Overall Accuracy Of Classification Research Articles

Performance evaluation of E-VGG19 model: Enhancing real-time skin cancer detection and classification

Skin cancer is a pervasive and potentially life-threatening disease. Early detection plays a crucial role in improving patient outcomes. Machine learning (ML) techniques, particularly when combined with pre-trained deep learning models, have shown promise in enhancing the accuracy of skin cancer detection. In this paper, we enhanced the VGG19 pre-trained model with max pooling and dense layer for the prediction of skin cancer. Moreover, we also explored the pre-trained models such as Visual Geometry Group 19 (VGG19), Residual Network 152 version 2 (ResNet152v2), Inception-Residual Network version 2 (InceptionResNetV2), Dense Convolutional Network 201 (DenseNet201), Residual Network 50 (ResNet50), Inception version 3 (InceptionV3), For training, skin lesions dataset is used with malignant and benign cases. The models extract features and divide skin lesions into two categories: malignant and benign. The features are then fed into machine learning methods, including Linear Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Decision Tree (DT), Logistic Regression (LR) and Support Vector Machine (SVM), our results demonstrate that combining E-VGG19 model with traditional classifiers significantly improves the overall classification accuracy for skin cancer detection and classification. Moreover, we have also compared the performance of baseline classifiers and pre-trained models with metrics (recall, F1 score, precision, sensitivity, and accuracy). The experiment results provide valuable insights into the effectiveness of various models and classifiers for accurate and efficient skin cancer detection. This research contributes to the ongoing efforts to create automated technologies for detecting skin cancer that can help healthcare professionals and individuals identify potential skin cancer cases at an early stage, ultimately leading to more timely and effective treatments.

Mapping Cropland Abandonment in the Cloudy Hilly Regions Surrounding the Southwest Basin of China

Cropland is a vital resource intricately connected to food security. Currently, the issue of cropland abandonment poses a serious threat to food production and supply, presenting a significant challenge to rural economies and the stability of the food supply chain. The hilly and cloudy regions of southwest China are particularly affected by cropland abandonment, presenting significant challenges in accurately mapping the distribution of abandoned cropland due to fragmentation and heavy cloud pollution. Therefore, this study focuses on Mingshan County, located in Ya’an City, Sichuan Province, China, as the study area. Utilizing Google Earth Engine (GEE) and a random forest algorithm, a method integrating multi-source data from Landsat 8, Sentinel-2, and Sentinel-1 is proposed to extract abandoned cropland spanning from 2018 to 2022. This study analyzes spatial and temporal characteristics, employing the Geodetector with optimal parameters to explore the underlying mechanisms. The findings reveal the following: (1) The method achieves an overall accuracy of land use classification surpassing 88.67%, with a Kappa coefficient exceeding 0.87. Specifically, the accuracy for identifying abandoned cropland reaches 87.00%. (2) From 2018 to 2022, the abandonment rate in Mingshan County fluctuated between 4.58% and 5.77%, averaging 5.03%. The lowest abandonment rate occurred in 2019–2020, while the highest was observed in 2020–2021. (3) Cropland abandonment is influenced by both natural and social factors. Elevation and slope are the main driving factors, alongside factors such as distance to road, town, and residential settlement that all significantly contribute to abandonment trends. These five factors exhibit positive correlation with the abandonment rate, with distance to the river showing relatively weaker explanatory power.

A Study of Species and Individual Differentiation of Burnt and Unburnt Bones Using X-Ray Fluorescence

Commingled remains analysis is a fundamental problem in bioarchaeology and forensic anthropology. In cases of commingling, the bones of all individuals represented must be accurately and reliably differentiated. X-ray fluorescence spectrometry (XRF) can identify species and individuals when bone fragments are found in archaeological or forensic contexts. The present study aimed to verify whether portable X-ray fluorescence spectrometry (pXRF) can be used to differentiate bone fragments at the species level (Bos taurus, Sus domesticus, Equus caballus, and Homo sapiens) and at the individual level of human individuals. We also aimed to verify whether species and individuals can be differentiated using pXRF even if the bones were burned. A total of 119 adult human bones from archaeological sites in Central Moravia and Silesia and 17 adult non-human bones from archaeological sites in Central Moravia (Czech Republic) were examined. All bones are dated to various periods (from the Bronze Age to Modern Times). When differentiating the unburnt and burnt bones at the species level, the overall accuracy of classification was 84.6% and 93.9%, respectively. When differentiating unburnt human bones at the individual level, the correct classification ranged from 88.1% to 72.7%. The differentiation of the burnt bones of humans at the individual level achieved an average success rate of more than 60%. The results confirmed that pXRF can be used for species and individual differentiation of unburnt bones and is almost equally applicable to burnt bones.

