Feature Model Research Articles

Preoperative prediction of HER-2 expression status in breast cancer based on MRI radiomics model

Objective: This study aims to explore the predictive value of T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC), and early-delayed phases enhanced magnetic resonance imaging (DCE-MRI) radiomics prediction model in determining human epidermal growth factor receptor 2 status in breast cancer. Methods: A retrospective study was conducted, involving 187 patients with confirmed breast cancer by postsurgical pathology at Zhenjiang First People's Hospital during January 2021 and May 2023. Immunohistochemistry or fluorescence in situ hybridization was used to determine the HER-2 status of these patients, with 48 cases classified as HER-2 positive and 139 cases as HER-2 negative. The training set was used to construct the prediction models and the validation set was used to verify the prediction models. Layers of T2WI, ADC, and early-delayed phase DCE-MRI images were used to delineate the volumeof interest and 960 radiomic features were extracted from each case using Pyradiomic. After screening and dimensionality reduction by intraclass correlation coefficient, Pearson correlation analysis, least absolute shrinkage, and selection operator, the radiomics labels were established. Logistic regression analysis was used to construct the T2WI radiomics model, ADC radiomics model, DCE-2 radiomics model, DCE-6 radiomics model, and the joint sequence radiomics model to predict the HER-2 expression status of breast cancer, respectively. Based on the clinical, pathological, and MRI image characteristics of patients, univariate and multivariate logistic regression analysis wasused to construct a clinicopathological MRI feature model. The radscore of every patient and the clinicopathological MRI features which were statistically significant after screening were used to construct a nomogram model. The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of each model and the decision curve analysis wasused to evaluate the clinical usefulness. Results: The T2WI, ADC, DCE-2, DCE-6, and joint sequence radiomics models, the clinicopathological MRI feature model, and the nomogram model were successfully constructed to predict the expression status of HER-2 in breast cancer. ROC analysis showed that in the training set and validation set, the areas under the curve (AUC) of the T2WI radiomics model were 0.797 and 0.760, of the ADC radiomics model were 0.776 and 0.634, of the DCE-2 radiomics model were 0.804 and 0.759, of the DCE-6 radiomics model were 0.869 and 0.798, of the combined sequence radiomics model were 0.908 and 0.847, of the clinicopathological MRI feature model were 0.703 and 0.693, and of the nomogram model were 0.938 and 0.859, respectively. In the training set, the combined sequence radiomics model outperformed the clinicopathological features model (P＜0.001). In the training and validation sets, the nomogram outperformed the clinicopathological features model (P＜0.05). In addition, the diagnostic performance of the nomogram was better than that of the four single-modality radiomics models in the training cohort (P＜0.05) and was better than that of DCE-2 and ADC models in the validation cohort (P＜0.05). Decision curve analysis indicated that the value of individualized prediction models was higher than clinical and pathological prediction models in clinical practice. The calibration curve showed that the multimodal radiomics model had a high consistency with the actual results in predicting HER-2 expression. Conclusions: T2WI, ADC and early-delayed phase DCE-MRI imaging histology models for HER-2 expression status in breast cancer are expected to provide a non-invasive virtual pathological basis for decision-making on preoperative neoadjuvant regimens in breast cancer.

A combined model based on radiomics features of Sonazoid contrast-enhanced ultrasound in the Kupffer phase for the diagnosis of well-differentiated hepatocellular carcinoma and atypical focal liver lesions: a prospective, multicenter study.

