Comparison of Automated Spike Detection Software in Detecting Epileptiform Abnormalities on Scalp-EEG of Genetic Generalized Epilepsy Patients.

PURPOSE: Despite availability of commercial EEG software for automated epileptiform detection, validation on real-world EEG datasets is lacking. Performance evaluation of two software packages on a large EEG dataset of patients with genetic generalized epilepsy was performed. METHODS: Three epileptologists labelled IEDs manually of EEGs from three centres. All Interictal epileptiform discharge (IED) markings predicted by two commercial software (Encevis 1.11 and Persyst 14) were reviewed individually to assess for suspicious missed markings and were integrated into the reference standard if overlooked during manual annotation during a second phase. Sensitivity, precision, specificity, and F1-score were used to assess the performance of the software packages against the adjusted reference standard. RESULTS: One hundred and twenty-five routine scalp EEG recordings from different subjects were included (total recording time, 310.7 hours). The total epileptiform discharge reference count was 5,907 (including spikes and fragments). Encevis demonstrated a mean sensitivity for detection of IEDs of 0.46 (SD 0.32), mean precision of 0.37 (SD 0.31), and mean F1-score of 0.43 (SD 0.23). Using the default medium setting, the sensitivity of Persyst was 0.67 (SD 0.31), with a precision of 0.49 (SD 0.33) and F1-score of 0.51 (SD 0.25). Mean specificity representing non-IED window identification and classification was 0.973 (SD 0.08) for Encevis and 0.968 (SD 0.07) for Persyst. CONCLUSIONS: Automated software shows a high degree of specificity for detection of nonepileptiform background. Sensitivity and precision for IED detection is lower, but may be acceptable for initial screening in the clinical and research setting. Clinical caution and continuous expert human oversight are recommended with all EEG recordings before a diagnostic interpretation is provided based on the output of the software.

[Hyperspectral estimation of leaf water content for winter wheat based on grey relational analysis (GRA)].

The objective of the present study was to compare two methods for the precision of estimating leaf water content (LWC) in winter wheat by combining stepwise regression method and partial least squares (SRM-PLS) or PLS based on the relational degree of grey relational analysis (GRA) between water vegetation indexes (WVIs) and LWC. Firstly, data utilized to analyze the grey relationships between LWC and the selected typical WVIs were used to determine the sensitivity of different WVIs to LWC. Secondly, the two methods of estimating LWC in winter wheat were compared, one was to directly use PLS and the other was to combine SRM and PLS, and then the method with the highest determination coefficient (R2) and lowest root mean square error (RMSE) was selected to estimate LWC in winter wheat. The results showed that the relationships between the first five WVI and LWC were stable by using GRA, and then LWC was estimated by using PLS and SRM-PLS at the whole stages with the R2 and RMSEs being 0.605 and 0.575, 4.75% and 7.35%, respectively. The results indicated that the estimation accuracy of LWC could be improved by using GRA firstly and then by using PLS and SRM-PLS.

[Monitoring the chlorophyll fluorescence parameter Fv/Fm in compact corn based on different hyperspectral vegetation indices].

In order to further assess the feasibility of monitoring the chlorophyll fluorescence parameter Fv/Fm in compact corn by hyperspectral remote sensing data, in the present study, hyperspectral vegetation indices from in-situ remote sensing measurements were utilized to monitor the chlorophyll fluorescence parameter Fv/Fm measured in the compact corn experiment. The relationships were analyzed between hyperspectral vegetation indices and Fv/Fm, and the monitoring models were established for Fv/Fm in the whole growth stages of compact corn. The results indicated that Fv/Fm was significantly correlated to the hyperspectral vegetation indices. Among them, structure-sensitive pigment index (SIPI) was the most sensitive remote sensing variable for monitoring Fv/Fm with correlation coefficient (r) of 0.88. The monitoring model of Fv/Fm was established on the base of SIPI, and the determination coefficients (r2) and the root mean square errors (RMSE) were 0.8126 and 0.082 respectively. The overall results suggest that hyperspectral vegetation indices can be potential indicators to monitor Fv/Fm during growth stages of compact corn.

Moving the field forward: detection of epileptiform abnormalities on scalp electroencephalography using deep learning-clinical application perspectives.