Classification algorithm for land use in the giant panda habitat of Jiajinshan based on spatial case-based reasoning

Jiajin Mountain, where the giant pandas reside, is an essential nature reserve in China. To comprehend the land use classification of the habitat, this article proposes a remote sensing interpretation algorithm based on spatial case reasoning, known as spatial case-based reasoning (SCBR). The algorithm incorporates specific spatial factors into its framework and does not require an extensive amount of domain knowledge and eliminates the need for a complex model training process, making it capable of completing land use classification in the study area. SCBR comprises a spatial case expression model and a spatial case similarity reasoning model. The paper conducted comparative experiments between the proposed algorithm and support vector machine (SVM), U-Net, vision transformer (ViT), and Trans-Unet, and the results demonstrate that spatial case-based reasoning produces superior classification outcomes. The land use classification experiment based on spatial case-based reasoning at the Jiajinshan giant panda habitat produced satisfactory experimental results. In the comparative experiments, the overall accuracy of SCBR classification reached 95%, and the Kappa coefficient reached 90%. The paper further analyzed the changes in land use classification from 2018 to 2022, and the average accuracy consistently exceeds 80%. We discovered that the ecological environment in the region where the giant pandas reside has experienced significant improvement, particularly in forest protection and restoration. This study provides a theoretical basis for the ecological environment protection of the area.

Using ZY1-02D satellite hyperspectral remote sensing to monitor landscape diversity and its spatial scaling change in the Yellow River Estuary

Monitoring and assessing wetland diversity is crucial for its accurate preservation. Hyperspectral satellites have been proven effective for detailed investigations of plant diversity in many places. However, it's unclear whether spectral diversity invert landscape diversity, and whether the inversion accuracy varies with spatial scale. In this study, the ZY1-02D hyperspectral remote sensing images of the Yellow River Estuary were supervised and classified by the support vector machine. Then, the landscape diversity indices (i.e., community richness, Shannon-Wiener index, Simpson index, and Pielou index) and spectral diversity indices (i.e., coefficient of variation, convex hull volume, and eight vegetation indices) were calculated. A random forest model was used to predict landscape diversity by using spectral diversity. The spatial scale relationship between spectral diversity and landscape diversity were explored lastly. Our results showed that the overall accuracy of plant community classification in the Yellow River Estuary was 91.53 %, with a Kappa coefficient of 0.90. Spectral diversity had the best inversion accuracy on the Shannon-Wiener index (14 ∼ 57 %, average = 38 %), while the intermediate on Pielou index (3 ∼ 56 %, average = 30 %) and community richness (2 ∼ 48 %, average = 30 %), but the lowest on the Simpson index (2 ∼ 43 %, average = 16 %). The inversion accuracy of landscape diversity index increased first and then stabilized with the increase of scales, reaching stability at a sampling size of 2880 m × 2880 m. Our results indicated that ZY1-02D hyperspectral data can be used to monitor spatial changes of landscape diversity in wetland systems. However, its accuracy is affected by diversity index type and spatial scaling effects. Our findings provide a new perspective for the conservation and management of large-scale wetland landscape diversity.

Useful blunders: Can automated speech recognition errors improve downstream dementia classification?

