Visual Interpretation Research Articles

Identification of texture MRI brain abnormalities on Fibromyalgia syndrome using interpretable machine learning models

To provide objective diagnostic markers for fibromyalgia symptoms (FMS) diagnosis, we have created interpretable extreme gradient boosting (XGBoost) models using radiomics to aid in the diagnosis of chronic pain (CP) and to develop nomogram models for diagnosing subgroups of FMS. A group of 54 patients with CP and 71 healthy controls was randomly separated into training and validation groups, using a 7:3 ratio. Radiomics features were extracted from grey-matter and white-matter in the filtered mwp0* image. The Mann-Whitney U test, Spearman’s rank correlation test, and least absolute shrinkage and selection operator (LASSO) were utilized to select features. An XGBoost model was created based on these features, and Shapley Additive exPlanations (SHAP) was used for personalization and visual interpretation. A nomogram was developed for the diagnosis of FMS subgroups, utilizing radiomics scores and clinical predictors. The efficacy of the nomogram was evaluated using the area under the receiver operating characteristic curve, while decision curve analysis was employed to evaluate its clinical efficacy. The XGBoost model displays stability in the training validation group, indicating lower overfitting of CP model. The nomogram model combined with the rad-score has a greater ability to distinguish between typical and sub-clinical than the clinical factor model alone. We developed and validated a CP diagnosis model by XGBoost and realized model visualization through SHAP. The rad-score obtained by machine learning was used to build a nomogram model that combines clinical scales to distinguish patients with typical and sub-clinical fibromyalgia.

Study of the patterns of ice lake variation and the factors influencing these changes in the western Nyingchi area

Abstract The current ice lake dataset in the western region of Nyingchi requires further improvement. Due to the intricate distribution of ice lakes and imprecise boundary delineation, research tends to overlook small-scale ice lakes in this area. Moreover, most related studies have focused solely on variations in ice lake areas within key regions, such as the Himalayas, with little attention given to changes occurring in southeastern Tibet. The frequency of ice and snow disasters in the study area has been steadily increasing over the years. Therefore, this study utilizes Landsat satellite images and employs visual interpretation methods to generate more precise and comprehensive maps depicting the distribution of ice lakes in the western region of Nyingchi Province for the years 1994, 2010, 2018, and 2022. Additionally, changes in scale and spatial patterns of different types of ice lakes were investigated. Between 1994 and 2022, the ice lake area in the study area significantly increased by 22.5%, reaching a total of 35.8 ± 3.0 km2. This expansion was primarily driven by glacier-fed lakes, which experienced a remarkable growth rate of 30.8%. In contrast, the non-glacier-fed lakes experienced an increase by only 15.6%. Notably, ice lakes at higher elevations exhibited a peak in expansion, with those above 5143.0 m experiencing the most substantial growth rate of 44.8%. The long-term expansion rate of ice lakes is investigated through the measurement of changes in their boundaries, with the aim to understand the factors contributing to their growth. These findings indicate the rapid expansion of the ice lake near the glacier, with an annual growth rate of 1.3% per annum. Specifically, the glacial-fed section exhibited an expansion rate of 1.1% per annum, while the nonglacial-fed section experienced a growth rate of 0.6% per annum. The seasonal variability in marine glaciers is the primary factor influencing the expansion of ice lakes in this region, with temperature and precipitation serving as the principal driving forces impacting the transformation of these lakes. The data provided by the research results will facilitate a comprehensive understanding of the dynamics and mechanisms governing the ice lake in western Nyingchi, thereby contributing to an enhanced scientific comprehension of potential disaster risks associated with this ice lake.

Ranking attention multiple instance learning for lymph node metastasis prediction on multicenter cervical cancer MRI.

