Flood susceptibility mapping using AutoML and a deep learning framework with evolutionary algorithms for hyperparameter optimization

The main goal of this research was to accurately diagnose diseases in pepper plants using convolutional neural network (CNN) models and visual parameters features. Thus, we aimed to build a computing system that can utilise visual cues to identify plant diseases with almost 100% reliability. A total of 2,200 images were trained and tested on a dataset, which was divided into healthy leaves and those with bacterial ailments. The images were preprocessed, resized to 300x300 pixels, and then segmented to extract the relevant traits from them. This CNN model, in a word, is accurate because it detects diseases with 99.65% accuracy. If we want to speak of a high accuracy rate, that's the CNN model. It learns to identify the intricate patterns and features in pepper plants that are associated with diseases. CNN can make revolutionary changes to the identification of plant diseases and help prevent crop damage through timely treatments. Automatic identification of the disease makes life easier for both agriculturists and scientists. The results underscore the importance of image preprocessing, segmentation, and feature extraction in improving the ability of CNN models to locate diseases. Additionally, the performance of the CNN model was compared with other machine learning models such as KNN, Decision Tree, ANN, and Random Forest. It was found that the developed CNN model can predict the response with greater accuracy than the others. Modern high-tech precision farming systems might utilise proposed CNN models for plant disease diagnosis.

Identifying patients with acute aortic dissection using an electrocardiogram with convolutional neural network

COVID-19 disease is known as a new disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) variant. The initial symptoms of the disease commonly include fever (83-98%), fatigue or myalgia, dry cough (76-82%), and shortness of breath (31-55%). Given the prevalence of coughing as a symptom, artificial intelligence has been employed as a means of detecting COVID-19 based on cough sounds. This study aims to compare the performance of six different Convolutional Neural Network (CNN) models (VGG-16, VGG-19, LeNet-5, AlexNet, ResNet-50, and ResNet-152) in detecting COVID-19 using mel-spectrogram images derived from cough sounds. The training and validation of these CNN models were conducted using the Virufy dataset. Audio data was processed to generate mel-spectrogram images, which were subsequently employed as inputs for the CNN models. The AlexNet model, utilizing an input size of 227x227, exhibited the best performance with the highest Area Under the Curve (AUC) value of 0.930303. This study provides compelling evidence of the efficacy of CNN models in detecting COVID-19 based on cough sounds through the utilization of mel-spectrogram images. Furthermore, the study underscores the impact of input size on model performance. The primary contribution of this research lies in identifying the CNN model that demonstrates the best performance in COVID-19 detection based on cough sounds. Additionally, this study establishes the fundamental groundwork for selecting an appropriate CNN methodology for early detection of COVID-19.

Freezing of gait (FOG) is one of the most debilitating motor symptoms experienced by patients with Parkinson's disease (PD). FOG detection is possible using acceleration data from wearable sensors, and a convolutional neural network (CNN) is often used to determine the presence of FOG epochs. We compared the performance of a standard CNN for the detection of FOG with two more complex networks, which are well suited for time series data, the MiniRocket and the InceptionTime. We combined acceleration data of people with PD across four studies. The final data set was split into a training (80%) and hold-out test (20%) set. A fifth study was included as an unseen test set. The data were windowed (2 s) and five-fold cross-validation was applied. The CNN, MiniRocket, and InceptionTime models were evaluated using a receiver operating characteristic (ROC) curve and its area under the curve (AUC). Multiple sensor configurations were evaluated for the best model. The geometric mean was subsequently calculated to select the optimal threshold. The selected model and threshold were evaluated on the hold-out and unseen test set. A total of 70 participants (23.7 h, 9% FOG) were included in this study for training and testing, and in addition, 10 participants provided an unseen test set (2.4 h, 11% FOG). The CNN performed best (AUC = 0.86) in comparison to the InceptionTime (AUC = 0.82) and MiniRocket (AUC = 0.76) models. For the CNN, we found a similar performance for a seven-sensor configuration (lumbar, upper and lower legs and feet; AUC = 0.86), six-sensor configuration (upper and lower legs and feet; AUC = 0.87), and two-sensor configuration (lower legs; AUC = 0.86). The optimal threshold of 0.45 resulted in a sensitivity of 77% and a specificity of 58% for the hold-out set (AUC = 0.72), and a sensitivity of 85% and a specificity of 68% for the unseen test set (AUC = 0.90). We confirmed that deep learning can be used to detect FOG in a large, heterogeneous dataset. The CNN model outperformed more complex networks. This model could be employed in future personalized interventions, with the ultimate goal of using automated FOG detection to trigger real-time cues to alleviate FOG in daily life.