Objectives:We aimed to investigate how errors from automatic speech recognition (ASR) systems affect dementia classification accuracy, specifically in the “Cookie Theft” picture description task. We aimed to assess whether imperfect ASR-generated transcripts could provide valuable information for distinguishing between language samples from cognitively healthy individuals and those with Alzheimer’s disease (AD). Methods:We conducted experiments using various ASR models, refining their transcripts with post-editing techniques. Both these imperfect ASR transcripts and manually transcribed ones were used as inputs for the downstream dementia classification. We conducted comprehensive error analysis to compare model performance and assess ASR-generated transcript effectiveness in dementia classification. Results:Imperfect ASR-generated transcripts surprisingly outperformed manual transcription for distinguishing between individuals with AD and those without in the “Cookie Theft” task. These ASR-based models surpassed the previous state-of-the-art approach, indicating that ASR errors may contain valuable cues related to dementia. The synergy between ASR and classification models improved overall accuracy in dementia classification. Conclusion:Imperfect ASR transcripts effectively capture linguistic anomalies linked to dementia, improving accuracy in classification tasks. This synergy between ASR and classification models underscores ASR’s potential as a valuable tool in assessing cognitive impairment and related clinical applications.

The influence of BRDF effects and representativeness of training data on tree species classification using multi-flightline airborne hyperspectral imagery

Airborne hyperspectral imagery provides detailed spatial and spectral information that can be used for tree species classification, but its application in practical forest inventories over large areas is limited. This study investigates the impact of (i) bidirectional reflectance distribution function (BRDF) effects on multi-flightline airborne imagery and (ii) the selection strategy for training data on tree species classification accuracy over large forested areas. Hyperspectral imagery was acquired using the Chinese Academy of Forestry’s LiDAR, CCD, and Hyperspectral system (CAF-LiCHy) over the Mengjiagang Forest Farm (MFF, ∼155 km2) in northeastern China, which includes common keystone coniferous species in northern China’s temperate-boreal ecozones. A kernel-driven BRDF model based on mNDVI704.50 stratification was applied to airborne hyperspectral data acquired from multi-flightlines. We developed three training data selection strategies, namely #1 sampling over the entire area, #2 localized sampling over the area for each flight date, and #3 gradually decreasing sampling in size. Additionally, two validation assessment strategies were respectively employed: independent data over the entire area and independent data in areas with inconsistent classification labeling between using BRDF corrected and uncorrected hyperspectral imagery. These strategies were implemented to perform a comprehensive cross-comparison of tree species classification accuracy using BRDF corrected and uncorrected airborne hyperspectral imagery. The study found that the representativeness of the training data (in terms of solar-viewing geometry distribution and size) has a stronger impact on the classification accuracy of tree species than the BRDF effects in airborne hyperspectral imagery. By utilizing BRDF corrected hyperspectral imagery with a majority of representative training data, the mean overall accuracy of tree species classification improves by ≤ 2.9 % compared to uncorrected imagery. Notably, BRDF correction has improved classification accuracy by 11.5 % in inconsistently labeled areas, mainly concentrated in the spatial overlap regions of adjacent flightlines. These findings provide valuable insights for improving tree species classification accuracy in large forested areas using multi-flightline airborne hyperspectral imagery.

Wildfire Burnt Area Severity Classification from UAV-Based RGB and Multispectral Imagery

Wildfires present a significant threat to ecosystems and human life, requiring effective prevention and response strategies. Equally important is the study of post-fire damages, specifically burnt areas, which can provide valuable insights. This research focuses on the detection and classification of burnt areas and their severity using RGB and multispectral aerial imagery captured by an unmanned aerial vehicle. Datasets containing features computed from multispectral and/or RGB imagery were generated and used to train and optimize support vector machine (SVM) and random forest (RF) models. Hyperparameter tuning was performed to identify the best parameters for a pixel-based classification. The findings demonstrate the superiority of multispectral data for burnt area and burn severity classification with both RF and SVM models. While the RF model achieved a 95.5% overall accuracy for the burnt area classification using RGB data, the RGB models encountered challenges in distinguishing between mildly and severely burnt classes in the burn severity classification. However, the RF model incorporating mixed data (RGB and multispectral) achieved the highest accuracy of 96.59%. The outcomes of this study contribute to the understanding and practical implementation of machine learning techniques for assessing and managing burnt areas.

