Image Analysis Research Articles

Automated Lymph Node Classification and Gastric Cancer Segmentation in CT Images Using Advanced Segmentation Techniques

The growing number of cases of stomach cancer makes it even more important to have accurate and quick diagnosis tools. This study uses cutting edge image processing techniques to try to solve this problem. The proposed method study uses a wide range of CT pictures that show different types of patients and imaging conditions. The method used to make sure the reliability of later studies includes pre-processing steps like noise reduction and intensity adjustment. For lymph node classification, the method created a complex program with advanced features that made it easier for lymph nodes to be automatically found and put into groups in the pictures. Similarly, the method used cutting-edge segmentation tools to correctly separate cancerous areas in the stomach cancer segmentation. The suggested method has been thoroughly tested using standard criteria, showing that it can accurately classify lymph nodes and divide stomach cancer into segments. Comparing the deep learning method to existing ones showed that it was better, showing that it has the ability to make a big difference in how medical picture analysis is done now. As part of the conversation, the proposed method strengths and weaknesses were emphasized, along with how the results were interpreted. The results show that the deep learning model could have a positive effect on patients by making diagnoses more accurately, making examination easier for the doctors, and eventually better patient outcomes. The proposed method for automatic lymph node labelling and the stomach cancer segmentation system are both exciting steps forward in the field of medical image analysis. The paper study's strong points and high levels of accuracy using deep learning network show that this approach has the ability to completely change how stomach cancer is diagnosed, leading to better patient care and results.

Deep learning enables the use of ultra-high-density array in DNBSEQ.

DNBSEQ employs a patterned array to facilitate massively parallel sequencing of DNA nanoballs (DNBs), leading to a considerable boost in throughput. By employing the ultra-high-density (UHD) array with an increased density of DNB binding sites, the throughput of DNBSEQ can be further expanded. However, the typical imaging system of the DNBSEQ sequencer is unable to resolve adjacent DNBs spaced smaller than the resolution limit, resulting in poor base-calling performance of the UHD array and hindering its practical application. In this study, we propose a deep-learning-based DNB image super-resolution network named DNBSRN to address this problem. DNBSRN has a specifically designed structure for DNB images and employs a histogram-matching-based preprocessing approach. For the eight DNB image datasets generated from the DNBSEQ sequencer using UHD arrays with 360nm pitch, the base-calling performances are significantly improved after super-resolution reconstruction by DNBSRN and reached a comparable level to those of the regular density array. In terms of reconstruction speed, DNBSRN takes only 7.61ms for an input image with 500 × 500 pixels, which minimizes its influence on throughput. Furthermore, compared with state-of-the-art super-resolution networks, DNBSRN demonstrates superior performance in terms of both the quality and speed of DNB image reconstruction. DNBSRN successfully addresses the DNB image super-resolution task. Integrating DNBSRN into the image analysis workflow of DNBSEQ will allow for the application of UHD array, hence enabling a considerable improvement in throughput as well as tremendous savings in unit reagent cost.

A semi-automated method using object-based image analysis (OBIA) to detect and enumerate beluga whales in summer from very high-resolution (VHR) satellite imagery.

Very high-resolution (VHR) satellite imagery has proven to be useful for detection of large to medium cetaceans, such as odontocetes and offers some significant advantages over traditional detection methods. However, the significant time investment needed to manually read satellite imagery is currently a limiting factor to use this method across large open ocean regions. The objective of this study is to develop a semi-automated detection method using object-based image analysis to identify beluga whales (Delphinapterus leucas) in open water (summer) ocean conditions in the Arctic using panchromatic WorldView-3 satellite imagery and compare the detection time between human read and algorithm detected imagery. The false negative rate, false positive rate, and automated count deviation were used to assess the accuracy and reliability of various algorithms for reading training and test imagery. The best algorithm, which used spectral mean and texture variance attributes, detected no false positives and the false negative rate was low (<4%). This algorithm was able to accurately and reliably identify all the whales detected by experienced readers in the ice-free panchromatic image. The autodetection algorithm does have difficulty separately identifying whales that are perpendicular to one another, whales below the surface, and may use multiple segments to define a whale. As a result, for determining counts of whales, a reader should manually review the automated results. However, object-based image analysis offers a viable solution for processing large amounts of satellite imagery for detecting medium-sized beluga whales while eliminating all areas of the imagery which are whale-free. This algorithm could be adapted for detecting other cetaceans in ice-free water.