The objective of this study was to develop a combined model based on radiomics features of Sonazoid contrast-enhanced ultrasound (CEUS) during the Kupffer phase and to evaluate its value in differentiating well-differentiated hepatocellular carcinoma (w-HCC) from atypical benign focal liver lesions (FLLs). A total of 116 patients with preoperatively Sonazoid-CEUS confirmed w-HCC or benign FLL were selected from a prospective multiple study on the clinical application of Sonazoid in FLLs conducted from August 2020 to March 2021. According to the randomization principle, the patients were divided into a training cohort and a test cohort in a 7:3 ratio. Seventy-nine patients were used for establishing and training the radiomics model and combined model. In comparison, 37 patients were used for validating and comparing the performance of the models. The diagnostic efficacy of the models for w-HCC and atypical benign FLLs was evaluated using ROCs curves and decision curves. A combined model nomogram was created to assess its value in reducing unnecessary biopsies. Among the patients, there were 55 cases of w-HCC and 61 cases of atypical benign FLLs, including 28 cases of early liver abscess, 16 cases of atypical hepatic hemangioma, 8 cases of hepatocellular dysplastic nodules (DN), and 9 cases of focal nodular hyperplasia (FNH). The radiomics model and combined model we established had AUCs of 0.905 and 0.951, respectively, in the training cohort, and the AUCs of the two models in the test cohort were 0.826 and 0.912, respectively. The combined model outperformed the radiomics feature model significantly. Decision curve analysis demonstrated that the combined model achieved a higher net benefit within a specific threshold probability range (0.25 to 1.00). A nomogram of the combined model was developed. The combined model based on the radiomics features of Sonazoid-CEUS in the Kupffer phase showed satisfactory performance in diagnosing w-HCC and atypical benign FLLs. It can assist clinicians in timely detecting malignant FLLs and reducing unnecessary biopsies for benign diseases.

Experimental Validation of Electrothermal and Aging Parameter Identification for Lithium-Ion Batteries

Modeling and predicting the long-term performance of Li-ion batteries is crucial for the effective design and efficient operation of integrated energy systems. In this paper, we introduce a comprehensive semi-empirical model for Li-ion cells, capturing electrothermal and aging features. This model replicates the evolution of cell voltage, capacity, and internal resistance, in relation to the cell actual operating conditions, and estimates the ongoing degradation in capacity and internal resistance due to the battery use. Thus, the model articulates into two sub-models, an electrothermal one, describing the battery voltage, and an aging one, computing the ongoing degradation. We first propose an approach to identify the parameters of both sub-models. Then, we validate the identification procedure and the accuracy of the electrothermal and aging models through an experimental campaign, also comprising two real cycle load tests at different temperatures, in which real measurements collected from real Li-ion cells are used. The overall model demonstrates good performances in simulating battery characteristics and forecasting degradation. The results show a Mean Absolute Percentage Error (MAPE) lower than 1% for battery voltage and capacity, and a maximum absolute error on internal resistance that is on par with the most up-to-date empirical models. The proposed approach is therefore well-suited for implementation in system modeling, and can be employed as an informative tool for enhancing battery design and operational strategies.

Alzheimer’s Disease Detection via Multiscale Feature Modelling Using Improved Spatial Attention Guided Depth Separable CNN

Early detection of Alzheimer's disease (AD) is critical due to its rising prevalence. AI-aided AD diagnosis has grown for decades. Most of these systems use deep learning using CNN. However, a few concerns must be addressed to identify AD: a. there is a lack of attention paid to spatial features; b. there is a lack of scale-invariant feature modelling; and c. the convolutional spatial attention block (C-SAB) mechanism is available in the literature, but it exploits limited feature sets from its input features to obtain a spatial attention map, which needs to be enhanced. The suggested model addresses these issues in two ways: through a backbone of multilayers of depth-separable CNN. Firstly, we propose an improved spatial convolution attention block (I-SAB) to generate an enhanced spatial attention map for the multilayer features of the backbone. The I-SAB, a modified version of the C-SAB, generates a spatial attention map by combining multiple cues from input feature maps. Such a map is forwarded to a multilayer of depth-separable CNN for further feature extraction and employs a skip connection to produce an enhanced spatial attention map. Second, we combine multilayer spatial attention features to make scale-invariant spatial attention features that can fix scale issues in MRI images. We demonstrate extensive experimentation and ablation studies using two open-source datasets, OASIS and AD-Dataset. The recommended model outperforms existing best practices with 99.75% and 96.20% accuracy on OASIS and AD-Dataset. This paper also performed a domain adaptation test on the OASIS dataset, which obtained 83.25% accuracy.

