F-measure Research Articles

Pathology diagnosis of intraoperative frozen thyroid lesions assisted by deep learning.

Thyroid cancer is a common thyroid malignancy. The majority of thyroid lesion needs intraoperative frozen pathology diagnosis, which provides important information for precision operation. As digital whole slide images (WSIs) develop, deep learning methods for histopathological classification of the thyroid gland (paraffin sections) have achieved outstanding results. Our current study is to clarify whether deep learning assists pathology diagnosis for intraoperative frozen thyroid lesions or not. We propose an artificial intelligence-assisted diagnostic system for frozen thyroid lesions that applies prior knowledge in tandem with a dichotomous judgment of whether the lesion is cancerous or not and a quadratic judgment of the type of cancerous lesion to categorize the frozen thyroid lesions into five categories: papillary thyroid carcinoma, medullary thyroid carcinoma, anaplastic thyroid carcinoma, follicular thyroid tumor, and non-cancerous lesion. We obtained 4409 frozen digital pathology sections (WSI) of thyroid from the First Affiliated Hospital of Sun Yat-sen University (SYSUFH) to train and test the model, and the performance was validated by a six-fold cross validation, 101 papillary microcarcinoma sections of thyroid were used to validate the system's sensitivity, and 1388 WSIs of thyroid were used for the evaluation of the external dataset. The deep learning models were compared in terms of several metrics such as accuracy, F1 score, recall, precision and AUC (Area Under Curve). We developed the first deep learning-based frozen thyroid diagnostic classifier for histopathological WSI classification of papillary carcinoma, medullary carcinoma, follicular tumor, anaplastic carcinoma, and non-carcinoma lesion. On test slides, the system had an accuracy of 0.9459, a precision of 0.9475, and an AUC of 0.9955. In the papillary carcinoma test slides, the system was able to accurately predict even lesions as small as 2mm in diameter. Tested with the acceleration component, the cut processing can be performed in 346.12s and the visual inference prediction results can be obtained in 98.61s, thus meeting the time requirements for intraoperative diagnosis. Our study employs a deep learning approach for high-precision classification of intraoperative frozen thyroid lesion distribution in the clinical setting, which has potential clinical implications for assisting pathologists and precision surgery of thyroid lesions.

Hierarchical classification for improving parcel-scale crop mapping using time-series Sentinel-1 data

Parcel-scale crop classification utilizing time-series satellite observations is of significant importance in precision agriculture. The prior knowledge that crop types can be organized in a hierarchical tree structure is beneficial for improving crop classification. Moreover, the crop hierarchy aligns with the coarse-to-fine cognitive process of geographic scenes. Based on the crop hierarchy, this study developed a general hierarchical classification framework for enhancing crop mapping using time-series Sentinel-1 data. Central to this method is a deep-learning-based hierarchical classification model that explores and makes use of crop hierarchical knowledge. First, preprocessed Sentinel-1 data were geometrically overlaid onto farmland parcel maps to derive parcel-scale time-series features. Second, we constructed a hierarchical crop type system for study areas based on the crop phenology of labeled crop-type samples. Third, we developed a deep-learning-based hierarchical classification model to identify crop types for each parcel, to generate final crop-type classification maps. The proposed approach was further discussed and verified through the implementation of parcel-scale time-series crop hierarchical classifications in a study area in France with farmland parcel maps and time-series Sentinel-1 data. The classification results, indicating significant improvements greater than 4.0% in overall accuracy and 5.0% in F1 score over comparative methods, demonstrated the effectiveness of the proposed method in learning multi-scale time-series features for hierarchical crop classification utilizing Sentinel-1 data sequences.