CardiLect: A combined cross-species lectin histochemistry protocol for the automated analysis of cardiac remodelling.

Cardiac remodelling, a crucial aspect of heart failure, is commonly investigated in preclinical models by quantifying cardiomyocyte cross-sectional area (CSA) and microvascular density (MVD) via histological methods, such as immunohistochemistry. To achieve this, optimized protocols are needed, and the species specificity is dependent on the antibody used. Lectin histochemistry offers several advantages compared to antibody-based immunohistochemistry, including as cost-effectiveness and cross-species applicability. Direct comparisons between the two methods are lacking from the literature. In this study, we compared antibody- and lectin-based methods for the histological assessment of cardiomyocyte CSA (with the use of anti-laminin and wheat germ agglutinin [WGA]) and microvascular density (utilizing anti-CD31 and isolectin B4 [ILB4]) using different embedding and antigen/carbohydrate retrieval techniques. Here, we describe a detailed, easy-to-use combined lectin histochemistry protocol (WGA and ILB4, 'CardiLect' protocol) for the histological assessment of cardiac remodelling. The lectin-based approach has been evaluated on a cross-species basis, and its efficacy has been demonstrated in zebrafish, rodents, large animals and human samples. We provide an ImageJ script ('CardiLect Analyser') for automated image analysis, validated in a preclinical heart failure model by correlating histological parameters with echocardiographic findings. CSA showed a significant positive correlation with left ventricular (LV) mass (P=0.0098, rS=0.7545) and significant negative correlation with markers of systolic function, such as ejection fraction (EF) (P=0.0402, rS=-0.6364). Microvascular density showed significant negative correlation with LV mass (P=0.0055, rS=-0.7622) and significant positive correlation with EF (P=0.0106, rS=0.7203). The described combined lectin histochemistry protocol with the provided ImageJ script is an easy-to-use, cost-effective, cross-species approach for the histological assessment of cardiac remodelling.

Abstract A055: Tumor muscle invasion promotes tumor heterogeneity and normal muscle reprogramming