In the current clinical diagnostic process, the gold standard for lymph node metastasis (LNM) diagnosis is histopathological examination following surgical lymphadenectomy. Developing a non-invasive and preoperative method for predicting LNM is necessary and holds significant clinical importance. We develop a ranking attention multiple instance learning (RA-MIL) model that integrates convolutional neural networks (CNNs) and ranking attention pooling to diagnose LNM from T2 MRI. Our RA-MIL model applies the CNNs to derive imaging features from 2D MRI slices and employs ranking attention pooling to create patient-level feature representation for diagnostic classification. Based on the MIL and attention theory, informative regions of top-ranking MRI slices from LNM-positive patients are visualized to enhance the interpretability of automatic LNM prediction. This retrospective study collected 300 female patients with cervical cancer who underwent T2-weighted magnetic resonance imaging (MRI) scanning and histopathological diagnosis from one hospital (289 patients) and one open-source dataset (11 patients). Our RA-MIL model delivers promising LNM prediction performance, achieving the area under the receiver operating characteristic curve (AUC) of 0.809 on the internal test set and 0.833 on the public dataset. Experiments show significant improvements in LNM status prediction using the proposed RA-MIL model compared with other state-of-the-art (SOTA) comparative deep learning models. The developed RA-MIL model has the potential to serve as a non-invasive auxiliary tool for preoperative LNM prediction, offering visual interpretability regarding informative MRI slices and regions in LNM-positive patients.

Incremental learning model for sustainable agricultural land assessment using multimodal satellite data

ABSTRACT The identification of agricultural lands is crucial for sustainable development in rural areas. In this paper, an augmented model that utilizes multimodal satellite-based data samples for identifying agricultural lands via incremental learning. The Whale Optimization Algorithm (WOA) is used for augmenting collected images and data samples to enhance the accuracy of the model under real-time conditions. To identify agricultural lands, Deep Convolutional Neural Networks (CNNs) are trained on the augmented data samples. Additionally, the incorporation of Q-Learning for continuous optimization of the model to enhance its efficiency and effectiveness in identifying agricultural lands. The proposed model offers many edges over existing methods. Firstly, the use of multimodal satellite-based data samples allows for a comprehensive and accurate analysis of agricultural lands. Secondly, the incorporation of the Whale Optimization Algorithm enables the augmentation of collected data samples, leading to improved accuracy and reliability of the model. Thirdly, Deep CNNs allows the extraction of complex features from the data, leading to more accurate identification of agricultural lands. Finally, the use of Q-Learning ensures that the model is continuously optimized to improve its efficiency and effectiveness. The need for this work arises from the limitations of existing methods in accurately identifying agricultural lands. Traditional methods based on manual surveys and visual interpretation are time-consuming, expensive, and prone to errors. Moreover, existing automated methods often lack the ability to analyse multimodal satellite-based data samples and fail to provide accurate results. Based on these observations, the proposed augmented model offers a promising solution for identifying agricultural lands for sustainable development. The use of multimodal satellite-based data samples, WOA, Deep CNNs, and Q-Learning allows for an accurate and efficient analysis of agricultural lands, which can aid in sustainable development planning and decision-making operations.

Reflectance spectrometry and multispectral data for mapping fractures and hydrothermal alterations in the northern edge of the eastern High Atlas, Morocco

The Marginal Folds, in the north-western axis of Moroccan eastern High Atlas, are mainly composed of the Infra-Cenomanian terrains. These latter contain continental detrital deposits (red beds) and are characterized by a network of fractures. These marginal folds host significant concentrations of metallic minerals, such as Cu, Pb, Ag, and Zn, particularly at the Merija (Cu) and Bou Sellam (Pb-Zn) sites.Several techniques were applied to the Landsat-8 Oli images to improve the clarity and visibility of the linear features. Radiometric and atmospheric corrections, color compositions, directional filters, and principal component analysis were adopted. The lineaments resulting from automatic extraction, as well as their statistical analysis were validated by visual interpretation and field investigations. Statistical analysis revealed three dominants fracture systems: N-S, NE-SW, and E-W; the first two systems show a high density in the study area.The mineral alteration of the outcropping rocks is well described by ASTER multispectral imaging techniques. In this work, the effectiveness of this techniques for mineral alteration mapping was demonstrated on the Merija area. Several occurrences of mineral alteration were highlighted, using a combination of false color and band ratios, the principal component analysis (PCA), and the band ratios. This research was completed by reflectance spectrometry, covering the visible, near-infrared (VNIR), and short-wave infrared (SWIR) domains, to identify mineral alteration.Hydrothermal alteration in the Merija region includes clay, phyllite, carbonate, and phyllosilicate alteration, such as kaolinite, illite, montmorillonite, palygorskite, and vermiculite. Regarding to oxidation, as evidenced by the presence of goethite, jarosite, and hematite, it is widespread in most of the analyzed samples, resulting from the alteration of pyrite, followed by jarosite and goethite.