Development of a new indicator for identifying vegetation destruction events using remote sensing data

Frequent climate change and intense anthropogenic activity increase the risk of vegetation destruction. Remote sensing technology, known for its timely observations and wide coverage, is a crucial tool for monitoring vegetation growth. However, accurately detecting vegetation destruction events remains challenging due to their spectral diversity, particularly in complex environments. Existing spectral indices (VIs) have limitations in effectively capturing vegetation dynamics as they are only sensitive to specific physiological parameters of vegetation, such as foliage, canopy, or water content, and are prone to background interference. To address this issue, we proposed the Slope Vegetation Index (SVI) based on Sentinel-2 imagery and PROSAIL model simulation data. Five representative VIs were selected for comprehensive comparison. The results showed that, compared with other VIs, SVI had the highest sensitivity to vegetation physiological parameters, with a correlation coefficient (R2) greater than 0.98. SVI performed best across all vegetation change scenes, with producer accuracy (PA), user accuracy (UA), and F1 score all exceeding 0.90. SVI proved effective in detecting various vegetation destruction events, including logging, insect infestation, landslides, and wildfires. Moreover, SVI was suitable for Landsat-8/9 imagery, achieving an F1 score of over 0.89. Overall, SVI is an effective and robust vegetation monitoring index, offering valuable insights for vegetation resource management and post-disaster ecological restoration.

Two-step deep learning models for detection and identification of the manufacturers and types of dental implants on panoramic radiographs.

The purpose of this study is to develop two-step deep learning models that can automatically detect implant regions on panoramic radiographs and identify several types of implants. A total of 1,574 panoramic radiographs containing 3675 implants were included. The implant manufacturers were Kyocera, Dentsply Sirona, Straumann, and Nobel Biocare. Model A was created to detect oral implants and identify the manufacturers using You Only Look Once (YOLO) v7. After preparing the image patches that cropped the implant regions detected by model A, model B was created to identify the implant types per manufacturer using EfficientNet. Model A achieved very high performance, with recall of 1.000, precision of 0.979, and F1 score of 0.989. It also had accuracy, recall, precision, and F1 score of 0.98 or higher for the classification of the manufacturers. Model B had high classification metrics above 0.92, exception for Nobel's class 2 (Parallel). In this study, two-step deep learning models were built to detect implant regions, identify four manufacturers, and identify implant types per manufacturer.

Euclid preparation. XLIX. Selecting active galactic nuclei using observed colours

The space mission will cover over $14\,000\ $ with two optical and near-infrared spectro-photometric instruments, and is expected to detect around ten million active galactic nuclei (AGN). This unique data set will make a considerable impact on our understanding of galaxy evolution in general, and AGN in particular. For this work we identified the best colour selection criteria for AGN, based only on photometry or including ancillary photometric observations, such as the data that will be available with the Rubin Legacy Survey of Space and Time (LSST) and observations already available from Spitzer /IRAC. The analysis was performed for unobscured AGN, obscured AGN, and composite (AGN and star-forming) objects. We made use of the spectro-photometric realisations of infrared-selected targets at all-$z$ ( ) to create mock catalogues mimicking both the Euclid Wide Survey (EWS) and the Euclid Deep Survey (EDS). Using these mock catalogues, we estimated the best colour selection, maximising the harmonic mean (F1) of: (a) completeness, that is, the fraction of AGN correctly selected with respect to the total AGN sample; and (b) purity, that is, the fraction of AGN inside the selection with respect to the selected sample. The selection of unobscured AGN in both surveys (Wide and Deep) is possible with photometry alone with $ F1=0.22$--0.23 (Wide and Deep), which can increase to $ F1=0.43$--0.38 (Wide and Deep) if we limit out study to objects at $z>0.7$. Such a selection is improved once the Rubin/LSST filters, that is, a combination of the $u$, $g$, $r$, or $z$ filters, are considered, reaching an F1 score of 0.84 and 0.86 for the EDS and EWS, respectively. The combination of a colour with the $ colour, which is possible only in the EDS, results in an F1 score of 0.59, improving the results using only filters, but worse than the selection combining and LSST colours. The selection of composite AGN =0.05$--0.65 at 8--40$\ and obscured AGN is challenging, with $ F1 even when including Rubin/LSST or IRAC filters. This is unsurprising since it is driven by the similarities between the broad-band spectral energy distribution of these AGN and star-forming galaxies in the wavelength range 0.3--5\,micron