Abstract Aggressive human prostate tumors traverse a contractile smooth muscle pseudo-capsule, in a process termed extracapsular extension (ECE), defining the pT3 pathologic stage associating with increased biochemical recurrence, bone metastases, and cancer-specific mortality. While T3 lesions can be detected by non-invasive mpMRI imaging, how the tumor invades into and through the contractile muscle is unknown. Using a mouse xenograft model system of ECE, we investigated the transcriptomic consequences of human tumor invasion into and through the contractile smooth muscle of the mouse diaphragm. Both human and mouse gene transcripts were documented and analyzed. Tumor cells were intraperitoneally injected into male NSG mice and 6 weeks later, three tumor compartments (pre-invasive, muscle-invasive, and super-invasive cells that had reached the superior diaphragm surface) were analyzed for differential bulk human and mouse gene expression. Whole genome transcriptomic sequencing and GO pathway analysis revealed that approximately 414 human genes were differentially expressed between the non- invasive, muscle-resident, and super-invasive tumor populations and at least 5 enriched pathways were involved. The unique expression gene patterns of muscle-resident tumor cells were reversible when compared to the pre-invasive and super-invasive expression patterns. Significant increases in g-H2AX and nuclear deformation were observed in the muscle-resident tumor clusters as compared to the non-invasive tumor mass. No differences in tumor cell proliferation, as detected by Ki67 staining, were found between the tumor clusters in the three compartments. Immunohistochemistry staining for a damage response cytokeratin (KRT6A) and integrin (CD49f) revealed heterogeneity within the muscle resident tumors as compared to the non- invasive cells. Taken together, these data suggest a hostile contractile muscle environment elicits specific responses by the invading tumor. Single cell analysis within each of the tumor compartments is currently underway to define the spatial heterogeneity of gene expression in invasive tumor cell clusters within the contractile muscle. A dynamic reciprocity of the tumor/muscle microenvironment is suggested by the analysis of the bulk transcriptomic sequencing demonstrating a significant increase in mouse muscle bio-synthetic gene transcription as a consequence of the human tumor penetrating the tissue. Taken together, these data indicate that human tumors, during the act of traversing the contractile muscle layer, respond to this unique microenvironment by transiently altering transcription, while sustaining nuclear damage which results in the reprogramming of the muscle. This new information suggests that novel tumor or muscle biomarkers might indicate early muscle invasion events and assist new high-resolution image analysis for precision diagnostic and/or therapeutic decisions. (Supported in part by P30 CA23074; F30 CA143924, UACC Team Science Award). Citation Format: Kendra D Marr, Beatrice S Knudsen, Rafael Sainz, Kelvin W Pond, Noel E Warfel, Belinda E Sun, Anne E Cress. Tumor muscle invasion promotes tumor heterogeneity and normal muscle reprogramming [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr A055.

PunctaFinder: an algorithm for automated spot detection in fluorescence microscopy images.

Fluorescence microscopy has revolutionised biological research by enabling the visualisation of subcellular structures at high resolution. With the increasing complexity and volume of microscopy data, there is a growing need for automated image analysis to ensure efficient and consistent interpretation. In this study, we introduce PunctaFinder, a novel Python-based algorithm designed to detect puncta, small bright spots, in raw fluorescence microscopy images without image denoising or signal enhancement steps. Furthermore, unlike other available spot detectors, PunctaFinder not only detects puncta, but also defines the cytoplasmic region, making it a valuable tool to quantify target molecule localisation in cellular contexts. PunctaFinder is a widely applicable punctum detector and size estimator, as evidenced by its successful detection of Atg9-positive vesicles, lipid droplets, aggregates of a destabilised luciferase mutant, and the nuclear pore complex. Notably, PunctaFinder excels in detecting puncta in images with a relatively low resolution and signal-to-noise ratio, demonstrating its capability to identify dim puncta and puncta of dynamic target molecules. PunctaFinder reliably detects puncta in fluorescence microscopy images where automated analysis was not possible before, providing researchers with an efficient and robust method for punctum quantification in fluorescence microscopy images.

Use of Data Augmentation and Fine-tuned Transfer Learning Models for Classification of Cervical Cancer

Innovative strategies for early and accurate diagnosis of cervical cancer continue to be a global health priority. In this research, we present a powerful framework for improving cervical cancer classification by combining data augmentation methods with tailored transfer learning models. The lack of properly labelled medical data is a major roadblock to developing reliable diagnosis models. The geometric transformations, contrast tweaks, and noise addition that we use to increase the dataset artificially help us overcome this difficulty. The model's robustness is improved as a result of the additional exposure to diverse real-world circumstances provided by this supplemented dataset. We use a transfer learning strategy based on pre-trained convolutional neural networks (CNNs) to harness the potential of deep learning. These networks are well-suited for medical image analysis due to their ability to quickly and accurately identify features in a wide variety of images. Our expanded cervical cancer dataset is used to fine-tune these algorithms for their diagnostic purpose. Our experiments show that utilising both data augmentation and fine-tuned transfer learning considerably raises the accuracy of cervical cancer categorization. In terms of sensitivity, specificity, and overall accuracy, the model performs at a state-of-the-art level. This suggests that it may be useful in the diagnosis and early identification of cervical cancer. Additionally, our method demonstrates the significance of efficient data utilisation and model adaptability in the field of medicine. Our method is scalable and affordable, and it can be easily implemented in healthcare settings since we handle the problem of data scarcity and take advantage of the features of pre-trained deep learning models.