The application of deep learning approaches for the detection of interictal epileptiform discharges is a nascent field, with most studies published in the past 5 years. Although many recent models have been published demonstrating promising results, deficiencies in descriptions of data sets, unstandardized methods, variation in performance evaluation and lack of demonstrable generalizability have made it difficult for these algorithms to be compared and progress to clinical validity. A few recent publications have provided a detailed breakdown of data sets and relevant performance metrics to exemplify the potential of deep learning in epileptiform discharge detection. This review provides an overview of the field and equips computer and data scientists with a synopsis of EEG data sets, background and epileptiform variation, model evaluation parameters and an awareness of the performance metrics of high impact and interest to the trained clinical and neuroscientist EEG end user. The gold standard and inter-rater disagreements in defining epileptiform abnormalities remain a challenge in the field, and a hierarchical proposal for epileptiform discharge labelling options is recommended. Standardized descriptions of data sets and reporting metrics are a priority. Source code-sharing and accessibility to public EEG data sets will increase the rigour, quality and progress in the field and allow validation and real-world clinical translation.

Deep learning for automated epileptiform discharge detection from scalp EEG: A systematic review.

Automated interictal epileptiform discharge (IED) detection has been widely studied, with machine learning methods at the forefront in recent years. As computational resources become more accessible, researchers have applied deep learning (DL) to IED detection with promising results. This systematic review aims to provide an overview of the current DL approaches to automated IED detection from scalp electroencephalography (EEG) and establish recommendations for the clinical research community. We conduct a systematic review according to the PRISMA guidelines. We searched for studies published between 2012 and 2022 implementing DL for automating IED detection from scalp EEG in major medical and engineering databases. We highlight trends and formulate recommendations for the research community by analyzing various aspects: data properties, preprocessing methods, DL architectures, evaluation metrics and results, and reproducibility. The search yielded 66 studies, and 23 met our inclusion criteria. There were two main DL networks, convolutional neural networks in 14 studies and long short-term memory networks in three studies. A hybrid approach combining a hidden Markov model with an autoencoder was employed in one study. Graph convolutional network was seen in one study, which considered a montage as a graph. All DL models involved supervised learning. The median number of layers was 9 (IQR: 5-21). The median number of IEDs was 11 631 (IQR: 2663-16 402). Only six studies acquired data from multiple clinical centers. AUC was the most reported metric (median: 0.94; IQR: 0.94-0.96). The application of DL to IED detection is still limited and lacks standardization in data collection, multi-center testing, and reporting of clinically relevant metrics (i.e. F1, AUCPR, and false-positive/minute). However, the performance is promising, suggesting that DL might be a helpful approach. Further testing on multiple datasets from different clinical centers is required to confirm the generalizability of these methods.

Research on remote sensing classification of fruit trees based on Sentinel-2 multi-temporal imageries.

Accurately obtaining the spatial distribution information of fruit tree planting is of great significance to the development of fruit tree growth monitoring, disease and pest control, and yield estimation. In this study, the Sentenel-2 multispectral remote sensing imageries of different months during the growth period of the fruit trees were used as the data source, and single month vegetation indices, accumulated monthly vegetation indices (∑VIs), and difference vegetation indices between adjacent months (∆VIs) were constructed as input variables. Four conventional vegetation indices of NDVI, PSRI, GNDVI, and RVI and four improved vegetation indices of NDVIre1, NDVIre2, NDVIre3, and NDVIre4 based on the red-edge band were selected to construct a decision tree classification model combined with machine learning technology. Through the analysis of vegetation indices under different treatments and different months, combined with the attribute of Feature_importances_, the vegetation indices of different periods with high contribution were selected as input features, and the Max_depth values of the decision tree model were determined by the hyperparameter learning curve. The results have shown that when the Max_depth value of the decision tree model of the vegetation indices under the three treatments was 6, 8, and 8, the model classification was the best. The accuracy of the three vegetation index processing models on the training set were 0.8936, 0.9153, and 0.8887, and the accuracy on the test set were 0.8355, 0.7611, and 0.7940, respectively. This method could be applied to remote sensing classification of fruit trees in a large area, and could provide effective technical means for monitoring fruit tree planting areas with medium and high resolution remote sensing imageries.