Automated in-season rice crop mapping using Sentinel time-series data and Google Earth Engine: A case study in climate-risk prone Bangladesh

High-resolution mapping of rice fields is crucial for understanding and managing rice cultivation in countries like Bangladesh, particularly in the face of climate change. Rice is a vital crop, cultivated in small scale farms that contributes significantly to the economy and food security in Bangladesh. Accurate mapping can facilitate improved rice production, the development of sustainable agricultural management policies, and formulation of strategies for adapting to climatic risks.To address the need for timely and accurate rice mapping, we developed a framework specifically designed for the diverse environmental conditions in Bangladesh. We utilized Sentinel-1 and Sentinel-2 time-series data to identify transplantation and peak seasons and employed the multi-Otsu automatic thresholding approach to map rice during the peak season (April–May). We also compared the performance of a random forest (RF) classifier with the multi-Otsu approach using two different data combinations: D1, which utilizes data from the transplantation and peak seasons (D1 RF) and D2, which utilizes data from the transplantation to the harvest seasons (D2 RF).Our results demonstrated that the multi-Otsu approach achieved an overall classification accuracy (OCA) ranging from 61.18% to 94.43% across all crop zones. The D2 RF showed the highest mean OCA (92.15%) among the fourteen crop zones, followed by D1 RF (89.47%) and multi-Otsu (85.27%). Although the multi-Otsu approach had relatively lower OCA, it proved effective in accurately mapping rice areas prior to harvest, eliminating the need for training samples that can be challenging to obtain during the growing season.In-season rice area maps generated through this framework are crucial for timely decision-making regarding adaptive management in response to climatic stresses and forecasting area-wide productivity. The scalability of our framework across space and time makes it particularly suitable for addressing field data scarcity challenges in countries like Bangladesh and offers the potential for future operationalization.

Random forest and support vector machine classifiers for coastal wetland characterization using the combination of features derived from optical data and synthetic aperture radar dataset

Abstract Mapping mangrove forests is crucial for their conservation, but it is challenging due to their complex characteristics. Many studies have explored machine learning techniques that use Synthetic Aperture Radar (SAR) and optical data to improve wetland classification. This research compares the random forest (RF) and support vector machine (SVM) algorithms, employing Sentinel-1 dual polarimetric C-band data and Sentinel-2 optical data for mapping mangrove forests. The study also incorporates various derived parameters. The Jeffries–Matusita distance and Spearman’s rank correlation are used to evaluate the significance of commonly used spectral indices and SAR parameters in wetland classification. Only significant parameters are retained, reducing data dimensionality from 63 initial features to 23–33 essential features, resulting in an 18% improvement in classification accuracy. The combination of SAR and optical data yields a substantial 33% increase in the overall accuracy for both SVM and RF classification. Consistently, the fusion of SAR and optical data produces higher classification accuracy in both RF and SVM algorithms. This research provides an effective approach for monitoring changes in Pichavaram wetlands and offers a valuable framework for future wetland monitoring, supporting the planning and sustainable management of this critical area.

DEEP LEARNING BASED MULTI TASK ROAD EXTRACTOR MODEL FOR PARCEL EXTRACTION AND CROP CLASSIFICATION USING KNOWLEDGE BASED NDVI TIME SERIES DATA

Abstract. The role of agriculture in food security is critical by the rapid increase of the global population. Food production should be strongly increased to secure and provide life necessities. Remote sensing provides a fruitful tool in agriculture management, and it provides advantages in terms of saving time and effort. This study aims to examine the feasibility of deep learning edge detection algorithm in order to automatically extract the agricultural parcel boundaries from the open-source Google Earth data. The potential of using Normalized Difference Vegetation Index (NDVI) in crop classification crop was also presented after analyzing the pattern of Rabi major crops (Wheat and Sugarcane) in Haridwar district, Uttarakhand, India. The advantage of Google Earth Engine cloud-based platform was exploited in generating NDVI data from Sentinel-2 satellite between October 2022 and February 2023 in order to save time and effort. To check the accuracy of the deep learning model, the value of the Mean Intersection of Union (mIoU) was tested and reached 0.79.To examine the results, ground truth data were collected in the study area using Unmanned Aerial Vehicles. The overall accuracy of the rule set-based classification reached 91.17%, and the kappa coefficient value was 0.82.

Correlation between membrane proteins and sizes of extracellular vesicles and particles: A potential signature for cancer diagnosis.