WITHDRAWN: Surging dynamics of unnamed glaciers at the Abi Gamin Peak in the central Himalayas

Surge-type glaciers in the Himalayas have been extensively documented, however, understanding of the mechanisms behind glacier surges remain limited. Herein, based on multi-source remote sensing data, we systematically investigated the characteristics and subglacial processes of the unnamed glacier at Abi Gamin Peak in central Himalayas. Our approach integrated optical stereo photogrammetry, feature tracking, visual interpretation, and glacier velocity decomposition. Our study reveals that the glacier surge was initiated in August 2019 and lasted for less than six months, with rapid acceleration and deceleration phases. During the surge, the glacier transported approximately 0.233 km3 of mass from higher to lower elevations, leading to a thickness increase of >70 m at the glacier terminus. The glacier terminus advanced by >800 m, accompanied by significant expansion of the crevasses. Furthermore, we quantitatively revealed the evolution of basal stresses, strain rates, and sliding velocities during the surge, indirectly characterising the influence of water on it. We contend that the surge in this glacier was primarily driven by subglacial sliding induced by glacier surface meltwater. Significant changes in regional climate serve as external factors that perturb the glacial dynamic equilibrium. The observed characteristics of the unnamed glacier at Abi Gamin Peak indicate that its surge was controlled by a hydrological switch. Notably, our work contributes to an enhanced understanding of glacier surge mechanisms in High Mountain Asia.

An Explainable Deep Learning Approach for Classifying Monkeypox Disease by Leveraging Skin Lesion Image Data

According to the World Health Organization's (WHO) external situation report on the multi-country outbreak of Monkeypox in 2023, from 11 countries in Southeast Asia Regions, Thailand recorded the highest reported cases, totaling 461. The ongoing Monkeypox outbreak has raised significant public health concerns due to its rapid spread across several nations. Early detection and diagnosis are imperative for effectively treating and controlling Monkeypox. Given this context, this study aimed to determine the most efficient model for detecting Monkeypox by employing interpretable deep learning techniques. This study utilizes deep learning techniques to diagnose Monkeypox based on images of skin lesions. We evaluate based on four models—convolutional neural network (CNN), gated recurrent unit (GRU), long short-term memory (LSTM), and bidirectional long short term memory (BiLSTM)—using a publicly available dataset. Additionally, we incorporate Local Interpretable Model-Agnostic Explanations (LIME) and techniques for explainable AI, facilitating visual interpretation of model predictions for healthcare practitioners. The CNN model's performance and LSTM model's performance have an accuracy of 100%, while the GRU model's performance and BiLSTM model's performance have an accuracy of 99.88% and 99.45%. Our findings demonstrate the effectiveness of deep learning models, including the suggested CNN model leveraging the pre-trained MobileNetV2 and LSTM. These models can play a pivotal role in combating the Monkeypox virus. Doi: 10.28991/ESJ-2024-08-05-013 Full Text: PDF

Machine Learning Model Based on Prognostic Nutritional Index for Predicting Long-Term Outcomes in Patients With HCC Undergoing Ablation.

To develop multiple machine learning (ML) models based on the prognostic nutritional index (PNI) and determine the optimal model for predicting long-term survival outcomes in hepatocellular carcinoma (HCC) patients after local ablation. From January 2009 to December 2019, we analyzed data from 848 primary HCC patients who underwent local ablation. ML models were constructed and evaluated using the concordance index (C-index), concordance-discordance area under curve (C/D AUC), and Brier scores. The optimal ML model was interpreted using the partial dependence plot (PDP) and SHapley Additive exPlanations (SHAP) framework. Additionally, the prognostic performance of our model was compared with other models. Alkaline phosphatase, preoperation alpha-fetoprotein level, PNI, tumor number, and tumor size were identified as independent prognostic factors for ML model construction. Among the 19 ML algorithms tested, the Aorsf model showed superior performance in both the training cohort (C/D AUC: 0.733; C-index: 0.736; Brier score: 0.133) and validation cohort (C/D AUC: 0.713; C-index: 0.793; Brier score: 0.117). The time-dependent AUC of the Aorsf model for predicting overall survival was as follows: 1-, 3-, 5-, 7-, and 9-year were 0.828, 0.765, 0.781, 0.817, and 0.812 in the training cohort, 0.846, 0.859, 0.824, 0.845, and 0.874 in the validation cohort, respectively. The PDP and SHAP algorithms were employed for visual interpretation. Furthermore, time-AUC and decision curve analysis demonstrated that the Aorsf model provided superior clinical benefits compared to other models. The PNI-based Aorsf model effectively predicts long-term survival outcomes after ablation therapy, making a significant contribution to HCC research by improving surveillance, prevention, and treatment strategies.