Predicting nitrous oxide contaminants in Cauvery basin using region-based convolutional neural network

Nitrous oxide (N2O) in riverbeds affects hydrological processes by contributing to the greenhouse effect, indicating poor water quality, disrupting biogeochemical cycling, and linking to eutrophication. Elevated N2O levels signal environmental issues, impacting aquatic life and necessitating precise forecasting for effective environmental management and reduced greenhouse gas emissions. Precisely forecasting nitrous oxide (N2O) emissions from riverbeds is paramount for effective environmental management, given its significant potency as a greenhouse gas. This study focuses on the difficulties related to spatial feature extraction and modeling accuracy in predicting N2O in riverbeds in Tamil Nadu. To address the obstacles, the research suggests utilizing the Deep Learning Based Prediction of Nitrous Oxide Contaminants (DL-PNOC), which studies the N2O contaminants in water using Region-based Convolutional Neural Network (RCNN) for spatial feature extraction, to predict nitrous oxide contaminants. The study is centered on the Cauvery River Basin located in Tamil Nadu, where the emission of N2O is a matter of environment. The outcomes encompass the specialized N2O contaminant model for riverbeds and the implementation of RCNN achieves precise N2O forecasting. The DL-PNOC approach combines a contaminant model with RCNN deep learning techniques to capture spatial characteristics and predict N2O pollutants accurately. Furthermore, using the River Bed Dynamics Simulator reinforces the dependability of the findings. The DL-PNOC approach has exhibited encouraging results, as evidenced by the following metrics: a high IoU of 88.66%, precision of 88.96%, recall of 90.03%, F1 score of 89.22%, and low RMSE and MAE values of 9.14% and 7.59%, respectively. The findings highlight the efficacy of the DL-PNOC approach in precisely forecasting N2O pollutants in river sediments.

Improved random forest classification model combined with C5.0 algorithm for vegetation feature analysis in non-agricultural environments

In response to the challenges posed by the high computational complexity and suboptimal classification performance of traditional random forest algorithms when dealing with high-dimensional and noisy non-agricultural vegetation satellite data, this paper proposes an enhanced random forest algorithm based on the C5.0 algorithm. The paper focuses on the Liaohe Plain, selecting two distinct non-agricultural landscape patterns in Shenbei New District and Changtu County as research objects. High-resolution satellite data from GF-2 serves as the experimental dataset. This paper introduces an ensemble feature method based on the bagging concept to improve the original random forest classification model. This method enhances the likelihood of selecting features beneficial to classifying positive class samples, avoiding excessive removal of useful features from negative samples. This approach ensures feature importance and model diversity. The C5.0 algorithm is then employed for feature selection, and the enhanced vegetation index (EVI) is utilized for vegetation coverage estimation. Results indicate that employing a multi-scale parameter selection tool, combined with limited field-measured data, facilitates the identification and classification of plant species in forest landscapes. The C5.0 algorithm effectively selects classification features, minimizing information redundancy. The established object-oriented random forest classification model achieves an impressive accuracy of 94.02% on the aerial imagery for forest classification dataset, with EVI-based vegetation coverage estimation demonstrating high accuracy. In experiments on the same test set, the proposed algorithm attains an average accuracy of 90.20%, outperforming common model algorithms such as bidirectional encoder representation from transformer, FastText, and convolutional neural network, which achieve average accuracies ranging from 84.41 to 88.33% in identifying non-agricultural artificial habitat vegetation features. The proposed algorithm exhibits a competitive edge compared to other algorithms. These research findings contribute scientific evidence for protecting agricultural ecosystems and restoring agricultural ecosystem biodiversity.

Wind Shear Operation-Based Competency Assessment Model for Civil Aviation Pilots

Undesirable meteorological conditions are typical aviation flight safety threats. Although most meteorological radar and flight augmentation computers have avionic system wind shear alarms, the preferred approach is that pilots avoid unsafe wind shear events. Therefore, effective pilot competency evaluations are needed to assess pilots’ abilities to deal with these events. This study developed a wind shear operation competency model that includes observable behavior indicators, sub-task decomposition, and competency check items. An adapted competency model and a quantitative data-driven competency evaluation criteria optimization method were then developed using three-dimensional competency feature modeling, after which wind shear simulation flight training data were used to verify the effectiveness of the proposed method. The competency assessment grades were significantly correlated with the results from experienced examiners with a 93.33% evaluation accuracy. The proposed quantitative data-driven competency assessment method can provide effective pilot competency assessments for a range of aviation meteorological threats.

