Search

Objective: To analyze the radiomic and clinical features extracted from 2D ultrasound images of thyroid tumors in patients with Hashimoto's thyroiditis (HT) combined with papillary thyroid carcinoma (PTC) using machine learning (ML) models, and to explore the diagnostic performance of the method in making preoperative noninvasive identification of cervical lymph node metastasis (LNM)., Methods: A total of 528 patients with HT combined with PTC were enrolled and divided into two groups based on their pathological results of the presence or absence of LNM. The groups were subsequently designated the With LNM Group and the Without LNM Group. Three ultrasound doctors independently delineated the regions of interest and extracted radiomic features. Two modes, radiomic features and radiomics-clinical features, were used to construct random forest (RF), support vector machine (SVM), LightGBM, K-nearest neighbor (KNN), and XGBoost models. The performance of these five ML models in the two modes was evaluated by the receiver operating characteristic (ROC) curves on the test dataset, and SHapley Additive exPlanations (SHAP) was used for model visualization., Results: All five ML models showed good performance, with area under the ROC curve (AUC) ranging from 0.798 to 0.921. LightGBM and XGBoost demonstrated the best performance, outperforming the other models ( P <0.05). The ML models constructed with radiomics-clinical features performed better than those constructed using only radiomic features ( P <0.05). The SHAP visualization of the best-performing models indicated that the anteroposterior diameter, superoinferior diameter, original&#95;shape&#95;VoxelVolume, age, wavelet-LHL&#95;firstorder&#95;10Percentile, and left-to-right diameter had the most significant effect on the LightGBM model. On the other hand, the superoinferior diameter, anteroposterior diameter, left-to-right diameter, original&#95;shape&#95;VoxelVolume, original&#95;firstorder&#95;InterquartileRange, and age had the most significant effect on the XGBoost model., Conclusion: ML models based on radiomics and clinical features can accurately evaluate the cervical lymph node status in patients with HT combined with PTC. Among the 5 ML models, LightGBM and XGBoost demonstrate the best evaluation performance., Competing Interests: 利益冲突　所有作者均声明不存在利益冲突, (© 2024《四川大学学报（医学版）》编辑部 版权所有Copyright ©2024 Editorial Office of Journal of Sichuan University (Medical Sciences).)

Background: Acute Kidney Injury (AKI) is a shared complication among Intensive Care Unit (ICU), marked by high cost, high morbidity and high mortality. As the early prediction of AKI is critical for patients' outcomes and data mining is such a powerful prediction tool, many AKI prediction models based on machine learning methods have been proposed. Our motivation is inspired by the fact that the incidence of AKI is a changing temporal sequence affected by the joint action of patients' daily drug combinations and their physiological indexes. However, most existing models have not considered such a temporal correlation. Besides, due to great challenges caused by sparse, high-dimensional and highly imbalanced clinical data, it is hard to achieve ideal performance., Methods: We develop a fast, simple and less-costly model based on an ensemble learning algorithm, named Ensemble Time Series Model (ETSM). Besides benefiting from vital signs and laboratory results as explicit indicators, ETSM explores the effect of drug combinations as possible implicit indicators for the AKI prediction. The model transforms temporal medication information into a multidimensional vector to consider and measure drug cumulative effects that may cause AKI., Results: We compare ETSM with state-of-the-art models on ICUC and MIMIC III datasets. On the basis of the experimental results, our model obtains satisfactory performance (ICUC: AUC 24 hours ahead: 0.81, 48 hours ahead: 0.78; MIMIC III: AUC 24 hours ahead: 0.95, 48 hours ahead: 0.95). Meanwhile, we compare the effects of different sampling and feature generation methods on the model performance. In the ablation study, we validate that medication information improves model performance (24 hours ahead: AUC increased from 0.74 to 0.81). We also find that the model's performance is closely related to the balanced level of the derivation dataset. The optimal ratio of major class size to minor class size for the model is found for AKI prediction., Conclusions: ETSM is an effective method for the early prediction of AKI. The model verifies that AKI incidence is related to the clinical medication. In comparison with other prediction methods, ETSM provides comparable performance results and better interpretability.

Purpose: To enhance clinician's decision-making by diagnosing hepatocellular carcinoma (HCC) in cirrhotic patients with indeterminate liver nodules using quantitative imaging features extracted from triphasic CT scans., Material and Methods: We retrospectively analyzed 178 cirrhotic patients from 27 institutions, with biopsy-proven liver nodules classified as indeterminate using the European Association for the Study of the Liver (EASL) guidelines. Patients were randomly assigned to a discovery cohort (142 patients (pts.)) and a validation cohort (36 pts.). Each liver nodule was segmented on each phase of triphasic CT scans, and 13,920 quantitative imaging features (12 sets of 1160 features each reflecting the phenotype at one single phase or its change between two phases) were extracted. Using machine-learning techniques, the signature was trained and calibrated (discovery cohort), and validated (validation cohort) to classify liver nodules as HCC vs. non-HCC. Effects of segmentation and contrast enhancement quality were also evaluated., Results: Patients were predominantly male (88%) and CHILD A (65%). Biopsy was positive for HCC in 77% of patients. LI-RADS scores were not different between HCC and non-HCC patients. The signature included a single radiomics feature quantifying changes between arterial and portal venous phases: Delta V-A &#95; DWT1&#95;LL&#95;Variance-2D and reached area under the receiver operating characteristic curve (AUC) of 0.70 (95%CI 0.61-0.80) and 0.66 (95%CI 0.64-0.84) in discovery and validation cohorts, respectively. The signature was influenced neither by segmentation nor by contrast enhancement., Conclusion: A signature using a single feature was validated in a multicenter retrospective cohort to diagnose HCC in cirrhotic patients with indeterminate liver nodules. Artificial intelligence could enhance clinicians' decision by identifying a subgroup of patients with high HCC risk., Key Points: • In cirrhotic patients with visually indeterminate liver nodules, expert visual assessment using current guidelines cannot accurately differentiate HCC from differential diagnoses. Current clinical protocols do not entail biopsy due to procedural risks. Radiomics can be used to non-invasively diagnose HCC in cirrhotic patients with indeterminate liver nodules, which could be leveraged to optimize patient management. • Radiomics features contributing the most to a better characterization of visually indeterminate liver nodules include changes in nodule phenotype between arterial and portal venous phases: the "washout" pattern appraised visually using EASL and EASL guidelines. • A clinical decision algorithm using radiomics could be applied to reduce the rate of cirrhotic patients requiring liver biopsy (EASL guidelines) or wait-and-see strategy (AASLD guidelines) and therefore improve their management and outcome.