Accuracy is not enough: a heterogeneous ensemble model versus FGSM attack

AbstractIn this paper, based on facial landmark approaches, the possible vulnerability of ensemble algorithms to the FGSM attack has been assessed using three commonly used models: convolutional neural network-based antialiasing (A_CNN), Xc_Deep2-based DeepLab v2, and SqueezeNet (Squ_Net)-based Fire modules. Firstly, the three individual deep learning classifier-based Facial Emotion Recognition (FER) classifications have been developed; the predictions from all three classifiers are then merged using majority voting to develop the HEM_Net-based ensemble model. Following that, an in-depth investigation of their performance in the case of attack-free has been carried out in terms of the Jaccard coefficient, accuracy, precision, recall, F1 score, and specificity. When applied to three benchmark datasets, the ensemble-based method (HEM_Net) significantly outperforms in terms of precision and reliability while also decreasing the dimensionality of the input data, with an accuracy of 99.3%, 87%, and 99% for the Extended Cohn-Kanade (CK+), Real-world Affective Face (RafD), and Japanese female facial expressions (Jaffee) data, respectively. Further, a comprehensive analysis of the drop in performance of every model affected by the FGSM attack is carried out over a range of epsilon values (the perturbation parameter). The results from the experiments show that the advised HEM_Net model accuracy declined drastically by 59.72% for CK + data, 42.53% for RafD images, and 48.49% for the Jaffee dataset when the perturbation increased from A to E (attack levels). This demonstrated that a successful Fast Gradient Sign Method (FGSM) can significantly reduce the prediction performance of all individual classifiers with an increase in attack levels. However, due to the majority voting, the proposed HEM_Net model could improve its robustness against FGSM attacks, indicating that the ensemble can lessen deception by FGSM adversarial instances. This generally holds even as the perturbation level of the FGSM attack increases.

A lightweight CNN-Transformer network for pixel-based crop mapping using time-series Sentinel-2 imagery

Deep learning approaches have provided state-of-the-art performance in crop mapping. Recently, several studies have combined the strengths of two dominant deep learning architectures, Convolutional Neural Networks (CNNs) and Transformers, to classify crops using remote sensing images. Despite their success, many of these models utilize patch-based methods that require extensive data labeling, as each sample contains multiple pixels with corresponding labels. This leads to higher costs in data preparation and processing. Moreover, previous methods rarely considered the impact of missing values caused by clouds and no-observations in remote sensing data. Therefore, this study proposes a lightweight multi-stage CNN-Transformer network (MCTNet) for pixel-based crop mapping using time-series Sentinel-2 imagery. MCTNet consists of several successive modules, each containing a CNN sub-module and a Transformer sub-module to extract important features from the images, respectively. An attention-based learnable positional encoding (ALPE) module is designed in the Transformer sub-module to capture the complex temporal relations in the time-series data with different missing rates. Arkansas and California in the U.S. are selected to evaluate the model. Experimental results show that the MCTNet has a lightweight advantage with the fewest parameters and memory usage while achieving the superior performance compared to eight advanced models. Specifically, MCTNet obtained an overall accuracy (OA) of 0.968, a kappa coefficient (Kappa) of 0.951, and a macro-averaged F1 score (F1) of 0.933 in Arkansas, and an OA of 0.852, a Kappa of 0.806, and an F1 score of 0.829 in California. The results highlight the importance of each component of the model, particularly the ALPE module, which enhanced the Kappa of MCTNet by 4.2% in Arkansas and improved the model’s robustness to missing values in remote sensing data. Additionally, visualization results demonstrated that the features extracted from CNN and Transformer sub-modules are complementary, explaining the effectiveness of the MCTNet.