Radiomics Analysis of Pericoronary Adipose Tissue From Baseline Coronary Computed Tomography Angiography Enables Prediction of Coronary Plaque Progression.

The relationship between plaque progression and pericoronary adipose tissue (PCAT) radiomics has not been comprehensively evaluated. We aim to predict plaque progression with PCAT radiomics features and evaluate their incremental value over quantitative plaque characteristics. Between January 2009 and December 2020, 500 patients with suspected or known coronary artery disease who underwent serial coronary computed tomography angiography (CCTA) ≥2 years apart were retrospectively analyzed and randomly stratified into a training and testing data set with a ratio of 7:3. Plaque progression was defined with annual change in plaque burden exceeding the median value in the entire cohort. Quantitative plaque characteristics and PCAT radiomics features were extracted from baseline CCTA. Then we built 3 models including quantitative plaque characteristics (model 1), PCAT radiomics features (model 2), and the combined model (model 3) to compare the prediction performance evaluated by area under the curve. The quantitative plaque characteristics of the training set showed the values of noncalcified plaque volume (NCPV), fibrous plaque volume, lesion length, and PCAT attenuation were larger in the plaque progression group than in the nonprogression group ( P < 0.05 for all). In multivariable logistic analysis, NCPV and PCAT attenuation were independent predictors of coronary plaque progression. PCAT radiomics exhibited significantly superior prediction over quantitative plaque characteristics both in the training (area under the curve: 0.814 vs 0.615, P < 0.001) and testing (0.736 vs 0.594, P = 0.007) data sets. NCPV and PCAT attenuation were independent predictors of coronary plaque progression. PCAT radiomics derived from baseline CCTA achieved significantly better prediction than quantitative plaque characteristics.

Performance Evaluation of VGG_19 Model for Brain Tumor Datasets

This paper presents a comprehensive assessment of the VGG19 model for detecting brain tumors using deep learning methods, particularly convolutional neural networks (CNNs). The study evaluates key performance metrics such as accuracy, precision, recall, F1 score, and area under the curve (AUC). The investigation delves into the influence of different activation functions, learning rates, and epochs on the VGG19 model's performance. Activation functions are crucial for feature extraction and model convergence, with a comparison made between ReLU and Leaky ReLU to determine their effectiveness in enhancing brain tumor detection accuracy. Additionally, the study examines the impact of various learning rates (0.01, 0.001, 0.0001) on the model's convergence speed and overall performance. Different numbers of training epochs (15, 25, 50) are also considered to strike a balance between computational efficiency and model effectiveness. Evaluation metrics encompass accuracy, precision, recall, F1 score, and AUC, providing a comprehensive assessment of the model's performance across different configurations. The goal is to identify optimal combinations of activation functions, learning rates, and epochs to maximize the VGG19 model's accuracy in detecting brain tumors. This research contributes to advancing efficient and accurate methodologies for brain tumor detection, facilitating early diagnosis and treatment planning. The insights gained from this study can guide the development of more effective deep learning-based systems for medical image analysis, benefiting both patients and healthcare professionals.