Extracellular vesicles and particles (EVPs) are recognized as ideal liquid biopsy tools for cancer detection, and membrane proteins are commonly used EVP biomarkers. However, bulk analysis of EVP membrane protein biomarkers typically fails to meet the clinical requirement for diagnostic accuracy. We investigated the correlation between the membrane protein expression level, the binding kinetics to aptamers and the sizes of EVPs with interferometric plasmonic microscopy (iPM), and demonstrated the implementation of the correlative signature to determine cancer types. Using EVPs collected from both cell model and clinical plasma samples with liver, lung, breast, or prostate cancer, we found that the selective set of membrane protein expression levels of five protein markers and their binding kinetics were highly heterogeneous across various sizes of EVPs, resulting in the low overall accuracy(<50%) in cancer classification with bulk analysis of all populations. By grouping the EVPs into three subpopulations according to their sizes, the overall accuracy could be increased to about 70%. We further grouped the EVPs into subpopulations with a 10nm interval in sizes and analysed the correlation between the membrane proteins and sizes with a machine learning algorithm. The results show that the overall accuracy to discriminate cancer types could be improved to 85%. Therefore, this work highlights the significance of size-dependent subtyping of EVPs and suggests that the correlation between the selective set of membrane proteins and sizes of EVP can serve as a signature for clinical cancer diagnosis.

Federated Recognition Architecture Based on Voting and Feedback Mechanisms for Accuracy Object Classification in Distributed Edge Intelligence Environment

Deep learning has recently become a crucial tool to solve many complex problems and has the potential to revolutionize industries. With the widespread adoption of the Internet of Things, there are now many devices with limited computational resources that are capable of running deep learning models. It has opened up new opportunities to implement deep learning in the edge environment so that decisions can be made locally without sending data to a cloud server for processing. However, because of limited resources, the model’s performance and communication overhead are challenges when deploying learning models to edge devices. The focus of this paper is to investigate and propose a federated recognition architecture for object classification in a distributed edge intelligence environment. Specifically, we build an edge server that includes a voting module and a feedback module to improve the overall accuracy of object classification. The voting module aggregates predictions of multiple edge devices whereas the feedback module sends the voting results to edge devices to adjust the local deep learning model. We build edge devices based on the EdgeX platform which makes it easy to manage data and optimize communication overheads. Because the edge server and edge nodes only exchange prediction results, our proposed architecture ensures security with sensitive data as well as deep learning model architecture. By testing on the image dataset, we evaluate the proposed architecture’s performance and show that it outperforms individual local models in terms of accuracy. Furthermore, our experiments demonstrate that with the feedback mechanism, the deep learning model is constantly updated with new data to maintain accuracy and avoid being outdated. Besides, we prove the real-time processing speed by collecting the delay time of the proposed model. The results show that our proposed architecture has the potential to be deployed in practical applications such as smart cities, and surveillance systems.

A New Approach for Feeding Multispectral Imagery into Convolutional Neural Networks Improved Classification of Seedlings

Tree species information is important for forest management, especially in seedling stands. To mitigate the spectral admixture of understory reflectance with small and lesser foliaged seedling canopies, we proposed an image pre-processing step based on the canopy threshold (Cth) applied on drone-based multispectral images prior to feeding classifiers. This study focused on (1) improving the classification of seedlings by applying the introduced technique; (2) comparing the classification accuracies of the convolutional neural network (CNN) and random forest (RF) methods; and (3) improving classification accuracy by fusing vegetation indices to multispectral data. A classification of 5417 field-located seedlings from 75 sample plots showed that applying the Cth technique improved the overall accuracy (OA) of species classification from 75.7% to 78.5% on the Cth-affected subset of the test dataset in CNN method (1). The OA was more accurate in CNN (79.9%) compared to RF (68.3%) (2). Moreover, fusing vegetation indices with multispectral data improved the OA from 75.1% to 79.3% in CNN (3). Further analysis revealed that shorter seedlings and tensors with a higher proportion of Cth-affected pixels have negative impacts on the OA in seedling forests. Based on the obtained results, the proposed method could be used to improve species classification of single-tree detected seedlings in operational forest inventory.