Landslide Recognition Based on Machine Learning Considering Terrain Feature Fusion

Landslides are one of the major disasters that exist worldwide, posing a serious threat to human life and property safety. Rapid and accurate detection and mapping of landslides are crucial for risk assessment and humanitarian assistance in affected areas. To achieve this goal, this study proposes a landslide recognition method based on machine learning (ML) and terrain feature fusion. Taking the Dawan River Basin in Detuo Township and Tianwan Yi Ethnic Township as the research area, firstly, landslide-related data were compiled, including a landslide inventory based on field surveys, satellite images, historical data, high-resolution remote sensing images, and terrain data. Then, different training datasets for landslide recognition are constructed, including full feature datasets that fusion terrain features and remote sensing features and datasets that only contain remote sensing features. At the same time, different ratios of landslide to non-landslide (or positive/negative, P/N) samples are set in the training data. Subsequently, five ML algorithms, including Extreme Gradient Boost (XGBoost), Adaptive Boost (AdaBoost), Light Gradient Boost (LightGBM), Random Forest (RF), and Convolutional Neural Network (CNN), were used to train each training dataset, and landslide recognition was performed on the validation area. Finally, accuracy (A), precision (P), recall (R), F1 score (F1), and intersection over union (IOU) were selected to evaluate the landslide recognition ability of different models. The research results indicate that selecting ML models suitable for the study area and the ratio of the P/N samples can improve the A, R, F1, and IOU of landslide identification results, resulting in more accurate and reasonable landslide identification results; Fusion terrain features can make the model recognize landslides more comprehensively and align better with the actual conditions. The best-performing model in the study is LightGBM. When the input data includes all features and the P/N sample ratio is optimal, the A, P, R, F1, and IOU of landslide recognition results for this model are 97.47%, 85.40%, 76.95%, 80.95%, and 71.28%, respectively. Compared to the landslide recognition results using only remote sensing features, this model shows improvements of 4.51%, 35.66%, 5.41%, 22.27%, and 29.16% in A, P, R, F1, and IOU, respectively. This study serves as a valuable reference for the precise and comprehensive identification of landslide areas.

Optimizing machine learning algorithms for diabetes data: A metaheuristic approach to balancing and tuning classifiers parameters

Diabetes mellitus poses a global health concern, prompting the development of machine learning algorithms designed to construct a model for the accurate classification of patients, enabling precise diagnoses and early-stage treatment. However, the efficacy of classifying diabetes patients through machine learning relies on datasets, often plagued by imbalance, leading to biased classification and inaccurate diagnoses. Previous research attempts, employing techniques like random sampling (under-sampling and oversampling) and the Synthetic Minority Oversampling Technique (SMOTE), have struggled to achieve optimally balanced datasets. Additionally, setting the best parameters for machine learning classifiers remains a challenging task. To address these issues, this research focuses on devising a methodological metaheuristic optimization, a machine learning algorithm tailored for diabetes data balancing, and classifier hyperparameter tuning. Leveraging Particle Swarm Optimization (PSO) algorithm for diabetes data balancing and a genetic algorithm to select the optimal architecture for various machine learning classifiers. The study compares the performance of the proposed metaheuristic data balancer and classifier architecture parameter tuner using classification metrics (F1 score, Average Precision–Recall (APR), AUC, and accuracy). The PSO balanced dataset emerges as the most effective in classifying diabetes, with an Average Percentage Improvement (API) in classification performance metrics: 20.78% accuracy, 16.79% area under the curve for receiver operating characteristics, and a significant 32.78% enhancement in APR. Moreover, the XGBOOST classifier trained with a genetic algorithm demonstrates minimal computational training time for the Centre for Disease Control and Prevention (CDC) diabetes dataset compared to the artificial neural network and random forest classifier. Notably, the imbalanced CDC diabetes dataset exhibits the least APR compared to random under-sampling and the PSO data balancing technique.

Prediction of treatment outcome for branch retinal vein occlusion using convolutional neural network-based retinal fluorescein angiography.