Abstract A031: Harnessing the potential of circulating tumour cells for precision diagnostics via multiplexed imaging of the levels and subcellular localizations of over 50 cell identity, state and signaling markers per cell

Abstract Due to their longitudinal accessibility through liquid biopsy, circulating tumour cells (CTCs) hold huge potential as sources of molecular information capable of underpinning precision diagnostics in cancer. Yet key methodological limitations have thus far constrained the practical utility of CTCs as diagnostic analytes, including: a) reliance on positive CTC selection strategies that can bias against distinctive CTC subpopulations (e.g. low EpCAM, low size, low density), and; b) standard fluorescence imaging methods that capture only 4-6 molecular markers per cell. Since 3-4 markers are required to identify CTCs with even moderate confidence, such low marker plexity acutely restricts derivation of additional molecular insights about the cellular states and molecular signals that may be driving disease. Yet these insights are essential to guide selection of optimal therapies, to adapt therapeutic strategies as cycles of resistance arise, and even to stratify patients to the trials needed to expand the therapeutic arsenal. To overcome these challenges and harness the potential of CTCs for precision diagnostics, we have developed a comprehensive pipeline for CTC isolation, storage and batch processing via multiplexed immunofluorescence imaging of 50+ molecular markers per cell. These markers illuminate CTC identity, state (epithelial vs mesenchymal vs stemness), fate (proliferation, death, senescence) and signalling (spanning ∼10 alternate ‘driver’ signalling pathways), providing truly unprecedented molecular insights per CTC, across heterogeneous CTC populations, and regarding potential therapeutic targets. Downstream single-cell quantitative image analyses enable robust but adaptable computational parsing of CTC identity, removing false-positive and false-negative instances otherwise resulting from standard 4-plex CTC definitions (e.g. DAPI, CD45, EpCAM, pan-Cytokeratin) and complementing our upstream use of negative selection methods that capture the full, unbiased diversity of CTCs. The remaining 40+ markers facilitate integrated analyses of relationships between CTC state, fate and signalling pathways, capturing not only protein and phospho-protein expression levels per cell, but also variations in subcellular localisation that constitute additional layers of functional regulation not captured by most omics methods. Excitingly, when combined with machine learning, our approach can discern therapy- resistant cell subpopulations with over 98% accuracy in models of prostate cancer therapy resistance. We are now assessing CTC diversity across patients ranging from treatment naïve to advanced castrate-resistance prostate cancers. Overall, our unique high-plexity analysis now enables interrogation of detailed states, signals and dependencies in each patient CTC, providing the foundations for next-generation precision diagnostics to continuously match individual patients to optimal therapies at each stage of their disease journey. We believe this marks a significant step towards harnessing the true potential of CTCs for precision diagnostics. Citation Format: John G Lock, Tim J Mann, Ye Zheng, Tanzila Khan, Daniel Neumann, Alexander James, Therese Becker, Tara Roberts. Harnessing the potential of circulating tumour cells for precision diagnostics via multiplexed imaging of the levels and subcellular localizations of over 50 cell identity, state and signaling markers per cell [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr A031.