Computer Vision Promising Innovations

Computer vision, an interdisciplinary field bridging artificial intelligence and image processing, seeks to bestow machines with the capability to interpret and make decisions based on visual data. As the digital age propels forward, the ubiquity of visual content underscores the importance of efficient and effective automated interpretation. This paper delves deeply into the modern advancements and methodologies of computer vision, emphasizing its transformative role in various applications ranging from medical imaging to autonomous driving. With the increasing complexity of visual data, challenges arise pertaining to real-time processing, scalability, and the ethical implications of automated decision-making. Through an exhaustive literature review and novel experimentation, this research demystifies the multifaceted domain of computer vision, elucidating its potential and constraints. The study culminates in a visionary outlook, highlighting future avenues for research, including the fusion of augmented reality with computer vision, novel deep learning architectures, and ensuring ethical AI practices in visual interpretation.

Visual interpretation of deep learning model in ECG classification: A comprehensive evaluation of feature attribution methods

Feature attribution methods can visually highlight specific input regions containing influential aspects affecting a deep learning model's prediction. Recently, the use of feature attribution methods in electrocardiogram (ECG) classification has been sharply increasing, as they assist clinicians in understanding the model's decision-making process and assessing the model's reliability. However, a careful study to identify suitable methods for ECG datasets has been lacking, leading researchers to select methods without a thorough understanding of their appropriateness. In this work, we conduct a large-scale assessment by considering eleven popular feature attribution methods across five large ECG datasets using a model based on the ResNet-18 architecture. Our experiments include both automatic evaluations and human evaluations. Annotated datasets were utilized for automatic evaluations and three cardiac experts were involved for human evaluations. We found that Guided Grad-CAM, particularly when its absolute values are utilized, achieves the best performance. When Guided Grad-CAM was utilized as the feature attribution method, cardiac experts confirmed that it can identify diagnostically relevant electrophysiological characteristics, although its effectiveness varied across the 17 different diagnoses that we have investigated.

Map of anthropogenic soil disturbance in Russia

Soils play an important role in maintaining the sustainability of the biosphere, as well as providing food, clothing and the basis for human life on Earth. In the process of irrational land use, soils are often degraded and sometimes completely destroyed. But, inventory of destroyed soils on a systematic basis is still not conducted in any country of the world. In Russia, traditional soil maps do not reflect the destroyed soils either. We have made an attempt to create the first map of the country, which shows soils destroyed as a result of directed anthropogenic impact. The map shows areas where soil cover was destroyed as a result of construction of buildings, structures, roads and railroads, quarries and embankments during mining. The crowdsourcing database OpenStreetMap was used as the main source of information, as well as the results of visual interpretation of disturbed soils using GoogleEarthTM satellite data. The data on disturbed soils were aggregated to soil-geographical sections of the Unified State Register of Soil Resources of Russia (scale 1 : 2500000) (UGRSR). The map is presented in GIS format (shapefile). The map contains information on the area and share of disturbed soils, as well as the type of impact that resulted in disturbance.