Deep learning techniques were used in ophthalmology to develop artificial intelligence (AI) models for predicting the short-term effectiveness of anti-VEGF therapy in patients with macular edema secondary to branch retinal vein occlusion (BRVO-ME). 180 BRVO-ME patients underwent pre-treatment FFA scans. After 3 months of ranibizumab injections, CMT measurements were taken at baseline and 1-month intervals. Patients were categorized into good and poor prognosis groups based on macular edema at the 4th month follow-up. FFA-Net, a VGG-based classification network, was trained using FFA images from both groups. Class activation heat maps highlighted important locations. Benchmark models (DesNet-201, MobileNet-V3, ResNet-152, MansNet-75) were compared for training results. Performance metrics included accuracy, sensitivity, specificity, F1 score, and ROC curves. FFA-Net predicted BRVO-ME treatment effect with an accuracy of 88.63% and an F1 score of 0.89, with a sensitivity and specificity of 79.40% and 71.34%, respectively.The AUC of the ROC curve for the FFA-Net model was 0.71. The use of FFA based on deep learning technology has feasibility in predicting the treatment effect of BRVO-ME. The FFA-Net model constructed with the VGG model as the main body has good results in predicting the treatment effect of BRVO-ME. The typing of BRVO in FFA may be an important factor affecting the prognosis.

Optimized Prescreen Survey Tool for Predicting Sleep Apnea Based on Deep Neural Network: Pilot Study

Obstructive sleep apnea (OSA) is one of the common sleep disorders related to breathing. It is important to identify an optimal set of questions among the existing questionnaires, using a data-driven approach, that can prescreen OSA with high sensitivity and specificity. The current study proposes reliable models that are based on machine learning techniques to predict the severity of OSA. A total of 66 participants consisted of 45 males and 21 females (average age = 52.4 years old; standard deviation ± 14.6). Participants were asked to fill out the questionnaire items. If the value of the Respiratory Disturbance Index (RDI) was more than 30, the participant was diagnosed with severe OSA. Several different modeling techniques were applied, including deep neural networks with a scaled principal component analysis (DNN-PCA), random forest (RF), Adaptive Boosting Classifier (ABC), Decision Tree Classifier (DTC), K-nearest neighbors classifier (KNC), and support vector machine classifier (SVMC). Among the participants, 27 participants were diagnosed with severe OSA (RDI > 30). The area under the receiver operating characteristic curve (AUROC) was used to evaluate the developed models. As a result, the AUROC values of DNN-PCA, RF, ABC, DTC, KNC, and SVMC models were 0.95, 0.62, 0.53, 0.53, 0.51, and 0.78, respectively. The highest AUROC value was found in the DNN-PCA model with a sensitivity of 0.95, a specificity of 0.75, a positive predictivity of 0.95, an F1 score of 0.95, and an accuracy of 0.95. The DNN-PCA model outperforms the existing screening questionnaires, scores, and other models.

Machine learning based stellar classification with highly sparse photometry data.

Identifying stars belonging to different classes is vital in order to build up statistical samples of different phases and pathways of stellar evolution. In the era of surveys covering billions of stars, an automated method of identifying these classes becomes necessary. Many classes of stars are identified based on their emitted spectra. In this paper, we use a combination of the multi-class multi-label Machine Learning (ML) method XGBoost and the PySSED spectral-energy-distribution fitting algorithm to classify stars into nine different classes, based on their photometric data. The classifier is trained on subsets of the SIMBAD database. Particular challenges are the very high sparsity (large fraction of missing values) of the underlying data as well as the high class imbalance. We discuss the different variables available, such as photometric measurements on the one hand, and indirect predictors such as Galactic position on the other hand. We show the difference in performance when excluding certain variables, and discuss in which contexts which of the variables should be used. Finally, we show that increasing the number of samples of a particular type of star significantly increases the performance of the model for that particular type, while having little to no impact on other types. The accuracy of the main classifier is ∼0.7 with a macro F1 score of 0.61. While the current accuracy of the classifier is not high enough to be reliably used in stellar classification, this work is an initial proof of feasibility for using ML to classify stars based on photometry.