Handwritten digit recognition, a task in computer vision, is critical for applications such as postal automation, banking, and digitization of forms. Traditional approaches have leveraged statistical models, but the rise of deep learning, particularly Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM), has revolutionized the field. However, a comprehensive performance comparison of CNN and LSTM architectures in the context of handwritten digit classification remains underexplored. This study aimed to address this gap by evaluating and comparing CNN and LSTM models for the classification of handwritten digits. Two machine learning models – Convolutional Neural Network (CNN) and Long-Short Term Memory (LSTM) were trained with the preprocessed handwritten digit dataset. The CNN model was designed with multiple convolutional and pooling layers, along with dropout for regularization. The LSTM model was designed with LSTM layers to capture sequential patterns in the data, followed by a dense layer for classification. The models were implemented in python, evaluated and compared based on accuracy, precision, recall and F1-score. The evaluation and comparison results indicate that CNN achieved 99.31% accuracy, 99.0% precision, 99.0% recall, and a 99.0% F1-score, while LSTM achieved 98.90% accuracy, 99.0% precision, 99.0% recall, and a 99.0% F1-score. The results demonstrated that CNN outperformed LSTM in terms of accuracy and misclassification errors, making it the optimal choice for image-based handwritten digit recognition. This finding underscores the efficiency of CNN in addressing challenges related to digit recognition, contributing to the advancement of automated digit classification systems and improving the accuracy of image-based classification tasks.

In the context of surgical interventions for lung adenocarcinoma (LADC), precise determination of the extent of LADC infiltration plays a pivotal role in shaping the surgeon's strategic approach to the procedure. The prevailing diagnostic standard involves the expeditious intraoperative pathological diagnosis of areas infiltrated by LADC. Nevertheless, current methodologies rely on the visual interpretation of tissue images by proficient pathologists, introducing an error margin of up to 15.6%. In this study, we investigated the utilization of Micro-Raman technique on isolated specimens of human LADC with the objective of formulating and validating a workflow for the pathological diagnosis of LADC featuring diverse degrees of infiltration. Our strategy encompasses a thorough pathological characterization of LADC, spanning different tissue types and levels of infiltration. Through the integration of Raman spectroscopy with advanced deep learning models for simultaneous diagnosis, this approach offers a swift, precise, and clinically relevant means of analysis. The diagnostic performance of the convolutional neural network (CNN) model, coupled with the microscopic Raman technique, was found to be exceptional and consistent, surpassing the traditional support vector machine (SVM) model. The CNN model exhibited an area under the curve (AUC) value of 96.1% for effectively distinguishing normal tissue from LADC and an impressive 99.0% for discerning varying degrees of infiltration in LADCs. To comprehensively assess its clinical utility, Raman datasets from patients with intraoperative rapid pathologic diagnostic errors were utilized as test subjects and input into the established CNN model. The results underscored the substantial corrective capacity of the Micro-Raman technique, revealing a misdiagnosis correction rate exceeding 96% in all cases. Ultimately, our discoveries highlight the Micro-Raman technique's potential to augment the intraoperative diagnostic precision of LADC with varying levels of infiltration. And compared to the traditional SVM model, the CNN model has better generalization ability in diagnosing different infiltration levels. This method furnishes surgeons with an objective groundwork for making well-informed decisions concerning subsequent surgical plans.