Background: Motion-onset visual evoked potentials (mVEP) can provide a softer stimulus with reduced fatigue, and it has potential applications for brain computer interface(BCI)systems. However, the mVEP waveform is seriously masked in the strong background EEG activities, and an effective approach is needed to extract the corresponding mVEP features to perform task recognition for BCI control., New Method: In the current study, we combine deep learning with compressed sensing to mine discriminative mVEP information to improve the mVEP BCI performance., Results: The deep learning and compressed sensing approach can generate the multi-modality features which can effectively improve the BCI performance with approximately 3.5% accuracy incensement over all 11 subjects and is more effective for those subjects with relatively poor performance when using the conventional features., Comparison With Existing Methods: Compared with the conventional amplitude-based mVEP feature extraction approach, the deep learning and compressed sensing approach has a higher classification accuracy and is more effective for subjects with relatively poor performance., Conclusions: According to the results, the deep learning and compressed sensing approach is more effective for extracting the mVEP feature to construct the corresponding BCI system, and the proposed feature extraction framework is easy to extend to other types of BCIs, such as motor imagery (MI), steady-state visual evoked potential (SSVEP)and P300., (Copyright © 2016 Elsevier B.V. All rights reserved.)

Flame combustion diagnostics is a technique that uses different methods to diagnose the flame combustion process and study its physical and chemical basis. As one of the most important parameters of the combustion process, the flame equivalence ratio has a significant influence on the entire flame combustion, especially on the combustion efficiency and the emission of pollutants. Therefore, the measurement of the flame equivalence ratio has a huge impact on efficient combustion and environment protection. In view of this, several effective measuring methods were proposed, which were based on the different characteristics of flames radicals such as spectral properties. With the rapid growth of machine learning, more and more scholars applied it in the combustion diagnostics due to the excellent ability to fit parameters. This paper presents a review of various measuring techniques of hydrocarbon flame equivalent ratio and the applications of machine learning in combustion diagnostics, finally making a brief comparison between different measuring methods.

Radio frequency fingerprinting (RFF) is the concept arising from classification of wireless emitters due to their unique radio frequency features. RFF has been further extended to applications including both RF devices classification and wireless signal identification. In this paper, we adopt Gaussian Mixture Models (GMM) technique as feature extraction approach and firstly apply it to extract RFF of antennas. 9 classical antennas with 3 different load conditions (open, short, match) were studied in our experiment. Moreover, we also made a theoretical analysis about the reason scattered signal has the unique features. Specifically, we adopt the Random Noise Radar (RNR) technique to obtain reflected RF signals of antenna under test (AUT) and apply the GMM technique to fit RF signals and then extract the RFF of AUT. A support vector machine (SVM) is proposed to recognize the RFF at different signal-to-noise ratio (SNR) environment. Compared with the conventional feature extraction approaches, for example, from variance, skewness and kurtosis (VSK) values, our method demonstrates better performance on large datasets with classification accuracy above 88% using a SVM classifier. Moreover, the accuracy remains higher than 75% even when the Signal to Noise Ratio (SNR) is equal to 0dB, indicating that the proposed approach has the strong capability of noise immunity.

In the area of radar imaging at any frequency band from microwave to optics, the technique of compressive sensing (CS) enables high resolution with reduced number of antenna elements and measurements. However, CS methods suffer from high computational complexity and require parameter tuning to ensure good image reconstruction under different noise, sparsity and undersampling levels. To alleviate such issues, we present a machine learning approach that combines CS and convolutional neural network (CNN) for radar imaging. This CS based CNN (CS-CNN) method maintains good characteristics of CS methods, such as sparse sampling and high resolving power but is free from time-consuming computer optimization and demanding spaces for data storage. In the meantime, it is also robust to environment changes like noise, target sparsity and sampling rate. We have conducted extensive computer simulations for both qualitative and quantitative evaluations. Finally, we experimentally validate the technique with a demonstration of stable high resolution imaging using a sparse multiple-input multiple-output (MIMO) array where traditional imaging methods suffer from serious grating lobes. This approach is generic and can be easily extended to other applications of electromagnetic imaging and sensing.