An automated sample generation method by integrating phenology domain optical-SAR features in rice cropping pattern mapping

Accurate spatio-temporal information on rice cropping patterns is vital for predicting grain production, managing water resource and assessing greenhouse gas emissions. However, current automated mapping of rice cropping patterns at regional scale is heavily constrained by insufficient training samples and frequent cloudy weathers in major rice-producing areas. To tackle this challenge, we proposed a Phenology domain Optical-SAR feature inTegration method to Automatically generate single (SC-Rice) and double cropping Rice (DC-Rice) sample (POSTAR) for efficient and refined rice mapping. POSTAR includes three major steps: (1) generating a potential rice map using a phenology- and object-based classification method with optical data (Sentinel-2 MSI) to select candidate rice samples; (2) employing K-means to identify SC- and DC-Rice candidate samples according to unique SAR-based (Sentinel-1 SAR) phenological features; (3) implementing a two-step refinement strategy to filter high-confidence SC- and DC-Rice samples, maintaining a balance between intraclass phenological variance and sample purity. Test areas selected for validation include the Dongting Lake plain and Poyang Lake plain in South China, as well as Fujin county located in the Sanjiang plain of North China. POSTAR proved effective in producing reliable SC- and DC-Rice samples, achieving a high spectral correlation similarity (>0.85) and low dynamic time wrapping distance (<8.5) with field samples. Applying POSTAR-derived samples to random forest classifier yielded an overall accuracy of 89.6%, with F1 score of 0.899 for SC-Rice and 0.938 for DC-Rice in the Dongting Lake plain. Owing to the incorporation of knowledge-based optical and SAR phenological features, POSTAR exhibited strong spatial transferability, achieving an overall accuracy of 96.0% in the Poyang Lake plain and 97.8% in the Fujin county. These results demonstrated the effectiveness of the POSTAR method in accurately mapping rice cropping patterns without extensive field visits, providing valuable insights for crop monitoring in large, diverse, and cloudy regions through the integration of optical and SAR data.

Predicting improvement in biofeedback gait training using short-term spectral features from minimum foot clearance data

Stroke rehabilitation interventions require multiple training sessions and repeated assessments to evaluate the improvements from training. Biofeedback-based treadmill training often involves 10 or more sessions to determine its effectiveness. The training and assessment process incurs time, labor, and cost to determine whether the training produces positive outcomes. Predicting the effectiveness of gait training based on baseline minimum foot clearance (MFC) data would be highly beneficial, potentially saving resources, costs, and patient time. This work proposes novel features using the Short-term Fourier Transform (STFT)-based magnitude spectrum of MFC data to predict the effectiveness of biofeedback training. This approach enables tracking non-stationary dynamics and capturing stride-to-stride MFC value fluctuations, providing a compact representation for efficient processing compared to time-domain analysis alone. The proposed STFT-based features outperform existing wavelet, histogram, and Poincaré-based features with a maximum accuracy of 95%, F1 score of 96%, sensitivity of 93.33% and specificity of 100%. The proposed features are also statistically significant (p&lt;0.001) compared to the descriptive statistical features extracted from the MFC series and the tone and entropy features extracted from the MFC percentage index series. The study found that short-term spectral components and the windowed mean value (DC value) possess predictive capabilities regarding the success of biofeedback training. The higher spectral amplitude and lower variance in the lower frequency zone indicate lower chances of improvement, while the lower spectral amplitude and higher variance indicate higher chances of improvement.

Patent transformation prediction: When a patent can be transformed

Patent transformation is a pivotal pathway for realizing technological advancements, and patent transformation prediction is a potential strategy for improving the patent transformation rate. Existing automated patent transformation prediction models do not predict the transformation time, causing invalid conclusions for these valid patents. In this study, we propose a patent transformation prediction model to predict patent transformation time. (1) To obtain patent features in different time periods, the years elapsed since the patent application are segmented into multiple time slots; (2) For each patent, we extract static features and dynamic features of each time slot after constructing and embedding a dynamic graph of the patent; (3) The features for each time slot are concatenated as the input of the dynamic model which utilizes a neural network to predict the patent transformation of the time slot. We measure the model in diverse domains, each of which includes 10,000 patent transformation data. The experimental results show that precision, recall, and F1 scores are approximately 80% for predicting patent transformation in the next 3 years. Additionally, our study yields some novel findings: (1) later applied patents have a higher transformation speed; (2) over 90% of patent transformations occur within 13 years since the patent application; (3) dynamic features, especially dynamic structured features, have a significantly greater impact on patent transformation prediction compared to static features; (4) our model performs stably on different experiment data.