Methodology For Extracting Poplar Planted Fields From Very High-Resolution Imagery Using Object-Based Image Analysis and Feature Selection Strategy

Abstract. Poplars (Populus sp.), a tree that grows rapidly species, are significant as industrial forest products. The delineation and monitoring of poplar cultivated areas are invaluable for decision-making processes. With the remote sensing technology, accurate detection of poplar planted areas could be determined much faster, more economically, and with minimum labor requirements. The objective of this research is to create a map of poplar plantations in the Sakarya region of Turkey utilizing Worldview-3 satellite imagery. Object-based image analysis (OBIA) through the application of the multi-resolution segmentation method (MRS) was employed to generate image segments, and then three prevailing machine learning algorithms, namely Support Vector Machine (SVM), Random Forest (RF) and Rotation Forest (RotFor) were implemented to produce LULC maps of the study area including 11 landscape features. The most effective and contributing object features that assure high separability between landscape features were determined using a filter-based Chi-square algorithm for the prediction models constructed with SVM, RF, and RotFor classifiers. Results revealed that the SVM classifier achieved the highest overall accuracy (91.73%) with 38 features out of 88 features, about 3% improvement compared to the other algorithms. According to the SHAP analysis, the IHS feature was the most effective one in the constructed RF model, followed by the CI (red edge), NDVI-1 and NDVI-2 vegetation indices.

Retraction Note: Cataract eye detection by optik image analysis using encoder basis Boltzmann architecture integrated with internet of things and data mining

Retraction Note: Cataract eye detection by optik image analysis using encoder basis Boltzmann architecture integrated with internet of things and data mining

Evaluation of podocytopathy in patients with no abnormality detected by light microscopy

ABSTRACT When podocytes are injured, they may detach leading to podocytopenia, which represents a critical step in the development of podocytopathy. . This study aimed to study the diagnostic role of colloidal iron stain in assessment of podoctyopathy. This is a retrospective study, conducted on 50 archival renal biopsies with the clinical diagnosis of steroid resistant or dependent nephrotic syndrome and with no abnormality detected by light microscopy (LM). Colloidal iron special stain was used to delineate the sialomucin coat of podocytes. Desmin immunohistochemical stain for assessment of podocyte injury was also applied. Electron microscopic (EM) examination was performed. Assessment of the mean percent of colloidal iron and desmin stained pixels in relation to the surface glomerular tuft area; i.e. mean percent of stained area (PSA) was done using image analysis software. Minimal change disease (MCD) was observed in 38/50 (76%) cases. Podocytopathy with detached podocytes was observed in 12/50 (24%) cases. Mean PSA in cases diagnosed as MCD was 19.17 ± 2.42 and was 18.95 ± 3.18 in cases diagnosed as podocytopathy with detached podocytes, while in control cases, mean PSA was 23.61 ± 2.18 which showed statistically significant difference between (p value = ≤0.001). Colloidal iron stain is an easy routine-friendly method for assessment of podocytopathy.

MalenoCare - Skin Cancer Detection and Prescription using CNN and ML

Machine learning (ML) and convolutional neural networks (CNNs) have driven significant advancements in healthcare, particularly in dermatology, by enabling the automation of diagnostic processes for skin cancer. Skin cancer, being one of the most common types of cancer, requires early detection and accurate classification to improve patient outcomes and reduce mortality rates. This review paper explores various studies and CNN-based tools that enhance skin cancer detection and support preventive care through image analysis. The paper also discusses the effectiveness of different CNN architectures, including VGG16, ResNet, and Inception, in achieving high accuracy rates in skin lesion classification. A combination of recent research findings, model evaluation metrics, and graphical data highlights the accuracy, interpretability, and real-world applications of ML models, offering insights into their potential for integration into clinical practice.

Different Evaluation Strategies of Oncological CT Examinations with Regard to Professional Experience: A Clinical Study Using Eye-tracking.