Abstract PO1-07-07: Radiomics features for distinguishing true recurrence versus new primary tumor following breast-conserving treatment

Abstract Background Despite modern surgical and irradiation techniques, ipsilateral breast tumor recurrence (IBTR) accounts for 5-15% of all cancer recurrences in women treated with breast conservative treatment. Historically, the methods to define true recurrence (TR) and new primary tumor (NP) for IBTR mainly rely on clinical and pathological criteria, limiting the accuracy of the discerption and causing misclassification. This study aimed to develop a preoperative, noninvasive model for distinguishing IBTR by integrating clinicopathological review with paired dynamic contrast-enhanced breast magnetic resonance imaging (DCE-MRI) at diagnosis and following in-breast recurrence. MethodsWe retrospectively extracted radiomics features from MRI to develop a radiomic cohort of IBTR (n =46), among which all patients underwent paired preoperative DCE-MRI. This radiomic cohort was divided into a training cohort (n =27) and a validation cohort (n =19) with stratified random sampling. Classification of IBTR as TR or NP on the basis of tumor location, histologic subtype, estrogen receptor, and HER2 status was set as the gold standard. The least absolute shrinkage and selection operator (LASSO) regression and logistic regression were utilized to perform radiomics feature selection and model training, respectively. The clinical utility of the model was determined via decision curve analysis (DCA). Results We selected three radiomics features (first-order feature Kurtosis from the first postcontrast phase on IBTR DCE-MRI, GLCM feature Imc2 from the first postcontrast phase on IBTR DCE-MRI, and the delta feature Correlation from the precontrast phase of the change between IBTR and primary tumor DCE-MRI) to develop an IBTR-classification predicting radiomic model, which performed well in the validation cohort (AUC 0.867, 95% confidence interval (CI) 0.694-1). Further investigation for sensitivity (73.3%) and specificity (100%) verified a favorable concordance between the radiomic classification and the conventional standard, with a diagnostic accuracy of 79%. Conclusions: Our study demonstrated the feasibility of the radiomics model in predicting IBTR classification and provided preoperative information about the nature of “recurrence”. This might have important implications in surgical approaches and multidisciplinary care for IBTR. Further efforts are needed to improve the reproducibility of radiomics features and models in multiple centers. Citation Format: Feilin Qu, Guan-Hua Su, Jin-Hui Li, Chao You, Zhi-Ming Shao. Radiomics features for distinguishing true recurrence versus new primary tumor following breast-conserving treatment [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-07-07.

Developing and validating multi-omics prediction models for late patient-reported dysphagia in head and neck radiotherapy.

To investigate models developed using radiomic and dosiomic (multi-omics) features from planning and treatment imaging for late patient-reported dysphagia in head and neck radiotherapy.&#xD;&#xD;Materials and methods: Training (n=64) and testing (n=23) cohorts of head and neck cancer patients treated with curative intent chemo-radiotherapy with a follow-up time greater than 12 months were retrospectively examined. Patients completed the MD Anderson Dysphagia Inventory and a composite score ≤60 was interpreted as patient-reported dysphagia. A chart review collected baseline dysphagia and clinical factors. Multi-omic features were extracted from planning and last synthetic CT images using the pharyngeal constrictor muscle contours as a region of interest. Late patient-reported dysphagia models were developed using a random forest backbone, with feature selection and up-sampling methods to account for the imbalanced data. Models were developed and validated for multi-omic feature combinations for both timepoints. &#xD;&#xD;Results: A clinical and radiomic feature model developed using the planning CT achieved good performance (validation: sensitivity=80±27% / balanced accuracy=71±23%, testing: sensitivity =80±10% / balanced accuracy=73±11%). The synthetic CT models did not show improvement over the plan CT multi-omics models, with poor reliability of the radiomic features on these images. Dosiomic features extracted from the synthetic CT showed promise in predicting late patient-reported dysphagia.&#xD;&#xD;Conclusion: Multi-omics models can predict late patient-reported dysphagia in head and neck radiotherapy patients. Synthetic CT dosiomic features show promise in developing successful models to account for changes in delivered dose distribution. Multi-center or prospective studies are required prior to clinical implementation of these models.&#xD.