Automated ischemic stroke detection and classification according to its vascular territory is an essential step in stroke image evaluation, especially at hyperacute stage where mechanical thrombectomy may improve patients’ outcome. This study aimed to evaluate the performance of various convolutional neural network (CNN) models on hyperacute staged diffusion-weighted images (DWI) for detection of ischemic stroke and classification into anterior circulation infarct (ACI), posterior circulation infarct (PCI) and normal image slices. In this retrospective study, 253 cases of hyperacute staged DWI were identified, downloaded and reviewed. After exclusion, DWI from 127 cases were used and we created a dataset containing total of 2119 image slices, and separates it into three groups, namely ACI (618 slices), PCI (149 slices) and normal (1352 slices). Two transfer learning based CNN models, namely Inception-v3, EfficientNet-b0 and one self-derived modified LeNet model were used. The performance of the models was evaluated and activation maps using gradient-weighted class activation mapping (Grad-Cam) technique were made. Inception-v3 had the best overall accuracy (86.3%), weighted F1 score (86.2%) and kappa score (0.715), followed by the modified LeNet (85.2% accuracy, 84.7% weighted F1 score and 0.693 kappa score). The EfficientNet-b0 had the poorest performance of 83.6% accuracy, 83% weighted F1 score and 0.662 kappa score. The activation map showed that one possible explanation for misclassification is due to susceptibility artifact. A sufficiently high performance can be achieved by using CNN model to detect ischemic stroke on hyperacute staged DWI and classify it according to vascular territory.

Understanding the learning mechanism of convolutional neural networks in spectral analysis

For the purpose of applying information technology to the creation of ink style painting, the algorithm of ink painting rendering based on the deep learning framework and convolutional neural network model is designed and improved. Firstly, the ink style rendering program is written in Python. Secondly, VGG under Caffe architecture and Illustration 2Vec models are transplanted to TensorFlow architecture, and the image is rendered in ink style based on deep learning framework and convolutional neural network model. Finally, based on Node.js, the server-side program for image ink style rendering is built. Among them, Express is adopted as the Web-side framework, and the front-end page effect is completed. The results show that the ink rendering logic program is applicable, and the expected purpose is achieved.

This study leveraged Convolutional Neural Network (CNN) models to estimate canopy height in Southeast Australian forests before and after the 2019–2020 bushfire event, using inputs from Sentinel-1, Sentinel-2, spectral indices, and terrain features and GEDI canopy height (rh99) as target variables. Our research consisted of three primary objectives: (1) an evaluation of GEDI rh99 canopy height in relation to an nDSM derived from airborne LiDAR, (2) the development of two CNN models for pre- and post-fire scenarios, and (3) utilizing the trained CNN models to generate pre- and post-fire canopy height maps and canopy height change maps across the continuous landscape. GEDI rh99 accuracy analysis revealed R 2 = 0.85, RMSE = 7.25 m and nRMSE = 0.23 for pre-fire GEDI and R 2 = 0.78, RMSE = 10.1 m and nRMSE = 0.3 for post-fire GEDI. The pre-fire CNN model achieved R 2 = 0.75, RMSE = 7.29 m and nRMSE = 0.22, and post-fire CNN model achieved R 2 = 0.62, RMSE = 11.67 m, and nRMSE = 0.34 when validated against nDSM. Bias analysis was conducted, revealing that GEDI rh99 underestimated nDSM across all canopy height ranges, and slope has no significant impact on GEDI accuracy. Notably, post-fire GEDI rh99 exhibited significant underestimation in high to extreme fire severity. The CNN models’ prediction displayed overestimation in short forests < 20 m and increasing underestimation in taller forests > 20 m and underestimation in high and extreme fire severity. Subsequent investigation indicated that canopy layer thinning in high fire severity conditions resulted in weak GEDI waveform signals and consequent bias. Additionally, the GEDI rh99 bias was propagated to the CNN models, together with insensitivity of spectral band values to canopy height in dense and burnt forests potentially contributed to models’ bias. Finally, we demonstrated CNN models’ ability to monitor forest recovery by generating canopy height maps spanning East Gippsland from 2019 to 2023.

Plants are an integral part of human life, and the ability to identify a fruit plant from its leaf image is both fascinating and challenging. Advances in image processing and pattern recognition have made it feasible to perform plant identification using digital images. Machine learning (ML) and convolutional neural network (CNN) models have demonstrated strong capabilities in handling texture-related features in image processing tasks, including segmentation. In this Paper, we present an approach that utilises ML and CNN models, including AlexNet, Inception, ResNet, LeNet, VGG Net, MobileNet, DenseNet, and GoogLeNet. These models are employed for classifying fruit plants through leaf images, achieving promising performance on leaf image datasets. Among the evaluated CNN models, MobileNet achieved the highest performance with 94.81% training, 99.57% validation, and 99.44% test accuracy, outperforming all others. LeNet, AlexNet, and ResNet also showed strong results above 93%, while DenseNet, GoogLeNet, and VGGNet achieved moderate accuracy. Inception performed the weakest, confirming MobileNet as the most efficient and reliable model for fruit plant leaf classification.