Maize, an important cereal crop and crucial industrial material, is widely used in various fields, including food, feed, and industry. Maize is also a highly adaptable crop, capable of thriving under various climatic and soil conditions. Against the backdrop of intensified climate change, studying the classification of maize growth stages can aid in adjusting planting strategies to enhance yield and quality. Accurate classification of the growth stages of maize breeding materials is important for enhancing yield and quality in breeding endeavors. Traditional remote sensing-based crop growth stage classifications mainly rely on time series vegetation index (VI) analyses; however, VIs are prone to saturation under high-coverage conditions. Maize phenotypic traits at different growth stages may improve the accuracy of crop growth stage classifications. Therefore, we developed a method for classifying maize growth stages during the vegetative growth phase by combining maize phenotypic traits with different classification algorithms. First, we tested various VIs, texture features (TFs), and combinations of VI and TF as input features to estimate the leaf chlorophyll content (LCC), leaf area index (LAI), and fractional vegetation cover (FVC). We determined the optimal feature inputs and estimation methods and completed crop height (CH) extraction. Then, we tested different combinations of maize phenotypic traits as input variables to determine their accuracy in classifying growth stages and to identify the optimal combination and classification method. Finally, we compared the proposed method with traditional growth stage classification methods based on remote sensing VIs and machine learning models. The results indicate that (1) when the VI+TFs are used as input features, random forest regression (RFR) shows a good estimation performance for the LCC (R2: 0.920, RMSE: 3.655 SPAD units, MAE: 2.698 SPAD units), Gaussian process regression (GPR) performs well for the LAI (R2: 0.621, RMSE: 0.494, MAE: 0.397), and linear regression (LR) exhibits a good estimation performance for the FVC (R2: 0.777, RMSE: 0.051, MAE: 0.040); (2) when using the maize LCC, LAI, FVC, and CH phenotypic traits to classify maize growth stages, the random forest (RF) classification method achieved the highest accuracy (accuracy: 0.951, precision: 0.951, recall: 0.951, F1: 0.951); and (3) the effectiveness of the growth stage classification based on maize phenotypic traits outperforms that of traditional remote sensing-based crop growth stage classifications. [ABSTRACT FROM AUTHOR]

Contrast-enhanced computed tomography scans (CECT) are routinely used in the evaluation of different clinical scenarios, including the detection and characterization of hepatocellular carcinoma (HCC). Quantitative medical image analysis has been an exponentially growing scientific field. A number of studies reported on the effects of variations in the contrast enhancement phase on the reproducibility of quantitative imaging features extracted from CT scans. The identification and labeling of phase enhancement is a time-consuming task, with a current need for an accurate automated labeling algorithm to identify the enhancement phase of CT scans. In this study, we investigated the ability of machine learning algorithms to label the phases in a dataset of 59 HCC patients scanned with a dynamic contrast-enhanced CT protocol. The ground truth labels were provided by expert radiologists. Regions of interest were defined within the aorta, the portal vein, and the liver. Mean density values were extracted from those regions of interest and used for machine learning modeling. Models were evaluated using accuracy, the area under the curve (AUC), and Matthew's correlation coefficient (MCC). We tested the algorithms on an external dataset (76 patients). Our results indicate that several supervised learning algorithms (logistic regression, random forest, etc.) performed similarly, and our developed algorithms can accurately classify the phase of contrast enhancement. [ABSTRACT FROM AUTHOR]

Objectives: To evaluate the performance of automatic deep learning (DL) algorithm for size, mass, and volume measurements in predicting prognosis of lung adenocarcinoma (LUAD) and compared with manual measurements. Methods: A total of 542 patients with clinical stage 0-I peripheral LUAD and with preoperative CT data of 1-mm slice thickness were included. Maximal solid size on axial image (MSSA) was evaluated by two chest radiologists. MSSA, volume of solid component (SV), and mass of solid component (SM) were evaluated by DL. Consolidation-to-tumor ratios (CTRs) were calculated. For ground glass nodules (GGNs), solid parts were extracted with different density level thresholds. The prognosis prediction efficacy of DL was compared with that of manual measurements. Multivariate Cox proportional hazards model was used to find independent risk factors. Results: The prognosis prediction efficacy of T-staging (TS) measured by radiologists was inferior to that of DL. For GGNs, MSSA-based CTR measured by radiologists (RMSSA%) could not stratify RFS and OS risk, whereas measured by DL using 0HU (2D−AIMSSA0HU%) could by using different cutoffs. SM and SV measured by DL using 0 HU (AISM0HU% and AISV0HU%) could effectively stratify the survival risk regardless of different cutoffs and were superior to 2D−AIMSSA0HU%. AISM0HU% and AISV0HU% were independent risk factors. Conclusion: DL algorithm can replace human for more accurate T-staging of LUAD. For GGNs, 2D−AIMSSA0HU% could predict prognosis rather than RMSSA%. The prediction efficacy of AISM0HU% and AISV0HU% was more accurate than of 2D−AIMSSA0HU% and both were independent risk factors. Clinical relevance statement: Deep learning algorithm could replace human for size measurements and could better stratify prognosis than manual measurements in patients with lung adenocarcinoma. Key Points: • Deep learning (DL) algorithm could replace human for size measurements and could better stratify prognosis than manual measurements in patients with lung adenocarcinoma (LUAD). • For GGNs, maximal solid size on axial image (MSSA)–based consolidation-to-tumor ratio (CTR) measured by DL using 0 HU could stratify survival risk than that measured by radiologists. • The prediction efficacy of mass- and volume-based CTRs measured by DL using 0 HU was more accurate than of MSSA-based CTR and both were independent risk factors. [ABSTRACT FROM AUTHOR]

Phthalate esters (PAEs) are widely exposed in the environment as plasticizers in plastics, and they have been found to cause significant environmental and health hazards, especially in terms of endocrine disruption in humans. In order to investigate the processes underlying the endocrine disruption effects of PAEs, three machine learning techniques were used in this study to build an adverse outcome pathway (AOP) for those effects on people. According to the results of the three machine learning techniques, the random forest and XGBoost models performed well in terms of prediction. Subsequently, sensitivity analysis was conducted to identify the initial events, key events, and key features influencing the endocrine disruption effects of PAEs on humans. Key features, such as Mol.Wt, Q+, QH+, ELUMO, minHCsats, MEDC-33, and EG, were found to be closely related to the molecular structure. Therefore, a 3D-QSAR model for PAEs was constructed, and, based on the three-dimensional potential energy surface information, it was discovered that the hydrophobic, steric, and electrostatic fields of PAEs significantly influence their endocrine disruption effects on humans. Lastly, an analysis of the contributions of amino acid residues and binding energy (BE) was performed, identifying and confirming that hydrogen bonding, hydrophobic interactions, and van der Waals forces are important factors affecting the AOP of PAEs' molecular endocrine disruption effects. This study defined and constructed a comprehensive AOP for the endocrine disruption effects of PAEs on humans and developed a method based on theoretical simulation to characterize the AOP, providing theoretical guidance for studying the mechanisms of toxicity caused by other pollutants. [ABSTRACT FROM AUTHOR]