FXAM: A unified and fast interpretable model for predictive analytics

Predictive analytics aims to build machine learning models to predict behavior patterns and use predictions to guide decision-making. Predictive analytics is human involved, thus the machine learning model is preferred to be interpretable. In literature, Generalized Additive Model (GAM) is a standard for interpretability. However, due to the one-to-many and many-to-one phenomena which appear commonly in real-world scenarios, existing GAMs have limitations to serve predictive analytics in terms of both accuracy and training efficiency. In this paper, we propose FXAM (Fast and eXplainable Additive Model), a unified and fast interpretable model for predictive analytics. FXAM extends GAM’s modeling capability with a unified additive model for numerical, categorical, and temporal features. FXAM conducts a novel training procedure called Three-Stage Iteration (TSI). TSI corresponds to learning over numerical, categorical, and temporal features respectively. Each stage learns a local optimum by fixing the parameters of other stages. We design joint learning over categorical features and partial learning over temporal features to achieve high accuracy and training efficiency. We prove that TSI is guaranteed to converge to the global optimum. We further propose a set of optimization techniques to speed up FXAM’s training algorithm to meet the needs of interactive analysis. Thorough evaluations conducted on diverse data sets verify that FXAM significantly outperforms existing GAMs in terms of training speed, and modeling categorical and temporal features. In terms of interpretability, we compare FXAM with the typical post-hoc approach XGBoost+SHAP on two real-world scenarios, which shows the superiority of FXAM’s inherent interpretability for predictive analytics.

Prediction and modeling of tie-dye polyester fabric's texture and color feature in supercritical carbon dioxide

To mitigate the water-intensive challenges of tie-dye clothing production and address the absence of objective evaluation for finished products, we have explored the correlation between tie-dye techniques in supercritical carbon dioxide (SC-CO2) and image patterns. Employing digital image processing, we extracted the average HSV three-component value of the valid tie-dye area, the proportion of incompletely dyed area in the HSV color space, and the initial three texture features of the digital tie-dye image using the Tamura method. We scrutinized five factors in the supercritical CO2 dyeing process: rotational speed, dyeing temperature, dyeing pressure, dye concentration, and dyeing time. The single-factor climbing test revealed that temperature and pressure are the predominant factors influencing tie-dyeing performance. We propose a correlation between the texture parameters of tie-dyeing and the swelling effect of polyester fabric in a supercritical CO2 environment. Subsequently, we conducted two-factor tests on temperature and pressure to investigate their specific effects on the parameters of supercritical polyester tie-dyeing. The results indicated significant changes in regularity, saturation, and the proportion of undyed area with variations in dyeing temperature and pressure. Building upon these findings, we developed an artificial neural network for predicting tie-dye dyeing temperature and pressure in supercritical CO2, utilizing features regularity, saturation, and proportion of undyed area as inputs. In the 105st round, experimental results demonstrated an accuracy of 0.99445, substantiating the efficacy of tie-dye pattern features in supercritical CO2 for technique forecasting.

Addressing subjectivity in paralinguistic data labeling for improved classification performance: A case study with Spanish-speaking Mexican children using data balancing and semi-supervised learning

Paralinguistics is an essential component of verbal communication, comprising elements that provide additional information to the language, such as emotional signals. However, the subjective nature of perceiving affective aspects, such as emotions, poses a significant challenge to the development of quality resources for training recognition models of paralinguistic features. Labelers may have different opinions and perceive different emotions from others, making it difficult to achieve a diverse and sufficient representation of considered categories. In this study, we focused on the automatic classification of paralinguistic aspects in Spanish-speaking Mexican children of elementary school age. However, the dataset presents a strong imbalance in all labeled aspects and a low agreement between the labelers. Furthermore, the audio samples were too short, making it challenging to accurately classify affective speech. To address these challenges, we propose a novel method that combines data balancing algorithms and semisupervised learning to improve the classification performance of the trained models. Our method aims to mitigate the subjectivity involved in labeling paralinguistic data, thus advancing the development of robust and accurate recognition models of affective aspects in speech.