Mangrove forest mapping from object-oriented multi-feature ensemble classification using Sentinel-2 images

Accurate mapping of mangrove forests is crucial for understanding their ecosystem function and developing effective management policies. However, the absence of an operational multi-feature fusion approach and an ensemble classification system restricts the achievement of this goal. This study aims to develop an object-oriented multi-feature ensemble classification scheme (OMEC). First, an enhanced mangrove spectral index (EMSI) is established by analyzing the spectral reflectance differences between mangrove forests and other land cover types. Sentinel-2 images are segmented into objects using the multi-resolution segmentation method. Then, spectral, textural, and geometric features are extracted, and these features (including EMSI) are inputted into the nearest neighbor classifier to implement mangrove classification. The experiment was conducted in three typical mangrove areas in China using Sentinle-2 images. The results demonstrate that EMSI exhibits good spectral separability for mangroves and performs well in the ensemble classification scheme. The overall accuracy of mangrove classification exceeds 90%, with a Kappa coefficient greater than 0.88. The object-oriented multi-feature ensemble classification scheme significantly improves accuracy and exhibits excellent performance. The method enhances the accuracy of mangrove classification, enriches the approach to mangrove remote sensing interpretation, and offers data support and scientific references for the restoration, management, and protection of coastal wetlands.

Denoising and classification of urban ICESat-2 photon data fused with Sentinel-2 spectral images

ABSTRACTThe ICESat-2 (Ice, Cloud, and Land Elevation Satellite-2) can collect earth surface elevation data with high precision on a global scale. However, the collected photon data contains a large amount of background noise due to the influence of sunlight, cloud reflection, and other factors. For photon data of different scenes, how to effectively denoise and achieve accurate classification of photon point clouds is crucial for subsequent applications. This study proposes a random forest based method for denoising and classifying ICESat-2 photon data in urban areas by fusing spectral features from Sentinel-2 images and spatial distribution features from photon data. The experimental results show that the method can effectively identify various types of photons. Compared with the reference data, the overall accuracy of photon denoising and classification is 95.97% on average, and the average kappa coefficient is 94.18%. Further analysis demonstrates that the addition of sentinel-2 spectral information can effectively improve the classification accuracy of photon point clouds in urban areas, and the photon classification method of combining photon lidar data and optical images can be a promising solution to improve classification accuracy.

Remote Sensing Classification of Temperate Grassland in Eurasia Based on Normalized Difference Vegetation Index (NDVI) Time-Series Data

The Eurasian temperate grassland is the largest temperate grassland ecosystem and vegetation transition zone globally. The spatiotemporal distribution and changes of grassland types are vital for grassland monitoring and management. However, there is currently a lack of a unified classification method and standard distribution map of Eurasian temperate grassland types. The Normalized Difference Vegetation Index (NDVI) from remote sensing data is commonly used in grassland monitoring. In this paper, the Accumulated Rate of NDVI Change Index (ARNCI) was proposed to characterize the annual NDVI trend of different temperate grassland types, and four transitional categories were introduced to account for the overlap between them. Based on survey data on the distribution of Eurasian temperate grassland types in the 1980s, the study area was divided into three sub-regions: Northern China, Central Asia, and Mongolia. Regionally, pixel-based ARNCI maps in the 1980s and 1990s were successfully calculated from using NOAA’s AVHRR NDVI time-series products. The ARNCI classification thresholds for different sub-regions were determined, and classification experiments and validation were conducted for each sub-region. The overall accuracies of grasslands types classification for Northern China, Central Asia, and Mongolia in the 1980s were 75.3%, 64.2%, and 84.6%, respectively, which demonstrated that there were variations in classification accuracy in the three sub-regions, and the overall performance was favorable. Finally, distribution maps of Eurasian temperate grassland types in the 1980s and 1990s were obtained, and the spatiotemporal changes of grassland types were analyzed and discussed. The ARNCI method is simple to operate and easy to obtain data, and it can be conveniently used in grassland type classification. The maps firstly address the lack of remote sensing classification maps of Eurasian temperate grassland types, and provide a promising tool for monitoring grassland degradation, management, and utilization.

Vocal cord leukoplakia classification using deep learning models in white light and narrow band imaging endoscopy images.