Accurate and rapid estimation of the crop yield is essential to precision agriculture. Critical to crop improvement, yield is a primary index for selecting excellent genotypes in crop breeding. Recently developed unmanned aerial vehicle (UAV) platforms and advanced algorithms can provide powerful tools for plant breeders. Genotype category information such as the maturity group information (M) can significantly influence soybean yield estimation using remote sensing data. The objective of this study was to improve soybean yield prediction by combining M with UAV-based multi-sensor data using machine learning methods. We investigated three types of maturity groups (Early, Median and Late) of soybean, and collected the UAV-based hyperspectral and red–green–blue (RGB) images at three key growth stages. Vegetation indices (VI) and texture features (Te) were extracted and combined with M to predict yield using partial least square regression (PLSR), Gaussian process regression (GPR), random forest regression (RFR) and kernel ridge regression (KRR). The results showed that (1) the method of combining M with remote sensing data could significantly improve the estimation performances of soybean yield. (2) The combinations of three variables (VI, Te and M) gave the best estimation accuracy. Meanwhile, the flowering stage was the optimal single time point for yield estimation (R2 = 0.689, RMSE = 408.099 kg/hm2), while using multiple growth stages produced the best estimation performance (R2 = 0.700, RMSE = 400.946 kg/hm2). (3) By comparing the models constructed by different algorithms for different growth stages, it showed that the models built by GPR showed the best performances. Overall, the results of this study provide insights into soybean yield estimation based on UAV remote sensing data and maturity information. [ABSTRACT FROM AUTHOR]

The transmission rate of COVID-19 varies by location and time. A proper measure of the transmissibility of an infectious disease should be place- and time-specific, which is currently unavailable. This research aims to better understand the spatiotemporally changing transmissibility of COVID-19. It contributes to COVID-19 research in three ways. First, it presents a generally applicable modeling framework to estimate the transmissibility of COVID-19 in a specific place and time based on daily reported case data, called space-time effective reproduction number, denoted as R s t. Then, the developed model is used to create a spatiotemporal data set of R s t values at the county level in the United States. Second, it investigates relationships between R s t and dynamically changing context factors with multiple machine learning and spatial modeling techniques. The research examines the relationships from a cross-sectional perspective and a longitudinal perspective separately. The longitudinal view allows us to understand how local human dynamics and policy factors influence changes in R s t over time in the place, whereas the cross-sectional view sheds light on the demographic, socioeconomic, and environmental factors behind spatial variations of R s t at a specific time slice. Some general trends of the relationships are found, but the level of impact by each context factor varies geographically. Third, the best performing local longitudinal models have promising potential to simulate or forecast future transmissibility. The random forest and the exponential regression models based on time-series data gave the best performances. These models were further evaluated against ground truth data of county-level reported cases. Their good prediction accuracies in the case study prove that these machine learning models are promising in their ability to predict transmissibility in hypothetical or foreseeable scenarios. [ABSTRACT FROM AUTHOR]

The moisture content of cigar leaves during drying is an important indicator for controlling the management of drying rooms. At present, the determination of cigar leaf moisture content is mainly dependent on traditional destructive detection methods, which are inefficient and damaging to plants. In this study, a Convolution Neural Network method consisting of digital images for monitoring the moisture content of cigar leaves during the drying process was proposed. In this study, the Convolution Neural Network model was trained to learn the relationship between the images and the corresponding moisture content using the extracted colour, shape, and texture features as input factors. In order to compare the Convolution Neural Network estimation results, a widely used traditional machine learning algorithm was applied. The results demonstrated that the estimated value of Convolution Neural Network agreed with the predicted value; the R² was 0.9044, and the average accuracy was 87.34%. These results were better than those produced by traditional machine learning methods. The generalization test of the proposed method was conducted using varieties of cigar leaves in other drying rooms. The results showed that Convolution Neural Network is a viable method for an accurate estimation of the moisture content, the R² was 0.8673 and the average accuracy was 86.81%. The Convolution Neural Network established by the features extracted from digital images could accurately estimate the moisture content of cigar leaves during drying and was therefore shown to be an effective monitoring tool. [ABSTRACT FROM AUTHOR]