Magnetic resonance imaging radiomic features stability in brain metastases: Impact of image preprocessing, image-, and feature-level harmonization

Background and purposeMagnetic resonance imaging (MRI) scans are highly sensitive to acquisition and reconstruction parameters which affect feature stability and model generalizability in radiomic research. This work aims to investigate the effect of image pre-processing and harmonization methods on the stability of brain MRI radiomic features and the prediction performance of radiomic models in patients with brain metastases (BMs). Materials and methodsTwo T1 contrast enhanced brain MRI data-sets were used in this study. The first contained 25 BMs patients with scans at two different time points and was used for features stability analysis. The effect of gray level discretization (GLD), intensity normalization (Z-score, Nyul, WhiteStripe, and in house-developed method named N-Peaks), and ComBat harmonization on features stability was investigated and features with intraclass correlation coefficient >0.8 were considered as stable. The second data-set containing 64 BMs patients was used for a classification task to investigate the informativeness of stable features and the effects of harmonization methods on radiomic model performance. ResultsApplying fixed bin number (FBN) GLD, resulted in higher number of stable features compare to fixed bin size (FBS) discretization (10 ± 5.5 % higher). Applying harmonization in feature domain was able to improve the stability for non-normalized and normalized images with Z-score and WhiteStripe methods. For the classification task, keeping the stable features resulted in good performance only for normalized images with N-Peaks along with FBS discretization. ConclusionsTo develop a robust MRI based radiomic model we recommend using an intensity normalization method based on a reference tissue (e.g N-Peaks) and then using FBS discretization.

Deep Neural Network for the Prediction of KRAS, NRAS and BRAF Genotypes in left-sided colorectal cancer based on histopathologic images

BackgroundThe KRAS, NRAS, and BRAF genotypes are critical for selecting targeted therapies for patients with metastatic colorectal cancer (mCRC). Here, we aimed to develop a deep learning model that utilizes pathologic whole-slide images (WSIs) to accurately predict the status of KRAS, NRAS, and BRAFV600E. Methods129 patients with left-sided colon cancer and rectal cancer from the Third Affiliated Hospital of Sun Yat-sen University were assigned to the training and testing cohorts. Utilizing three convolutional neural networks (ResNet18, ResNet50, and Inception v3), we extracted 206 pathological features from H&E-stained WSIs, serving as the foundation for constructing specific pathological models. A clinical feature model was then developed, with carcinoembryonic antigen (CEA) identified through comprehensive multiple regression analysis as the key biomarker. Subsequently, these two models were combined to create a clinical-pathological integrated model, resulting in a total of three genetic prediction models. Result103 patients were evaluated in the training cohort (1,782,302 image tiles), while the remaining 26 patients were enrolled in the testing cohort (489,481 image tiles). Compared with the clinical model and the pathology model, the combined model which incorporated CEA levels and pathological signatures, showed increased predictive ability, with an area under the curve (AUC) of 0.96 in the training and an AUC of 0.83 in the testing cohort, accompanied by a high positive predictive value (PPV 0.92). ConclusionThe combined model demonstrated a considerable ability to accurately predict the status of KRAS, NRAS, and BRAFV600E in patients with left-sided colorectal cancer, with potential application to assist doctors in developing targeted treatment strategies for mCRC patients, and effectively identifying mutations and eliminating the need for confirmatory genetic testing.

Saliency Detection Based on Multiple-Level Feature Learning

Finding the most interesting areas of an image is the aim of saliency detection. Conventional methods based on low-level features rely on biological cues like texture and color. These methods, however, have trouble with processing complicated or low-contrast images. In this paper, we introduce a deep neural network-based saliency detection method. First, using semantic segmentation, we construct a pixel-level model that gives each pixel a saliency value depending on its semantic category. Next, we create a region feature model by combining both hand-crafted and deep features, which extracts and fuses the local and global information of each superpixel region. Third, we combine the results from the previous two steps, along with the over-segmented superpixel images and the original images, to construct a multi-level feature model. We feed the model into a deep convolutional network, which generates the final saliency map by learning to integrate the macro and micro information based on the pixels and superpixels. We assess our method on five benchmark datasets and contrast it against 14 state-of-the-art saliency detection algorithms. According to the experimental results, our method performs better than the other methods in terms of F-measure, precision, recall, and runtime. Additionally, we analyze the limitations of our method and propose potential future developments.