Accurate vocal cord leukoplakia classification is critical for the individualized treatment and early detection of laryngeal cancer. Numerous deep learning techniques have been proposed, but it is unclear how to select one to apply in the laryngeal tasks. This article introduces and reliably evaluates existing deep learning models for vocal cord leukoplakia classification. We created white light and narrow band imaging (NBI) image datasets of vocal cord leukoplakia which were classified into six classes: normal tissues (NT), inflammatory keratosis (IK), mild dysplasia (MiD), moderate dysplasia (MoD), severe dysplasia (SD), and squamous cell carcinoma (SCC). Vocal cord leukoplakia classification was performed using six classical deep learning models, AlexNet, VGG, Google Inception, ResNet, DenseNet, and Vision Transformer. GoogLeNet (i.e., Google Inception V1), DenseNet-121, and ResNet-152 perform excellent classification. The highest overall accuracy of white light image classification is 0.9583, while the highest overall accuracy of NBI image classification is 0.9478. These three neural networks all provide very high sensitivity, specificity, and precision values. GoogLeNet, ResNet, and DenseNet can provide accurate pathological classification of vocal cord leukoplakia. It facilitates early diagnosis, providing judgment on conservative treatment or surgical treatment of different degrees, and reducing the burden on endoscopists.

Utilizing LANDSAT data and the Maximum Likelihood Classifier for Analysing Land Use Patterns in Shimoga, Karnataka

The loss of natural resources has been linked to rapid and invasive urbanization, which in turn worsens the local environment’s scenery and conditions. Preparation of a land use land cover(LULC) map is one of the methods to observe the changes in the geological structure of the study area. The LULU map gives an idea of changes that are occurring during the specified period which will in turn helps in suggesting the measures to be taken to prevent the chances of natural disasters that might occur because of these changes. This study uses a collection of LANDSAT images to evaluate changes in LULC in the Shimoga district for the years 2010, 2015, and 2020. For the classification and creation of LULC maps for the chosen periods, a supervised technique using a Maximum Likelihood Classifier(MLC) has been used. Waterbodies, urban areas, forest areas, and agricultural land have been recognized as the main classes of LULC. The overall accuracy of these maps has been evaluated while taking into account ground facts from Google Earth Pro. The overall accuracy for classification obtained is 85.03% for 2010, 85.27% for 2015, and 85.61% for 2020. The classifier created using LANDSAT scenes and the MLC approach performs well for the research area, as seen by the Kappa index values of 0.8, 0.8, and 0.81 for the years 2010, 2015, and 2020, respectively. The study’s findings indicate that over ten years, the proportion of built-up areas has expanded from 2.8% to 5.4%. When a 2.49% increase occurs in just 10 years, it is necessary to be concerned given the rise of only 1.6% over the previous 40 years. It can also be observed that the proportion of agricultural land has expanded while the fraction of forests has diminished in the study area. The findings of this study are useful in determining that LULC changes are one of the causes of natural disasters including landslides, floods, and forest fires.

The OCDA-Net: a 3D convolutional neural network-based system for classification and staging of ovarian cancer patients using [18F]FDG PET/CT examinations.

To create the 3D convolutional neural network (CNN)-based system that can use whole-body [18F]FDG PET for recurrence/post-therapy surveillance in ovarian cancer (OC). In this study, 1224 image sets from OC patients who underwent whole-body [18F]FDG PET/CT at Kowsar Hospital between April 2019 and May 2022 were investigated. For recurrence/post-therapy surveillance, diagnostic classification as cancerous, and non-cancerous and staging as stage III, and stage IV were determined by pathological diagnosis and specialists' interpretation. New deep neural network algorithms, the OCDAc-Net, and the OCDAs-Net were developed for diagnostic classification and staging of OC patients using [18F]FDG PET/CT images. Examinations were divided into independent training (75%), validation (10%), and testing (15%) subsets. This study included 37 women (mean age 56.3years; age range 36-83years). Data augmentation techniques were applied to the images in two phases. There were 1224 image sets for diagnostic classification and staging. For the test set, 170 image sets were considered for diagnostic classification and staging. The OCDAc-Net areas under the receiver operating characteristic curve (AUCs) and overall accuracy for diagnostic classification were 0.990 and 0.92, respectively. The OCDAs-Net achieved areas under the receiver operating characteristic curve (AUCs) of 0.995 and overall accuracy of 0.94 for staging. The proposed 3D CNN-based models provide potential tools for recurrence/post-therapy surveillance in OC. The OCDAc-Net and the OCDAs-Net model provide a new prognostic analysis method that can utilize PET images without pathological findings for diagnostic classification and staging.