The acquisition of maize tassel phenotype information plays a vital role in studying maize growth and improving yield. Unfortunately, detecting maize tassels has proven challenging because of the complex field environment, including image resolution, varying sunlight conditions, plant varieties, and planting density. To address this situation, the present study uses unmanned aerial vehicle (UAV) remote sensing technology and a deep learning algorithm to facilitate maize tassel identification and counting. UAVs are used to collect maize tassel images in experimental fields, and RetinaNet serves as the basic model for detecting maize tassels. Small maize tassels are accurately identified by optimizing the feature pyramid structure in the model and introducing attention mechanisms. We also study how mapping differences in image resolution, brightness, plant variety, and planting density affect the RetinaNet model. The results show that the improved RetinaNet model is significantly better at detecting maize tassels than the original RetinaNet model. The average precision in this study is 0.9717, the precision is 0.9802, and the recall rate is 0.9036. Compared with the original model, the improved RetinaNet improves the average precision, precision, and recall rate by 1.84%, 1.57%, and 4.6%, respectively. Compared with mainstream target detection models such as Faster R-CNN, YOLOX, and SSD, the improved RetinaNet model more accurately detects smaller maize tassels. For equal-area images of differing resolution, maize tassel detection becomes progressively worse as the resolution decreases. We also analyze how detection depends on brightness in the various models. With increasing image brightness, the maize tassel detection worsens, especially for small maize tassels. This paper also analyzes the various models for detecting the tassels of five maize varieties. Zhengdan958 tassels prove the easiest to detect, with R2 = 0.9708, 0.9759, and 0.9545 on 5, 9, and 20 August 2021, respectively. Finally, we use the various models to detect maize tassels under different planting densities. At 29,985, 44,978, 67,466, and 89,955 plants/hm2, the mean absolute errors for detecting Zhengdan958 tassels are 0.18, 0.26, 0.48, and 0.63, respectively. Thus, the detection error increases gradually with increasing planting density. This study thus provides a new method for high-precision identification of maize tassels in farmland and is especially useful for detecting small maize tassels. This technology can be used for high-throughput investigations of maize phenotypic traits. [ABSTRACT FROM AUTHOR]

Synthetic musks (SMs), as an indispensable odor additive, are widely used in various personal care products. However, due to their physico-chemical properties, SMs were detected in various environmental media, even in samples from arctic regions, leading to severe threats to human health (e.g., abortion risk). Environmentally friendly and functionally improved SMs have been theoretically designed in previous studies. However, the synthesizability of these derivatives has barely been proven. Thus, this study developed a method to verify the synthesizability of previously designed SM derivatives using machine learning, 2D-QSAR, 3D-QSAR, and high-throughput density functional theory in order to screen for synthesizable, high-performance (odor sensitivity), and environmentally friendly SM derivatives. In this study, three SM derivatives (i.e., D52, D37, and D25) were screened and recommended due to their good performances (i.e., high synthesizability and odor sensitivity; low abortion risk; and bioaccumulation ability in skin keratin). In addition, the synthesizability mechanism of SM derivatives was also analyzed. Results revealed that high intramolecular hydrogen bond strength, electrostatic interaction, qH+ value, energy gap, and low EHOMO would lead to a higher synthesizability of SMs and their derivatives. This study broke the synthesizability bottleneck of theoretically designed environment-friendly SM derivatives and advanced the mechanism of screening functional derivatives. [ABSTRACT FROM AUTHOR]

Three-dimensional (3D) printing has stood out as a reliable technology to construct carbon microlattice electrodes for supercapacitors (SCs) in the field of custom areal electrochemical performance needed. The complex structural parameters of 3D-printed (3DP) electrodes make customization of the 3DP electrodes low efficient and time-consuming. Herein, we integrate machine learning (ML) to deeply unravel the influence of typical structural parameters of 3DP electrodes made of graphene and carbon nanotubes (CNTs). The dependence of areal performance on electrode structures was established through selecting only 9 experimental points combined with random forest (RF) algorithm, where the valuable information and the hidden correlations were quickly extracted. By using the as-established model, the electrodes with desired performance could be printed based on structural parameters directly selected from the model, offering an essentially improved performance for target performance. Specifically, the areal performance could be tuned from 0.032 to 1.6 F cm−2, covering an extremely large range. Moreover, the electrochemical surface area (ECSA) and finite element analysis (FEA) were employed to analyze the dependence of areal performance on structural parameters, agreeing well with the model information. The idea proposed in this work could largely increase the efficiency of developing new electrode architectures for desired performance. [Display omitted] A machine learning framework is established to develop a prediction model between structural parameters and areal capacitance of 3DP electrodes. The well-fitted machine learning model constructed by random forest indicated that the feature importance is geometric area > electrode thickness > gap distance. Based on this model, the customization of areal capacitance could be realized, fulfilling the diverse requirements of actual applications. • 3D printing exhibits huge advantages in fabricating customized supercapacitors. • Machine learning model is established to guide 3D printing supercapacitors with desired performance. • Structural parameters influence areal capacitance by regulating electrochemical surface area and current distribution. [ABSTRACT FROM AUTHOR]

Accurate wind speed prediction plays a crucial role in wind power generation and disaster avoidance. However, stochasticity and instability increase the difficulty of wind speed prediction. In this study, we proposed a dual-attention mechanism multi-channel convolutional LSTM (DACLSTM), collected European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) near-ground element-grid data from some parts of North China, and selected elements with high correlations with wind speed to form multiple channels. We used a convolutional network for the feature extraction of spatial information, a Long Short-Term Memory (LSTM) network for the feature extraction of time-series information, and used channel attention with spatial attention for feature extraction. The experimental results show that the DACLSTM model can improve the accuracy of six-hour lead time wind speed prediction relative to the traditional ConvLSTM model and fully connected network long short-term memory (FC_LSTM). [ABSTRACT FROM AUTHOR]

Accurate and reliable runoff prediction is essential for water resources management. In this paper, a hybrid model STL-VMD-SFO-ESN which combines seasonal adjustment method (STL), variational mode decomposition (VMD), echo state network (ESN) and sailed fish optimizer (SFO) is proposed for daily runoff prediction. Daily runoff data from three different runoff monitoring stations in China's Yellow River basin is used to evaluate the performance of proposed model and other newly reported models. The results indicates that: (1) The proposed model performs significantly better than the traditional data-driven models and some newly reported models. (2) STL decomposition can effectively remove the seasonal component of runoff and improve modeling accuracy. (3) ESN has a strong potential in runoff prediction, and its performance can be greatly improved by using bio-optimization algorithms. Thus, this new model has strong potential for runoff prediction for further application. [ABSTRACT FROM AUTHOR]