Dependence of debris flow susceptibility maps on sampling strategy with data-driven grid-based model

Different sampling strategies produce varying sample data, serve as the primary input data and directly affect the accuracy of predictions in data-driven grid-based susceptibility models. This study analyzes the accuracy and variation of debris flow susceptibility maps (DFSMs) generated by various sampling strategies. The study area is the Yingxiu region in China, where six sampling strategies were applied, including three sampling locations (deposition area, runout area, and source area) and two sampling types (centroid and polygon) for the debris flow inventory. The effectiveness of 10 conditioning factors used to build the model was assessed by using Pearson correlation coefficient, variance inflation factor, and information gain ratio (IGR) techniques. We then used Weight of Evidence (WofE), Logistic Regression (LR), and Deep Neural Network (DNN) models to produce DFSMs and quantify their performance using the receiver operating characteristic curve (ROC), Accuracy (ACC), Precision, F1 score, and Recall. The results show that the WofE (AUC: 0.754–0.960), LR (AUC: 0.761–0.965), and DNN (AUC: 0.786–0.976) models all perform well, but the DFSMs and dominant factors depend strongly on sampling strategies, especially on sampling location. If the sample areas are excessively large and span across different factor class labels, or if there is a concentration of either large or small sample areas within a specific region, the results of centroid and polygon sampling strategies may differ or even be contradictory.We recommend: (1) determining sampling locations based on the research objectives to provide more accurate evaluation results; (2) selecting the sampling type by first considering the sample size. If the aforementioned conditions are not present, the quicker and more convenient centroid sampling strategy can be chosen; and (3) determining an appropriate sampling strategy and ensuring the accuracy of initial samples are paramount before producing DFSMs.

YOLOv8n-CSE: A Model for Detecting Litchi in Nighttime Environments

The accurate detection of litchi fruit cluster is the key technology of litchi picking robot. In the natural environment during the day, due to the unstable light intensity, uncertain light angle, background clutter and other factors, the identification and positioning accuracy of litchi fruit cluster is greatly affected. Therefore, we proposed a method to detect litchi fruit cluster in the night environment. The use of artificial light source and fixed angle can effectively improve the identification and positioning accuracy of litchi fruit cluster. In view of the weak light intensity and reduced image features in the nighttime environment, we proposed the YOLOv8n-CSE model. The model improves the recognition of litchi clusters in night environment. Specifically, we use YOLOv8n as the initial model, and introduce the CPA-Enhancer module with chain thinking prompt mechanism in the neck part of the model, so that the network can alleviate problems such as image feature degradation in the night environment. In addition, the VoVGSCSP design pattern in Slimneck was adopted for the neck part, which made the model more lightweight. The multi-scale linear attention mechanism and the EfficientViT module, which can be deeply divided, further improved the detection accuracy and detection rate of YOLOv8n-CSE. The experimental results show that the proposed YOLOv8n-CSE model can not only recognize litchi clusters in the night scene, but also has a significant improvement over previous models. In mAP@0.5 and F1, YOLOv8n-CSE achieved 98.86% and 95.54% respectively. Compared with the original YOLOv8n, RT-DETR-l and YOLOv10n, mAP@0.5 is increased by 4.03%, 3.46% and 3.96%, respectively. When the number of parameters is only 4.93 m, F1 scores are increased by 5.47%, 2.96% and 6.24%, respectively. YOLOv8n-CSE achieves an inference time of 36.5ms for the desired detection results. To sum up, the model can satisfy the criteria of the litchi cluster detection system for extremely accurate nighttime environment identification.