Contrast-enhanced CT is the standard imaging technique in oncological objectives. Rates of missed pathologies depend on work experience of the respective radiologists. Thus the aim of this study is to analyze the eye movements of professionals while reading CT images in order to evaluate whether the eye-fixation patterns and search strategies of experienced radiologists could explain higher detection rates of pathologies and whether such patterns can be learned.Anonymized images of 10 patients were presented to three medical students and six radiologists with different levels of work experience. During image analysis, ocular fixation positions were recorded using an eye-tracking software tool. The CT scans were analyzed retrospectively, considering the individual course of disease with the issue of successful detection of all pathologies. Visual attention and dwell time of ocular fixation on clinically important abnormalities or areas with pathological findings, general search patterns, and time efficiency were assessed. For statistical analysis, interobserver variability and accuracy of lesion detection were evaluated taking into account individual experience.The results revealed that observer sensitivity depends on work experience due to a more systematic order of inspection and a well-known course of disease, e.g. in case of metastatic spread. The areas of missed pathologies mostly included secondary findings. Inexperienced readers changed the stratification considerably more often and required more time for reporting or detecting pathologies.Our results suggest that experienced radiological physicians reduce their amount of missed findings by looking more systematically at images and by applying a more targeted inspection of clinically important regions. · CT interpretation by radiology residents is faster and less error-prone compared to postgraduate residents. · systematic image analysis is trainable. · engrams tend to be acquired through experience. · Kluge S, Schülke C, Ites HC et al. Different Evaluation Strategies of Oncological CT Examinations with Regard to Professional Experience: A Clinical Study Using Eye-tracking. Fortschr Röntgenstr 2024; DOI 10.1055/a-2452-2180.

Differential Glial Response and Neurodegenerative Patterns in CA1, CA3, and DG Hippocampal Regions of 5XFAD Mice

In this study, the distinct patterns of glial response and neurodegeneration within the CA1, CA3, and dentate gyrus (DG) regions of the hippocampus were examined in 5XFAD mice at 6 and 12 months of age. The primary feature of this transgenic mouse model is the rapid onset of amyloid pathology. We employed quantitative assessments via immunohistochemistry, incorporating double staining techniques, followed by observation with light microscopy and subsequent digital analysis of microscopic images. We identified significantly increased Aβ deposition in these three hippocampal regions at 6 and 12 months of transgenic mice. Moreover, the CA1 and CA3 regions showed higher vulnerability, with signs of reactive astrogliosis such as increased astrocyte density and elevated GFAP expression. Additionally, we observed a significant rise in microglia density, along with elevated inflammatory markers (TNFα) in these hippocampal regions. These findings highlight a non-uniform glial and neuronal response to Aβ plaque deposition within the hippocampal regions of 5xFAD mice, potentially contributing to the neurodegenerative and memory deficit characteristics of Alzheimer’s disease in this model.

Analyzing Multimodal Construction of China’s Image Based on the Illustrated Articles in The Economist

The shaping of national image is a two-way process, including “self-shaping” and “shaping by others”. Magazines are one of the important channels for disseminating information. Taking The Economist’s China-related reports in 2023 as a case, this article focuses on 195 reports on politics and diplomacy as well as economy, conducting quantitative and qualitative analyses of the visual images from the perspective of visual grammar, and further exploring the multimodal construction of China's image in The Economist. The research shows that the images in The Economist fully realize the representational meaning, interactive meaning and compositional meaning. In terms of the representational meaning, narrative representation is the major process. Images that reflect the reactional process, action process and symbolic process are used more frequently. In terms of the interactive meaning, “offer” images, long shot images and images with horizontal perspective appear more frequently. In addition, information value, salience, and framing are interrelated to achieve the compositional meaning. The political and diplomatic image of China constructed by The Economist is that China is a country with increasing power, which is authoritarian internally and threatening externally. In terms of economy, China is portrayed as a country with a fast growing economy which can withstand the global economic crisis, but its economic prospects are not optimistic, and the “China threat theory” still exists.

Predicting post-hepatectomy liver failure in patients with hepatocellular carcinoma: nomograms based on deep learning analysis of gadoxetic acid-enhanced MRI.