Frost Characterization for the Altiplano, Valleys and Chaco Regions of Bolivia

The analysis focused on frost events in the Altiplano, Valles, and Chaco regions of Bolivia, with an emphasis on sub-zero temperatures and related atmospheric patterns to identify the predominant type of frost. A retrospective, observational, and longitudinal design was employed to analyze minimum temperatures from nine meteorological stations between 2005 and 2019. Radiative frosts in the Altiplano showed stability in minimum temperatures from April to October, with low variability and the absence of outliers, while variability was higher in the Valles and Chaco regions. Frost events were more frequent and severe from May to August, with the Altiplano being the most affected region. Satellite imagery and atmospheric reanalysis data from 2010 to 2019 were used for the visual interpretation of atmospheric patterns. A mixed-method approach—quantitative for the statistical analysis of minimum temperatures and qualitative for image interpretation—identified 24 advection frost events between 2010 and 2019, primarily in July and August, associated with high-pressure systems over the Atlantic and Pacific. The Chaco region was the most affected by advection frosts, followed by the Altiplano and the Valles. These findings underscore the importance of spatial and temporal variability in frost occurrence and the need for regional strategies to manage climate risk. It is concluded that altitude, along with factors such as solar radiation, thermal inversion, and wind, is key in frost formation, with radiative frosts being the most common, although advection frosts are also significant, especially under cold front conditions.

Establishing a Landslide Traces Inventory for the Baota District, Yan’an City, China, Using High-Resolution Satellite Images

The Baota District of Yan’an City, located in the Loess Plateau, is an important patriotic education base in China. The region’s fragile geological environment and frequent geological disasters pose significant threats to the production and livelihood of residents. Establishing a landslide traces inventory can provide crucial assistance for studying regional land disaster distribution patterns and implementing disaster prevention and mitigation measures. However, the Baota District has not yet established a comprehensive and detailed landslide traces inventory, resulting in a lack of clear understanding and comprehensive knowledge regarding the threats and impacts of landslide disasters in the area. Therefore, this study employed high-resolution satellite images, applying a human–computer interactive visual interpretation method in conjunction with field survey verifications, to develop the most detailed and comprehensive landslide traces inventory for the Baota District to date. The results indicate that within the 3556 km2 area of the Baota District, there are 73,324 landslide traces, with an average landslide density of 20.62 km-2 and a total landslide area of 769.12 km2, accounting for 21.63% of the total land area. These landslides are relatively evenly distributed throughout the district, with a higher concentration in the east compared to the west. Most of the landslides are small in size. This study can support disaster prevention and mitigation efforts in the Baota District and serve as a reference for establishing landslide inventories in other regions of the Loess Plateau.

Enhancing the effectiveness of teaching neuroanatomy: A comparative study using stained and unstained brain sections to interpret cross sectional neuroanatomy

BackgroundNeuroanatomy is considered one of the most challenging parts of anatomy curriculum and the complexity in understanding neuroanatomy generates from the lack of a 3-dimensional view of the structures. Since everything appears grey and white and is difficult to differentiate the various structures in a cross section of brain, staining aids in the visual interpretation and retention. There are various staining methods employed such as Mulligan's method, Alston's method, Prussian blue reaction method, however the results obtained by Alston's method was the best as gathered from previous literature. Material and methodsThe study was conducted in the department of Anatomy, formalin fixed brain slices measuring 10 mm thickness were stained using Alston's method. 250 medical students of the first professional year were demonstrated features such as the internal capsule, caudate nucleus, lentiform and thalamus in the stained and unstained transverse sections of cerebrum during their neuroanatomy practical sessions. Following which a feedback was collected regarding both the specimens using Microsoft Forms. ResultsThere was a positive feedback from the students with 83 % of them preferring the stained over the unstained section. Excerpts from the open ended question where all in favour of the stained sections -“much easier to see and identify the parts in the stained section because of the clear distinction between white and grey matter”. ConclusionsMacroscopic staining of sections of the brain using Alston's method could be implemented as a valuable method for effective teaching of neuroanatomy since students appreciated the structures better in the stained sections which enhanced their retention of neuroanatomy. Since Alston stain produces the least shrinkage, these stained sections could be utilized for research studies and such specimens could also be plastinated.

Transition Radiation on a Conducting Target Shaped as a Right Dihedral Angle

The transition radiation of a charged particle in the simplest case of its incidence onto an infinite, perfectly conducting plane can be described using the method of images known from electrostatics. The same method makes it possible to find the field distribution in more complex cases, such as the field created by a point particle in the presence of two intercepting conducting planes, the angle between which divides exactly the angle of 180°. Based on the method of images, a description is given of the transition radiation that occurs when a fast-charged particle falls on a target in the form of two conducting half-planes intersecting at the right angle (on the inside of the dihedral angle). The features of the radiation emitted by fast and slow particles are qualitatively considered, and their visual interpretation is given. The possibility of using interference effects arising from radiation to monitor beams of charged particles is discussed.