Predicting Osteoporosis and Osteopenia by Fusing Deep Transfer Learning Features and Classical Radiomics Features Based on Single-Source Dual-energy CT Imaging

Rationale and ObjectivesTo develop and validate a predictive model for osteoporosis and osteopenia prediction by fusing deep transfer learning (DTL) features and classical radiomics features based on single-source dual-energy computed tomography (CT) virtual monochromatic imaging. MethodsA total of 606 lumbar vertebrae with dual-energy CT imaging and quantitative CT (QCT) evaluation were included in the retrospective study and randomly divided into the training (n=424) and validation (n=182) cohorts. Radiomics features and DTL features were extracted from 70-keV monochromatic CT images, followed by feature selection and model construction, radiomics and DTL features models were established. Then, we integrated the selected two types of features into a features fusion model. We developed a two-level classifier for the hierarchical pairwise classification of each vertebra. All the vertebrae were first classified into osteoporosis and non-osteoporosis groups, then non-osteoporosis group was classified into osteopenia and normal groups. QCT was used as reference. The predictive performance and clinical usefulness of three models were evaluated and compared. ResultsThe area under the curve (AUC) of the features fusion, radiomics and DTL models for the classification between osteoporosis and non-osteoporosis were 0.981, 0.999, 0.997 in the training cohort and 0.979, 0.943, 0.848 in the validation cohort. Furthermore, the AUCs of the abovementioned models for the differentiation between osteopenia and normal were 0.994, 0.971, 0.996 in the training cohort and 0.990, 0.968, 0.908 in the validation cohort. The overall accuracy of the abovementioned models for two-level classifications was 0.979, 0.955, 0.908 in the training cohort and 0.918, 0.885, 0.841 in the validation cohort. Decision curve analysis (DCA) showed that all models had high clinical value. ConclusionThe feature fusion model can be used for osteoporosis and osteopenia prediction with improved predictive ability over a radiomics model or a DTL model alone.

Chaotic characterization and chaotic prediction modelling of HFSWR ionospheric echoes excited by typhoon

ABSTRACT This research explores the intricate-coupling relationship between typhoons and the ionosphere by the high-frequency surface wave radar’s (HFSWR) over-the-horizon capabilities. This letter describes the gravity wave features in travelling ionospheric disturbances (TIDs) and ionospheric drift information. Further discussion is raised on the chaotic properties of gravity wave features using the Lyapunov exponents and chaotic attractors, particularly as they are influenced by typhoon. To predict these gravity wave features more accurately, a new model is developed by synergizing phase space reconstruction theory and statistical learning theory, a chaotic prediction model for gravity wave features constructed on a third-order Volterra filter. The experimental findings confirm that the gravity waves features detected via HFSWR present the hyperchaotic behaviour. More importantly, the innovative gravity wave prediction model based on chaotic attractor can capture the dynamics of the original chaotic system. Notably, the model’s prediction accuracy has been improved by 98.6% after the introduction of phase space reconstruction modelling.

Using Feature Selection Enhancement to Evaluate Attack Detection in the Internet of Things Environment

The rapid evolution of technology has given rise to a connected world where billions of devices interact seamlessly, forming what is known as the Internet of Things (IoT). While the IoT offers incredible convenience and efficiency, it presents a significant challenge to cybersecurity and is characterized by various power, capacity, and computational process limitations. Machine learning techniques, particularly those encompassing supervised classification techniques, offer a systematic approach to training models using labeled datasets. These techniques enable intrusion detection systems (IDSs) to discern patterns indicative of potential attacks amidst the vast amounts of IoT data. Our investigation delves into various aspects of supervised classification, including feature selection, model training, and evaluation methodologies, to comprehensively evaluate their impact on attack detection effectiveness. The key features selected to improve IDS efficiency and reduce dataset size, thereby decreasing the time required for attack detection, are drawn from the extensive network dataset. This paper introduces an enhanced feature selection method designed to reduce the computational overhead on IoT resources while simultaneously strengthening intrusion detection capabilities within the IoT environment. The experimental results based on the InSDN dataset demonstrate that our proposed methodology achieves the highest accuracy with the fewest number of features and has a low computational cost. Specifically, we attain a 99.99% accuracy with 11 features and a computational time of 0.8599 s.