This study aimed to develop nomograms for predicting post-hepatectomy liver failure (PHLF) in patients with hepatocellular carcinoma (HCC), using deep learning analysis of Gadoxetic acid-enhanced hepatobiliary (HBP) MRI. This retrospective study analyzed patients who underwent gadoxetic acid-enhanced MRI and hepatectomy for HCC between 2016 and 2020 at two referral centers. Using a deep learning algorithm, volumes and signal intensities of whole non-tumor liver, expected remnant liver, and spleen were measured on HBP images. Two multivariable logistic regression models were formulated to predict PHLF, defined and graded by the International Study Group of Liver Surgery: one based on whole non-tumor liver measurements (whole liver model) and the other on expected remnant liver measurements (remnant liver model). The models were presented as nomograms and a web-based calculator. Discrimination performance was evaluated using the area under the receiver operating curve (AUC), with internal validation through 1000-fold bootstrapping. The study included 1760 patients (1395 male; mean age ± standard deviation, 60 ± 10 years), with 137 (7.8%) developing PHLF. Nomogram predictors included sex, gamma-glutamyl transpeptidase, prothrombin time international normalized ratio, platelets, extent of liver resection, and MRI variables derived from the liver volume, liver-to-spleen signal intensity ratio, and spleen volume. The whole liver and the remnant liver nomograms demonstrated strong predictive performance for PHLF (optimism-corrected AUC of 0.78 and 0.81, respectively) and symptomatic (grades B and C) PHLF (optimism-corrected AUC of 0.81 and 0.84, respectively). Nomograms based on deep learning analysis of gadoxetic acid-enhanced HBP images accurately stratify the risk of PHLF. Question Can PHLF be predicted by integrating clinical and MRI-derived volume and functional variables through deep learning analysis of gadoxetic acid-enhanced MRI? Findings Whole liver and remnant liver nomograms demonstrated strong predictive performance for PHLF with the optimism-corrected area under the curve of 0.78 and 0.81, respectively. Clinical relevance These nomograms can effectively stratify the risk of PHLF, providing a valuable tool for treatment decisions regarding hepatectomy for HCC.

A-PoRM SIPLAS: Assessing The Post-Disaster Recovery Of Mangrove Forest In Siargao Island Protected Landscape And Seascape (SIPLAS) Area Using Sentinel-1 SAR Data

Abstract. Tropical cyclones and typhoons pose significant threats to coastal ecosystems, particularly mangrove forests. Assessing the extent of mangrove damage and monitoring recovery post-disaster is vital for effective conservation and management strategies. Space-based observations have become increasingly common in monitoring forest damage in tropical regions. However, the predominant methods for change detection rely on satellite optical imagery, which faces challenges in humid tropical zones. Technological advancements have led to the development of detection techniques utilizing synthetic aperture radar (SAR) to address cloud cover issues. The study specifically targets Siargao Island Protected Landscape and Seascape (SIPLAS), the largest contiguous forest in the Philippines, from destruction caused by Super typhoon Rai (Odette). The study utilized SAR time-series data, unsupervised log ratio change detection method and Cumulative Sum (CuSum) analysis via Google Earth Engine (GEE). Analysis of Sentinel-1 GRD images found VV polarization had the highest accuracy at 92% and an F-measure of 0.92. VH polarization had 87.5% accuracy and an F-measure of 0.87, while combining VV and VH and VV/VH achieved the same results of 80.5% overall accuracy with an F-measure of 0.82. With that, results indicated that 53% of mangroves in SIPLAS area were damaged, with 38.05% regenerating by 2022 and 56.75% by 2023, notably after September 16, 2022 detected from CuSum analysis using the VV polarization. These findings substantiate the robustness of the data and methodologies employed in monitoring mangrove resilience and supporting conservation initiatives within the SIPLAS area.

Possible inhibition effects of resveratrol on pancreatic tumorigenesis in the azaserine-rat model.