Identification and Characterization of Reclaimed and Underclaimed Mine Features Using Lidar and Temporal Remote Sensing Methods within the Coastal Plain Uranium Mining Region of Texas

We developed a spatiotemporal mapping approach utilizing multiple techniques for distinguishing and mapping known reclaimed mine sites from “unreclaimed” mine sites in a historic uranium mining district along the South Texas Coastal Plains. Lidar laser scanning penetrates the vegetation canopy to expose anthropogenic modifications to the landscape. The Lidar analysis (bare earth elevation surface, slope, topographic contours, topographic textures, and overland-flow hydrography) revealed mine features. Visual interpretation of Landsat imagery and time-series analysis augmented the Lidar analysis revealing the temporal life cycle of mining. The combination of bare earth texture with time-lapse and time-series analyses revealed areas of disturbance for reclaimed mines. The spatiotemporal mapping approach proved to be most useful in identifying and characterizing the known mine pit and pile features, reclamation status, and areas of disturbance due to mining. Two mine waste volume estimation methods resulted in a 21% difference indicating that although the approach helps to map mine features and areas of mining disturbance for the purposes of mine land inventory, additional information is needed to improve the estimation of buried mine waste at reclaimed mine sites.

Is It Reliable to Extract Gully Morphology Parameters Based on High-Resolution Stereo Images? A Case of Gully in a “Soil-Rock Dual Structure Area”

The gully morphology parameter is an important quantitative index for monitoring gully erosion development. Its extraction method and accuracy evaluation in the “soil-rock dual structure area” are of great significance to the evaluation of gully erosion in this type of area. In this study, unmanned aerial vehicle (UAV) tilt photography data were used to evaluate the accuracy of extracting gully morphology parameters from high-resolution remote sensing stereoscopic images. The images data (0.03 m) were taken as the reference in Zhangmazhuang and Jinzhongyu small river valleys in Yishui County, Shandong Province, China. The accuracy of gully morphology parameters were extracted from simultaneous high-resolution remote sensing stereo images data (0.5 m) was evaluated, and the parameter correction model was constructed. The results showed that (1) the average relative errors of circumference (P), area (A), linear length of bottom (L1), and curve length of bottom (L2) are mainly concentrated within 10%, and the average relative errors of top width (TW) are mainly within 20%. (2) The average relative error of three-dimensional (3D) parameters such as gully volume (V) and gully depth (D) is mainly less than 50%. (3) The larger the size of the gully, the smaller the 3D parameters extracted by visual interpreters, especially the absolute value of the mean relative error (Rmean) of V and D. (4) A relationship model was built between the V and D values obtained by the two methods. When V and D were extracted from high-resolution remote sensing stereo images, the relationship model was used to correct the measured parameter values. These findings showed that high-resolution remote sensing stereo images represents an efficient and convenient data source for monitoring gully erosion in a small watershed in a “soil-rock dual structure area”.

How Do Observational Astronomers Learn to Inspect Imaging Data

Astronomy is entering an unprecedented era of data collection. Upcoming large surveys will gather more data than ever before, generated at rates requiring real-time decision making. Looking ahead, it is inevitable that astronomers will need to rely more heavily on automated processes. Indeed, some instances have already arisen wherein the majority of the inspection process is automated. Visual discovery, performed traditionally by humans, is one key area where automation is now being integrated rapidly. Visual discovery comprises two aspects: (1) visual inspection, the skill associated with examining an image to identify areas or objects of interest; and (2) visual interpretation, the knowledge associated with the classification of the objects or features. Both skills and knowledge are vital for humans to perform visual discovery, however, there appears to have been limited investigation into how the skill of visual inspection in astronomy is acquired. In this work, we address this issue by setting out to identify the landscape within which observational astronomers develop the skills to perform visual inspection. We report on a survey of 70 professional observational astronomers, at various career stages and from different geographical regions. We found that between 63%and 73% of the astronomers surveyed had received formal and/or informal training in visual inspection of images, although formal training (21%) was less common than informal training (60%). Surprisingly, out of the 37% who did not recall having received training in visual inspection, 29% (20 astronomers) indicated that they provided training to others. This suggests the emergence of ``expertise without precedent'' where a first expert in the field provides a new way of achieving a task. These results, paired with a set of three pilot interviews, present a touchstone against which the training of future observational astronomers can be compared.