Automatic age and gender prediction from facial images is increasingly crucial for applications in security, marketing, and social media. Existing systems often face challenges related to accuracy, demographic generalization, and bias. This study addresses these issues by developing a deep learning-based system utilizing Convolutional Neural Networks (CNNs) for enhanced classification of age and gender. The key research gaps include limited accuracy, insufficient handling of diverse data, and model bias. The proposed approach encompasses data acquisition, preprocessing, and the design of a CNN architecture within a multi-class classification framework. Various CNN models are evaluated, incorporating transfer learning, hyperparameter optimization, and regularization techniques to improve performance. The system's effectiveness is assessed through metrics such as classification accuracy, precision, recall, and robustness across different demographic groups. Results indicate significant advancements in prediction accuracy and model generalization compared to existing methods. The technology holds practical applications in security, personalized marketing, and social networking. Challenges such as model bias and the need for diverse datasets are addressed, with future research aimed at further refining the model and expanding its applicability. This work highlights the substantial improvements deep learning offers to facial recognition technologies.

Automatic prediction of citrus fruit diseases requires the selection of a suitable classification model among various Convolutional Neural Network (CNN) models. The CNN models are the most effective for image classification. However, more than one CNN model occasionally performs closer to each other. Choosing the best classification model requires considerable effort in this situation. In this study, 17 CNN models were used for citrus fruit disease identification. Citrus fruit diseases include anthracnose, blackspot, canker, scab, and healthy. In this research, the performance of CNN models was analysed, and a two-phase statistical analysis was carried out to determine which CNN model provided the most accurate predictions of citrus crop disease. The 17 CNN models used were AlexNet, Darknet, Densenet, efficientnet, nasnetlarge, googlenet, inceptionresnetv2, inceptionv3, mobilenetv2, nasnetmobile, Xception, squeezenet, resnet18, resnet50, resnet101, vgg16, and vgg19. The two-phase statistical analysis was Duncan's multiple range test, followed by the Wilcoxon signed-rank test is the second one. Again, in order to do a statistical analysis, nine different indicators, including “accuracy, sensitivity, specificity, precision, FPR, F1 Score, MCC, Kappa, and computing time”, were taken into consideration. After the execution of 17 CNN models and twophase statistical analysis, it was revealed that resnet101 is superior among CNN models for citrus fruit disease prediction.

To better understand and improve the prediction of the seasonal surface temperature (TS) across the United States, we employed convolutional neural network (CNN) models trained on the Community Earth System Model Version 2 Large Ensemble (CESM2 LENS). We used lagged sea surface temperatures (SST) over the tropical Pacific region, containing the information of the El Niño Southern Oscillation (ENSO), as input for the CNN models. ENSO is the principal driver of variability in seasonal US surface temperatures (TSUS) and employing CNN models allows for spatiotemporal aspects of ENSO to be analyzed to make seasonal TSUS predictions. For predicting TSUS, the CNN models exhibited significantly improved skill over standard statistical multilinear regression (MLR) models and dynamical forecasts across most regions in the US, for lead times ranging from 1 to 6 months. Furthermore, we employed the CNN models to predict seasonal TSUS during extreme ENSO events. For these events, the CNN models outperformed the MLR models in predicting the effects on seasonal TSUS, suggesting that the CNN models are able to capture the ENSO‐TSUS teleconnection more effectively. Results from a heatmap analysis demonstrate that the CNN models utilize spatial features of ENSO rather than solely the magnitude of the ENSO events, indicating that the improved skill of seasonal TSUS is due to analyzing spatial variation in ENSO events. The proposed CNN model demonstrates a promising improvement in prediction skill compared to existing methods, suggesting a potential path forward for enhancing TSUS forecast skill from subseasonal to seasonal timescales.

Distinguishing pericarpium citri reticulatae of different origins using terahertz time-domain spectroscopy combined with convolutional neural networks