Time series extrinsic regression: Predicting numeric values from time series data.

This paper studies time series extrinsic regression (TSER): a regression task of which the aim is to learn the relationship between a time series and a continuous scalar variable; a task closely related to time series classification (TSC), which aims to learn the relationship between a time series and a categorical class label. This task generalizes time series forecasting, relaxing the requirement that the value predicted be a future value of the input series or primarily depend on more recent values. In this paper, we motivate and study this task, and benchmark existing solutions and adaptations of TSC algorithms on a novel archive of 19 TSER datasets which we have assembled. Our results show that the state-of-the-art TSC algorithm Rocket, when adapted for regression, achieves the highest overall accuracy compared to adaptations of other TSC algorithms and state-of-the-art machine learning (ML) algorithms such as XGBoost, Random Forest and Support Vector Regression. More importantly, we show that much research is needed in this field to improve the accuracy of ML models. We also find evidence that further research has excellent prospects of improving upon these straightforward baselines.

Using HJ-CCD image and PLS algorithm to estimate the yield of field-grown winter wheat.

Remote sensing has been used as an important means of estimating crop production, especially for the estimation of crop yield in the middle and late growth period. In order to further improve the accuracy of estimating winter wheat yield through remote sensing, this study analyzed the quantitative relationship between satellite remote sensing variables obtained from HJ-CCD images and the winter wheat yield, and used the partial least square (PLS) algorithm to construct and validate the multivariate remote sensing models of estimating the yield. The research showed a close relationship between yield and most remote sensing variables. Significant multiple correlations were also recorded between most remote sensing variables. The optimal principal components numbers of PLS models used to estimate yield were 4. Green normalized difference vegetation index (GNDVI), optimized soil-adjusted vegetation index (OSAVI), normalized difference vegetation index (NDVI) and plant senescence reflectance index (PSRI) were sensitive variables for yield remote sensing estimation. Through model development and model validation evaluation, the yield estimation model's coefficients of determination (R2) were 0.81 and 0.74 respectively. The root mean square error (RMSE) were 693.9 kg ha-1 and 786.5 kg ha-1. It showed that the PLS algorithm model estimates the yield better than the linear regression (LR) and principal components analysis (PCA) algorithms. The estimation accuracy was improved by more than 20% than the LR algorithm, and was 13% higher than the PCA algorithm. The results could provide an effective way to improve the estimation accuracy of winter wheat yield by remote sensing, and was conducive to large-area application and promotion.

Quantitative monitoring of leaf area index in wheat of different plant types by integrating NDVI and Beer-Lambert law.

Normalized difference vegetation index (NDVI) is one of the most important vegetation indices in crop remote sensing. It features a simple, fast, and non-destructive method and has been widely used in remote monitoring of crop growing status. Beer-Lambert law is widely used in calculating crop leaf area index (LAI), however, it is time-consuming detection and low in output. Our objective was to improve the accuracy of monitoring LAI through remote sensing by integrating NDVI and Beer-Lambert law. In this study, the Beer-Lambert law was firstly modified to construct a monitoring model with NDVI as the independent variable. Secondly, experimental data of wheat from different years and various plant types (erectophile, planophile and middle types) was used to validate the modified model. The results showed that at 130 DAS (days after sowing), the differences in NDVI, leaf area index (LAI) and extinction coefficient (k) of the three plant types with significantly different leaf orientation values (LOVs) reached the maximum. The NDVI of the planophile-type wheat reached saturation earlier than that of the middle and erectophile types. The undetermined parameters of the model (LAI = -ln (a1 × NDVI + b1)/(a2 × NDVI + b2)) were related to the plant type of wheat. For the erectophile-type cultivars (LOV ≥ 60°), the parameters for the modified model were, a1 = 0.306, a2 = -0.534, b1 = -0.065, and b2 = 0.541. For the middle-type cultivars (30° < LOV < 60°), the parameters were, a1 = 0.392, a2 = -0.881, b1 = 0.028, and b2 = 0.845. And for the planophile-type cultivars (LOV ≤ 30°), those parameters were, a1 = 0.596, a2 = -1.306, b1 = 0.014, and b2 = 1.130. Verification proved that the modified model based on integrating NDVI and Beer-Lambert law was better than Beer-Lambert law model only or NDVI-LAI direct model only. It was feasible to quantitatively monitor the LAI of different plant-type wheat by integrating NDVI and Beer-Lambert law, especially for erectophile-type wheat (R2 = 0.905, RMSE = 0.36, RE = 0.10). The monitoring model proposed in this study can accurately reflect the dynamic changes of plant canopy structure parameters, and provides a novel method for determining plant LAI.