Relation extraction for colorectal cancer via deep learning with entity-aware feature orthogonal decomposition

Relation extraction is significant for text structuring of colorectal cancer (CRC) pathological reports to facilitate doctors’ disease diagnoses. Although many relation extraction methods have been extensively studied for various natural language processing applications, they cannot be well transferred to be applied for CRC pathological reports since CRC pathological reports have some unique characteristics. To this end, a deep learning framework is designed in this paper to extract entity relations in CRC pathological reports, which is based on an encoder–decoder architecture with entity-aware feature orthogonal decomposition. Specifically, to effectively extract semantic features of long and short entities, a two-stream encoder is designed based on an edge-aware convolutional neural network and a dimension-aware dilated convolution residual network. To alleviate the influence of the blending of subject–object features, entity-aware feature orthogonal decomposition is designed to decompose the extracted semantic features into three types, i.e. subject features, object features and subject–object shared features. A stage-wise cross entropy loss is proposed to well train the network. Comparison experiments indicated that our designed network performs well on CRC pathological texts with the performance of 92.3% F1 score, 93.1% Precision, and 91.5% Recall, outperforming the existing relation extraction models.

Deep Learning-Based System Combining Chest X-Ray and Computerized Tomography Images for COVID-19 Diagnosis.

Aims/Background: The coronavirus disease 2019 (COVID-19) pandemic has highlighted the need for accurate and efficient diagnostic methods. This study aims to improve COVID-19 detection by integrating chest X-ray (CXR) and computerized tomography (CT) images using deep learning techniques, further improving diagnostic accuracy by using a combined imaging approach. Methods: The study used two publicly accessible databases, COVID-19 Questionnaires for Understanding the Exposure (COVID-QU-Ex) and Integrated Clinical and Translational Cancer Foundation (iCTCF), containing CXR and CT images, respectively. The proposed system employed convolutional neural networks (CNNs) for classification, specifically EfficientNet and ResNet architectures. The data underwent preprocessing steps, including image resizing, Gaussian noise addition, and data augmentation. The dataset was divided into training, validation, and test sets. Gradient-weighted Class Activation Mapping (Grad-CAM) was used for model interpretability. Results: The EfficientNet-based models outperformed the ResNet-based models across all metrics. The highest accuracy achieved was 99.44% for CXR images and 99.81% for CT images with EfficientNetB5. The models also demonstrated high precision, recall, and F1 scores. For statistical significance, the p-values were less than 0.05, indicating that the results are significant. Conclusion: Integrating CXR and CT images using deep learning significantly improves the accuracy of COVID-19 diagnosis. The EfficientNet-based models, with their superior feature extraction capabilities, show better performance than ResNet models. Grad-CAM Visualizations provide insights into the model's decision-making process, potentially reducing diagnostic errors and accelerating diagnosis processes. This approach can improve patient care and support healthcare systems in managing the pandemic more effectively.

Supervised classification and circuit parameter analysis of electrical bioimpedance spectroscopy data of water stress in tomato plants

Drought poses a significant challenge by inducing water stress in crops, urgently calling for effective monitoring and early intervention. In this work, we conducted a comprehensive 38-day investigation of the bioimpedance of water stressed tomato stems under controlled environmental conditions. A total of 8000 measurements were meticulously collected across a frequency spectrum spanning from 100 Hz to 10 MHz. The collected data was categorized into control, early stress, and late stress groups, corresponding to distinct phases of water stress treatment. To analyze the data, we employed established equivalent circuit models, including Cole, Randles, and double Cole. Subsequently, we evaluated the performance of eight machine learning algorithms widely-used in predicting the water stress stages of the plants. Remarkably, the Cole model, in conjunction with a multi-layer perceptron (MLP) algorithm, demonstrated robust performance, achieving an impressive F1 score of 0.89. In fact, in the control group, 378 out of 388 instances were accurately identified. However, misclassifications were observed in the early and late stress groups, with 31 out of 179 and 14 out of 208 instances mislabeled, respectively. Furthermore, we scrutinized the Cole model’s circuit parameters over time, providing insights aligned with the plant physiological behavior documented in the literature. Our findings suggest that MLP models trained on stem bioimpedance data hold promise as a valuable technique for estimating water stress in tomato plants, offering an essential tool for proactive management by farmers.