Rapidly developing remote sensing techniques are shedding new light on large-scale crop growth status monitoring, especially in recent applications of unmanned aerial vehicles (UAVs). Many inversion models have been built to estimate crop growth variables. However, the present methods focused on building models for each single crop stage, and the features generally used in the models are vegetation indices (VI) or joint VI with data derived from UAV-based sensors (e.g., texture, RGB color information, or canopy height). It is obvious these models are either limited to a single stage or have an unstable performance across stages. To address these issues, this study selected four key wheat growth parameters for inversion: above-ground biomass (AGB), plant nitrogen accumulation (PNA) and concentration (PNC), and the nitrogen nutrition index (NNI). Crop data and multispectral data were acquired in five wheat growth stages. Then, the band reflectance and VI were obtained from multispectral data, along with the five stages that were recorded as phenology indicators (PIs) according to the stage of Zadok's scale. These three types of data formed six combinations (C1–C6): C1 used all of the band reflectances, C2 used all VIs, C3 used bands and VIs, C4 used bands and PIs, C5 used VIs and PIs, and C6 used bands, Vis, and PIs. Some of the combinations were integrated with PIs to verify if PIs can improve the model accuracy. Random forest (RF) was used to build models with combinations of different parameters and evaluate the feature importance. The results showed that all models of different combinations have good performance in the modeling of crop parameters, such as R2 from 0.6 to 0.79 and NRMSE from 10.51 to 15.83%. Then, the model was optimized to understand the importance of PIs. The results showed that the combinations that integrated PIs showed better estimations and the potential of using PIs to minimize features while still achieving good predictions. Finally, the varied model results were evaluated to analyze their performances in different stages or fertilizer treatments. The results showed the models have good performances at different stages or treatments (R2 > 0.6). This paper provides a reference for monitoring and estimating wheat growth parameters based on UAV multispectral imagery and phenology information. [ABSTRACT FROM AUTHOR]

As a key functional trait, leaf photosynthetic pigment content (LPPC) plays an important role in the health status monitoring and yield estimation of apples. Hyperspectral features including vegetation indices (VIs) and derivatives are widely used in retrieving vegetation biophysical parameters. The fractional derivative spectral method shows great potential in retrieving LPPC. However, the performance of fractional derivatives and machine learning (ML) for retrieving apple LPPC still needs to be explored. The objective of this study is to test the capacity of using fractional derivative and ML methods to retrieve apple LPPC. Here, the hyperspectral data in the 400–2500 nm domains was used to calculate the fractional derivative order of 0.2–2, and then the sensitive bands were screened through feature dimensionality reduction to train ML to build the LPPC estimation model. Additionally, VIs-based ML methods and empirical regression models were developed to compare with the fractional derivative methods. The results showed that fractional derivative-driven ML methods have higher accuracy than the ML methods driven by the original spectra or vegetation index. The results also showed that the ML methods perform better than empirical regression models. Specifically, the best estimates of chlorophyll content and carotenoid content were achieved using support vector regression (SVR) at the derivative order of 0.2 (R2 = 0.78) and 0.4 (R2 = 0.75), respectively. The fractional derivative maintained a good universality in retrieving the LPPC of multiple phenological periods. Therefore, this study highlights that the fractional derivative and ML improved the estimation of apple LPPC. [ABSTRACT FROM AUTHOR]

Multi-label learning deals with training examples each represented by a single instance while associated with multiple class labels. Due to the exponential number of possible label sets to be considered by the predictive model, it is commonly assumed that label correlations should be well exploited to design an effective multi-label learning approach. On the other hand, class-imbalance stands as an intrinsic property of multi-label data which significantly affects the generalization performance of the multi-label predictive model. For each class label, the number of training examples with positive labeling assignment is generally much less than those with negative labeling assignment. To deal with the class-imbalance issue for multi-label learning, a simple yet effective class-imbalance aware learning strategy called cross-coupling aggregation (COCOA) is proposed in this article. Specifically, COCOA works by leveraging the exploitation of label correlations as well as the exploration of class-imbalance simultaneously. For each class label, a number of multiclass imbalance learners are induced by randomly coupling with other labels, whose predictions on the unseen instance are aggregated to determine the corresponding labeling relevancy. Extensive experiments on 18 benchmark datasets clearly validate the effectiveness of COCOA against state-of-the-art multi-label learning approaches especially in terms of imbalance-specific evaluation metrics. [ABSTRACT FROM AUTHOR]

Diabetic nephropathy (DN) is regarded as the leading cause of end-stage renal disease worldwide and lacks novel therapeutic targets. To screen and verify special biomarkers for glomerular injury in patients with DN, fifteen datasets were retrieved from the Gene Expression Omnibus (GEO) database, correspondingly divided into training and testing cohorts and then merged. Using the limma package, 140 differentially expressed genes (DEGs) were screened out between 81 glomerular DN samples and 41 normal ones from the training cohort. With the help of the ConsensusClusterPlus and WGCNA packages, the 81 glomerular DN samples were distinctly divided into two subclusters, and two highly associated modules were identified. By using machine learning algorithms (LASSO, RF, and SVM-RFE) and the Venn diagram, two overlapping genes (PRKAR2B and TGFBI) were finally determined as potential biomarkers, which were further validated in external testing datasets and the HFD/STZ-induced mouse models. Based on the biomarkers, the diagnostic model was developed with reliable predictive ability for diabetic glomerular injury. Enrichment analyses indicated the apparent abnormal immune status in patients with DN, and the two biomarkers played an important role in the immune microenvironment. The identified biomarkers demonstrated a meaningful correlation between the immune cells' infiltration and renal function. In conclusion, two robust genes were identified as diagnostic biomarkers and may serve as potential targets for therapeutics of DN, which were closely associated with multiple immune cells. [ABSTRACT FROM AUTHOR]

