GPTV3, and GPTV6 radiomics models, GPTV9 radiomics model exhibited better prediction performance with the AUCs of 0.82, 0.75, and 0.67 in the training, internal validation, and external validation cohorts, respectively. In the clinical model, smoking and clinical stage were independent predictors. The nomogram incorporating independent predictors and GPTV9-Radscore was clinically useful, with the AUCs of 0.89, 0.83, and 0.66 in three cohorts. Pathological LVI, GPTV9-Radscore-predicted, and Nomoscore-predicted LVI were associated with poor OS (p < 0.05). Conclusions: CT-based radiomics nomogram can predict LVI and OS in patients with NSCLC and may help in making personalized treatment strategies before surgery. Key Points: • Compared with GTV, GPTV3, and GPTV6radiomics models, GPTV9radiomics model showed better prediction performance for LVI status in NSCLC. • The radiomics nomogram based on GPTV9radiomics features and clinical independent predictors could effectively predict LVI status and OS in NSCLC and outperformed the clinical model. • The radiomics nomogram had a wider scope of clinical application. [ABSTRACT FROM AUTHOR]

Current risk stratification systems for thyroid nodules suffer from low specificity and high biopsy rates. Recently, machine learning (ML) is introduced to assist thyroid nodule diagnosis but lacks interpretability. Here, we developed and validated ML models on 3965 thyroid nodules, as compared to the American College of Radiology Thyroid Imaging, Reporting and Data System (ACR TI‐RADS). Subsequently, a SHapley Additive exPlanation (SHAP) algorithm was leveraged to interpret the results of the best‐performing ML model. Clinical characteristics including thyroid‐function tests were collected from medical records. Five ACR TI‐RADS ultrasonography (US) categories plus nodule size were assessed by experienced radiologists. Random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGBoost) were used to build US‐only and US‐clinical ML models. The ML models and ACR TI‐RADS were compared in terms of diagnostic performance and unnecessary biopsy rate. Among the ML models, the US‐only RF model (hereafter, Thy‐Wise) achieved the optimal performance. Compared to ACR TI‐RADS, Thy‐Wise showed higher accuracy (82.4% vs 74.8% for the internal validation; 82.1% vs 73.4% for external validation) and specificity (78.7% vs 68.3% for internal validation; 78.5% vs 66.9% for external validation) while maintaining sensitivity (91.7% vs 91.2% for internal validation; 91.9% vs 91.1% for external validation), as well as reduced unnecessary biopsies (15.3% vs 32.3% for internal validation; 15.7% vs 47.3% for external validation). The SHAP‐based interpretation of Thy‐Wise enables clinicians to better understand the reasoning behind the diagnosis, which may facilitate the clinical translation of this model. [ABSTRACT FROM AUTHOR]

Selective laser melting (SLM) is the most used and versatile technique for additively manufactured metallic parts. Due to its unique flexibility in enabling the fabrication of complex geometries, it has received great attention, especially in very demanding industries such as aerospace and automotive. Improving this process requires a better understanding of laser-metal interactions, something that can be achieved by relying on numerical modelling as a cost-effective alternative to experiments. This review focuses on laser-metal-powder interactions, with an emphasis on powder-scale modelling and simulation, to elucidate the powder-part relationship. Through a comprehensive meta-analysis of the physical processes and the existing numerical modelling methodologies, this review exposes the advancements in both virtual powder bed generation and ways of carrying out multi-physics computations. Process optimisation with meta-modelling and machine learning are also critically discussed, with clear evidence of their capabilities and limitations. [ABSTRACT FROM AUTHOR]

Carbon market is the most effective market tool for carbon emission reduction. China, the largest carbon emitter in the world, established the national carbon market in 2021, covering over 2000 key units in the power sector. Therefore, the forecasting of carbon price has profound implications for environmental and energy policies. In this paper, two traditional econometric model and three kinds of machine learning (ML) algorithms are used to predict carbon returns in the Chinese carbon trading market based on carefully selected predictors. Among all forecasting models, the Random Forest (RF) has the best forecasting performance, followed by some GARCH models using various factors. Compared with the traditional benchmark of ARMA model, the BP neural network (BP) and the GA improved BP method (GA-BP) are less competitive in predicting carbon returns in China because of their large forecasting errors. According to the optimal model, social attention to the carbon market, the international crude oil returns and the overall performance of the stock market are the most important predictors. The findings are robust to the change in the sample set. Overall, the ML approach shows an advantage in forecasting carbon returns, but the selection of predictors is also important. • Machine learning and econometric models are used to forecast carbon returns in China. • The selection of predictors is based on economic analysis. • The Random Forest (RF) outperforms the other ML and econometric methods. • The energy market is extremely important for predicting carbon returns. • Social concerns and stock market performance improve forecasting accuracy. [ABSTRACT FROM AUTHOR]