Assessment of Fv/Fm absorbed by wheat canopies employing in-situ hyperspectral vegetation indexes.

Chlorophyll fluorescence parameter of Fv/Fm, as an important index for evaluating crop yields and biomass, is key to guide crop management. However, the shortage of good hyperspectral data can hinder the accurate assessment of wheat Fv/Fm. In this research, the relationships between wheat canopy Fv/Fm and in-situ hyperspectral vegetation indexes were explored to develop a strategy for accurate Fv/Fm assessment. Fv/Fm had the highest coefficients with normalized pigments chlorophyll ratio index (NPCI) and the medium terrestrial chlorophyll index (MTCI). Both NPCI and MTCI were increased with the increase in Fv/Fm. However, NPCI value ceased to increase as Fv/Fm reached 0.61. MTCI had a descending trend when Fv/Fm value was higher than 0.61. A piecewise Fv/Fm assessment model with NPCI and MTCI regression variables was established when Fv/Fm value was ≤0.61 and >0.61, respectively. The model increased the accuracy of assessment by up to 16% as compared with the Fv/Fm assessment model based on a single vegetation index. Our study indicated that it was feasible to apply NPCI and MTCI to assess wheat Fv/Fm and to establish a piecewise Fv/Fm assessment model that can overcome the limitations from vegetation index saturation under high Fv/Fm value.

[Hyperspectral remote sensing diagnosis models of rice plant nitrogen nutritional status].

The correlations of rice plant nitrogen content with raw hyperspectral reflectance, first derivative hyperspectral reflectance, and hyperspectral characteristic parameters were analyzed, and the hyperspectral remote sensing diagnosis models of rice plant nitrogen nutritional status with these remote sensing parameters as independent variables were constructed and validated. The results indicated that the nitrogen content in rice plant organs had a variation trend of stem < sheath < spike < leaf. The spectral reflectance at visible light bands was leaf < spike < sheath < stem, but that at near-infrared bands was in adverse. The linear and exponential models with the raw hyperspectral reflectance at 796.7 nm and the first derivative hyperspectral reflectance at 738.4 nm as independent variables could better diagnose rice plant nitrogen nutritional status, with the decisive coefficients (R2) being 0.7996 and 0.8606, respectively; while the model with vegetation index (SDr - SDb) / (SDr + SDb) as independent variable, i. e., y = 365.871 + 639.323 ((SDr - SDb) / (SDr + SDb)), was most fit rice plant nitrogen content, with R2 = 0.8755, RMSE = 0.2372 and relative error = 11.36%, being able to quantitatively diagnose the nitrogen nutritional status of rice.

[Accuracy of winter wheat identification based on multi-temporal CBERS-02 images].

Chinese-Brazil Earth Resources Satellite No. 2 (CBERS-02) has good spatial resolution and abundant spectral information, and a strong ability in detecting vegetation. Based on five CBERS-02 images in winter wheat growth season, the spectral distance between winter wheat and other ground targets was calculated, and then, winter wheat was classified from each individual image or their combinations by using supervised classification. The train and validation samples were derived from high resolution Aerial Images and SPOT5 images. The accuracies and analyses were evaluated for CBERS-02 images at early growth stages, and the results were compared to those of TM images acquired in the same phenological calendars. The results showed that temporal information was the main factor affecting the classification accuracy in winter wheat, but the characteristics of different sensors could affect the classification accuracy. The multi-temporal images could improve the classification accuracy, compared with the results derived from signal stage, with the producer accuracy of optimum periods combination improved 20.0% and user accuracy improved 7.83%. Compared with TM sensor, the classification accuracy from CBERS-02 was a little lower.