The development of Internet and network applications has brought the development of encrypted communication technology. But on this basis, malicious traffic also uses encryption to avoid traditional security protection and detection. Traditional security protection and detection methods cannot accurately detect encrypted malicious traffic. In recent years, the rise of artificial intelligence allows us to use machine learning and deep learning methods to detect encrypted malicious traffic without decryption, and the detection results are very accurate. At present, the research on malicious encrypted traffic detection mainly focuses on the characteristics' analysis of encrypted traffic and the selection of machine learning algorithms. In this paper, a method combining natural language processing and machine learning is proposed; that is, a detection method based on TF-IDF is proposed to build a detection model. In the process of data preprocessing, this method introduces the natural language processing method, namely, the TF-IDF model, to extract data information, obtain the importance of keywords, and then reconstruct the characteristics of data. The detection method based on the TF-IDF model does not need to analyze each field of the data set. Compared with the general machine learning data preprocessing method, that is, data encoding processing, the experimental results show that using natural language processing technology to preprocess data can effectively improve the accuracy of detection. Gradient boosting classifier, random forest classifier, AdaBoost classifier, and the ensemble model based on these three classifiers are, respectively, used in the construction of the later models. At the same time, CNN neural network in deep learning is also used for training, and CNN can effectively extract data information. Under the condition that the input data of the classifier and neural network are consistent, through the comparison and analysis of various methods, the accuracy of the one-dimensional convolutional network based on CNN is slightly higher than that of the classifier based on machine learning. [ABSTRACT FROM AUTHOR]

Purpose: We aimed to explore potential confounders of prognostic radiomics signature predicting survival outcomes in clear cell renal cell carcinoma (ccRCC) patients and demonstrate how to control for them. Materials and Methods: Preoperative contrast enhanced abdominal CT scan of ccRCC patients along with pathological grade/stage, gene mutation status, and survival outcomes were retrieved from The Cancer Imaging Archive (TCIA)/The Cancer Genome Atlas—Kidney Renal Clear Cell Carcinoma (TCGA-KIRC) database, a publicly available dataset. A semi-automatic segmentation method was applied to segment ccRCC tumors, and 1,160 radiomics features were extracted from each segmented tumor on the CT images. Non-parametric principal component decomposition (PCD) and unsupervised hierarchical clustering were applied to build the radiomics signature models. The factors confounding the radiomics signature were investigated and controlled sequentially. Kaplan–Meier curves and Cox regression analyses were performed to test the association between radiomics signatures and survival outcomes. Results: 183 patients of TCGA-KIRC cohort with available imaging, pathological, and clinical outcomes were included in this study. All 1,160 radiomics features were included in the first radiomics signature. Three additional radiomics signatures were then modelled in successive steps removing redundant radiomics features first, removing radiomics features biased by CT slice thickness second, and removing radiomics features dependent on tumor size third. The final radiomics signature model was the most parsimonious, unbiased by CT slice thickness, and independent of tumor size. This final radiomics signature stratified the cohort into radiomics phenotypes that are different by cancer-specific and recurrence-free survival; HR (95% CI) = 3.0 (1.5–5.7), p <0.05 and HR (95% CI) = 6.6 (3.1–14.1), p <0.05, respectively. Conclusion: Radiomics signature can be confounded by multiple factors, including feature redundancy, image acquisition parameters like slice thickness, and tumor size. Attention to and proper control for these potential confounders are necessary for a reliable and clinically valuable radiomics signature. [ABSTRACT FROM AUTHOR]

Accurate segmentation of ore images plays a significant role in automatic geometric parameter detecting and composition analyzing of ore dressing progress. Semantic segmentation based on deep learning is a promising method for accurate ore image segmentation. However, the similar appearance with low contrast and blurry boundary of ores in image hamper segmentation accuracy. Moreover, it is difficult to train a deep network due to limited available ore images. In this work, an improved encoder-decoder network based on U-Net is proposed to handle above challenges. A contour awareness loss (CAL) is proposed to improve model sensitivity to misclassified pixels, pixels of similar appearance, and pixels near the boundary. The proposed scheme is verified on ore images by benchmarking against state-of-the-art segmentation methods. Experiment results show that the proposed scheme achieves 88.6% pixel-wise accuracy (PA) and 66.0% mean intersection over union (mIoU). [ABSTRACT FROM AUTHOR]

Various investigations have shown that driver fatigue is the main cause of traffic accidents. Research on the use of computer vision techniques to detect signs of fatigue from facial actions, such as yawning, has demonstrated good potential. However, accurate and robust detection of yawning is difficult because of the complicated facial actions and expressions of drivers in the real driving environment. Several facial actions and expressions have the same mouth deformation as yawning. Thus, a novel approach to detecting yawning based on subtle facial action recognition is proposed in this study to alleviate the abovementioned problems. A 3D deep learning network with a low time sampling characteristic is proposed for subtle facial action recognition. This network uses 3D convolutional and bidirectional long short-term memory networks for spatiotemporal feature extraction and adopts SoftMax for classification. A keyframe selection algorithm is designed to select the most representative frame sequence from subtle facial actions. This algorithm rapidly eliminates redundant frames using image histograms with low computation cost and detects outliers by median absolute deviation. A series of experiments are also conducted on YawDD benchmark and self-collected datasets. Compared with several state-of-the-art methods, the proposed method has high yawning detection rates and can effectively distinguish yawning from similar facial actions. [ABSTRACT FROM AUTHOR]