Taxonomic and functional diversity of foliage‐dwelling ants in conventional and organic agroecosystems in southern Brazil: A landscape perspective

Abstract In modern landscapes, understanding how land use and land cover, specifically agriculture, affect biodiversity and ecological functions is essential for conservation. According to Brazilian law, agroecosystems must include areas designated for biodiversity conservation called Legal Reserves. We evaluated the taxonomic and functional diversity of foliage‐dwelling ants in Legal Reserves of agroecosystems in the south of Rio Grande do Sul, Brazil. We aimed to understand how land use and land cover at different landscape scales affect the ant fauna. We sampled ant communities in organic and conventional agroecosystems using passive and active sampling methods (arboreal pitfall traps and active sample, respectively). For the analysis of functional diversity, we measured morphological attributes related to the ecological strategies of ants, such as perception of visual cues, foraging and resource acquisition. We characterized the landscape by visual interpretation of high‐resolution satellite imagery within a 100 m, 500 m and 1 km radius of the sampling areas. We did not identify any effects of cultivation systems on ant fauna. However, we found that landscape characteristics, such as the proportion of natural forest within 100 m, positively affected functional richness. Additionally, within 500 m, natural forest coverage, along with the proportion of cultivation, positively influenced ant incidence. We emphasize the function of Legal Reserves around farms, of serving as support to maintain the presence of ants and the diversity of ecological functions performed by these organisms. We also emphasize that new studies in the region should investigate how a higher homogenization of cultures and cultivation systems, including other phyto‐physiognomies, influence the myrmecofauna of these environments.

A novel GAN-based three-axis mutually supervised super-resolution reconstruction method for rectal cancer MR image

Background and objectiveThis study aims to enhance the resolution in the axial direction of rectal cancer magnetic resonance (MR) imaging scans to improve the accuracy of visual interpretation and quantitative analysis. MR imaging is a critical technique for the diagnosis and treatment planning of rectal cancer. However, obtaining high-resolution MR images is both time-consuming and costly. As a result, many hospitals store only a limited number of slices, often leading to low-resolution MR images, particularly in the axial plane. Given the importance of image resolution in accurate assessment, these low-resolution images frequently lack the necessary detail, posing substantial challenges for both human experts and computer-aided diagnostic systems. Image super-resolution (SR), a technique developed to enhance image resolution, was originally applied to natural images. Its success has since led to its application in various other tasks, especially in the reconstruction of low-resolution MR images. However, most existing SR methods fail to account for all anatomical planes during reconstruction, leading to unsatisfactory results when applied to rectal cancer MR images. MethodsIn this paper, we propose a GAN-based three-axis mutually supervised super-resolution reconstruction method tailored for low-resolution rectal cancer MR images. Our approach involves performing one-dimensional (1D) intra-slice SR reconstruction along the axial direction for both the sagittal and coronal planes, coupled with inter-slice SR reconstruction based on slice synthesis in the axial direction. To further enhance the accuracy of super-resolution reconstruction, we introduce a consistency supervision mechanism across the reconstruction results of different axes, promoting mutual learning between each axis. A key innovation of our method is the introduction of Depth-GAN for synthesize intermediate slices in the axial plane, incorporating depth information and leveraging Generative Adversarial Networks (GANs) for this purpose. Additionally, we enhance the accuracy of intermediate slice synthesis by employing a combination of supervised and unsupervised interactive learning techniques throughout the process. ResultsWe conducted extensive ablation studies and comparative analyses with existing methods to validate the effectiveness of our approach. On the test set from Shanxi Cancer Hospital, our method achieved a Peak Signal-to-Noise Ratio (PSNR) of 34.62 and a Structural Similarity Index (SSIM) of 96.34 %. These promising results demonstrate the superiority of our method.