Resveratrol, which is thought to have a preventive effect on the formation of different types of cancer, is abundant in grapes and other foods. Resveratrol has been shown to have anti-cancer effects by in vitro andin vivo studies, however this is the first time its effect on atypical acinar cell foci (AACF), known as precursor forms of pancreatic carcinoma, has been experimentally investigated. Male Sprague Dawley rats, each consisting of 5 experimental groups (Cont, AzCont, AzRes10, AzRes15, and AzRes20), 10 of which were 14 days old, were used in the study. In the azaserine groups (AzCont, AzRes10, AzRes15, and AzRes20), it was investigated how the development of Atypical Cell Foci (AACF) resulting from intraperitoneal injection (i.p.) of azaserine (30 mg/kg bw) in 14-day-old rats was affected by dietary restoration. Male rats in the resveratrol group (AzRes10, AzRes15, and AzRes20) were fed diets containing 10%, 15%, or 20% mmol resveratrol for an 8-month experimental period 1 week after the last azaserine injection. Pancreas preparations prepared from histological sections were examined for AACF burden and analyzed via a video image analyzer. One-way analysis of variance (ANOVA) non-parametric statistical analyses were performed to test whether there was a difference between the averages of the experimental and control groups. The AACF load in the azaserine group (AzCont) compared to the control group (Cont) was found to be statistically significant in all categories (p < 0.05). The calculated estimated mean AACF volume (mm3) values and the AACF ratio as a percentage of the calculated organ volume were statistically significantly lower in all resveratrol groups (AzRes10, AzRes15, and AzRes20) compared to the azaserine control group (AzCont). The calculated estimated mean AACF volume (mm3) values and the AACF ratio as a percentage of the calculated organ volume were statistically significantly lower in all resveratrol groups (AzRes10, AzRes15, and AzRes20) compared to the azaserine control group (AzCont) (p < 0.05). In addition, the calculated estimated mean AACF diameter (mm) in the AzRes10 and AzRes15 groups, in the AzRes15 group the calculated estimated mean AACF number in the whole organ and the calculated average AACF number per unit area were found to be statistically significant compared to the azaserine control group (AzCont) (p < 0.05). According to the results of our study, it has been shown that atypical acinar cell foci (AACF) formed in the exocrine pancreas of rats with azaserine can be reduced by a diet containing resveratrol. It was determined that the tumor burden decreased statistically significantly (p ≤ 0.05) in resveratrol-treated rats. Accordingly, it is thought that the inhibitory effects of resveratrol may contribute to studies that reduce the occurrence of pancreatic cancer.

Cancer-Associated Lymphoid Aggregates in Histology Images: Manual and Deep Learning-Based Quantification Approaches.

Quantification of lymphoid aggregates including tertiary lymphoid structures (TLS) with germinal centers in histology images of cancer is a promising approach for developing prognostic and predictive tissue biomarkers. In this article, we provide recommendations for identifying lymphoid aggregates in tissue sections from routine pathology workflows such as hematoxylin and eosin staining. To overcome the intrinsic variability associated with manual image analysis (such as subjective decision-making, attention span), we recently developed a deep learning-based algorithm called HookNet-TLS to detect lymphoid aggregates and germinal centers in various tissues. Here, we additionally provide a guideline for using manually annotated images for training and implementing HookNet-TLS for automated and objective quantification of lymphoid aggregates in various cancer types.

Development of a Hybrid Medical Softbot for Enhanced Decision Making using AI-tools

Health is wealth. The maintenance of health is of paramount importance. Due to increasing environmental decay, human health is threatened and therefore requires maintenance. Healthcare providers are few in number and therefore may not be able to cater for everyone. They tend to get fatigued because of the volume of work required on daily basis. This terribly affects their decision making and may lead to death as a result of wrong diagnosis or recommendations. This is the motivation behind this work; design and implementation of a hybrid medical softbot for enhanced decision making. This work was designed with the aid of machine learning algorithms for image analysis and classification and a rule based system that accepts input in the form of symptoms from user to make expert diagnosis and recommendation. Object oriented analysis and design methodology was employed in the analysis and design phase. The result is a hybrid softbot capable of analyzing and classifying X-ray images and giving expert diagnosis for patients.