To improve the production efficiency in the pulp and paper industry, the chemical composition of pulp wood species has to be measured in real-time, especially the holocellulose and acid insoluble lignin contents. Near infrared (NIR) spectroscopy, as a promising rapid and on-line technology, is an attractive and promising tool to determine holocellulose and lignin contents in pulp wood. Due to the high complexity and nonlinearity of the spectra of pulp wood, it is significant to select suitable chemometric methods. In this study, in order to eliminate noise and irrelevant information of the original spectra collected by a portable spectrometer, four methods were used to preprocess the original spectra, including the first derivative, moving average filtering, multiplicative scatter correction and standard normal variate transformation. Next a comparison was conducted using four modeling approaches, including partial least squares (PLS) regression, least square support vector machine (LSSVM), back-propagation neural network (BPNN), and kernel extreme learning machine (KELM). The last three approaches were calibrated using spectral features that reduced the dimensions by principal component analysis (PCA). Furthermore, regularization parameter and kernel function parameter of LSSVM and KELM were optimized by a particle swarm optimization (PSO) algorithm. The results indicated that multiplicative scatter correction efficiently eliminated the spectral noise and irrelative information, and that KELM displayed the best prediction performance compared to the other approaches. Therefore, an inexpensive and portable NIR spectrometer has been employed to accurately and efficiently determine the chemical composition of pulp wood when combined with multiplicative scatter correction and the KELM method. [ABSTRACT FROM AUTHOR]

Phthalate esters (PAEs) as plasticizers posed significant environmental and human health risks during the production and application processes. This study aims to investigate the synthesizability of previously designed environmentally friendly PAEs alternatives by in silico approaches and explore the key factors influencing their synthesizability. Firstly, a novel approach combining positive and unlabeled (PU) machine learning (ML) and k-Nearest Neighbor (kNN) models was utilized to screen three environmentally friendly PAEs alternatives with high synthesizability. Subsequently, the synthesizability and environmental friendliness of the selected PAEs alternatives were evaluated and confirmed using ADMETlab 2.0. Results demonstrated that all three PAEs alternatives exhibited high synthesizability and environmental friendliness. Finally, the SHapley Additive exPlanation (SHAP) method was employed to analyze the synthesis mechanisms of PAEs alternatives, revealing that the VE2_DzZ (molecular substitutability) and maxHCsats (molecular structure) could influence their synthesizability. Based on these findings, the synthesizability of the three PAEs alternatives was validated using density functional theory (DFT) and three-dimensional quantitative structure-activity relationship (3D-QSAR) methods. This cost-effective study provided the first evaluation of the synthesizability of PAEs alternatives and reduced the number of unlabeled samples by 95.83%. Moreover, these findings offered a theoretical foundation for developing environmentally friendly PAEs alternatives and provided new insights for substituting emerging pollutants and developing new materials. [Display omitted] • First study on synthesizability of environmentally friendly PAEs alternatives. • PU learning was firstly used to predict the synthesizability of PAEs alternatives. • High synthesizability of three PAEs alternatives was found and verified. • Mechanism for the synthesizability of PAEs alternatives was analyzed. [ABSTRACT FROM AUTHOR]

Motivation De novo peptide sequencing based on tandem mass spectrometry data is the key technology of shotgun proteomics for identifying peptides without any database and assembling unknown proteins. However, owing to the low ion coverage in tandem mass spectra, the order of certain consecutive amino acids cannot be determined if all of their supporting fragment ions are missing, which results in the low precision of de novo sequencing. Results In order to solve this problem, we developed pNovo 3, which used a learning-to-rank framework to distinguish similar peptide candidates for each spectrum. Three metrics for measuring the similarity between each experimental spectrum and its corresponding theoretical spectrum were used as important features, in which the theoretical spectra can be precisely predicted by the pDeep algorithm using deep learning. On seven benchmark datasets from six diverse species, pNovo 3 recalled 29–102% more correct spectra, and the precision was 11–89% higher than three other state-of-the-art de novo sequencing algorithms. Furthermore, compared with the newly developed DeepNovo, which also used the deep learning approach, pNovo 3 still identified 21–50% more spectra on the nine datasets used in the study of DeepNovo. In summary, the deep learning and learning-to-rank techniques implemented in pNovo 3 significantly improve the precision of de novo sequencing, and such machine learning framework is worth extending to other related research fields to distinguish the similar sequences. Availability and implementation pNovo 3 can be freely downloaded from http://pfind.ict.ac.cn/software/pNovo/index.html. Supplementary information Supplementary data are available at Bioinformatics online. [ABSTRACT FROM AUTHOR]

• C. osmophloeum , C. burmannii , and C. insularimontanum resembles each other. • Deep CNN classifiers were developed to identify the Cinnamomum species. • Score fusion was applied to improve the performance of the classifiers. • The fused CNN classifiers reached a testing accuracy of 96.7%. • Comparison indicated that the proposed approach outperformed SVM classifier. Plant adulteration is a rising problem in the forest industry. Cinnamomum osmophloeum Kanehira (Lauraceae) is an evergreen plant that yields cinnamaldehyde. Although two other species— C. burmannii (Nees & T. Nees) Blume and C. insularimontanum Hayata— morphologically resemble C. osmophloeum , they produce a minimum amount of cinnamaldehyde, and thus have lower economical values. The adulteration of C. osmophloeum using C. burmannii has been reported. However, even for experts, differentiating between the three species on the basis of their appearance is challenging due to their high degree of similarity in appearance. This study proposed to identify the three Cinnamomum species using leaf images and deep convolutional neural networks (CNN). In this study, leaf images of the three species were acquired using flatbed scanners. Leaf patches around the blade center of the leaves were extracted. Classifiers based on deep CNN models, VGG16, Inception-V3, and NASNet, were then developed using the leaf images or patches as inputs. Score fusion was then applied to improve the performance of the developed CNN classifiers. The fused CNN classifiers reached a test accuracy as high as 96.7%. A performance comparison indicated that three developed deep CNN classifiers outperformed support vector machine classifiers. [ABSTRACT FROM AUTHOR]