Jaccard Index Research Articles

Identifying important nodes of hypergraph: An improved PageRank algorithm

Abstract Hypergraphs can accurately capture complex higher-order relationships, but it is challenging to identify their important nodes. In this paper, an improved PageRank(ImPageRank) algorithm is designed to identify important nodes in a directed hypergraph. The algorithm introduces the Jaccard similarity of directed hypergraphs. By comparing the numbers of common neighbors between nodes with the total number of their neighbors, the Jaccard similarity measure takes into account the similarity between nodes that are not directly connected, and can reflect the potential correlation between nodes. An improved SI model in directed hypergraph is proposed, which considers nonlinear propagation mode and more realistic propagation mechanism. In addition, some important node evaluation methods are transferred from undirected hypergraphs and applied to directed hypergraphs. Finally, the ImPageRank algorithm is used to evaluate the performance of SI model, network robustness and monotonicity. Simulations of real networks demonstrate the excellent performance of the proposed algorithm and provide a powerful framework for identifying important nodes in directed hypergraphs.

From Data Silos to Health Records Without Borders: A Systematic Survey on Patient-Centered Data Interoperability

The widespread use of electronic health records (EHRs) and healthcare information systems (HISs) has led to isolated data silos across healthcare providers, and current interoperability standards like FHIR cannot address some scenarios. For instance, it cannot retrieve patients’ health records if they are stored by multiple healthcare providers with diverse interoperability standards or the same standard but different implementation guides. FHIR and similar standards prioritize institutional interoperability rather than patient-centered interoperability. We explored the challenges in transforming fragmented data silos into patient-centered data interoperability. This research comprehensively reviewed 56 notable studies to analyze the challenges and approaches in patient-centered interoperability through qualitative and quantitative analyses. We classified the challenges into four domains and categorized common features of the propositions to patient-centered interoperability into six categories: EMR integration, EHR usage, FHIR adaptation, blockchain application, semantic interoperability, and personal data retrieval. Our results indicated that “using blockchain” (48%) and “personal data retrieval” (41%) emerged as the most cited features. The Jaccard similarity analysis revealed a strong synergy between blockchain and personal data retrieval (0.47) and recommends their integration as a robust approach to achieving patient-centered interoperability. Conversely, gaps exist between semantic interoperability and personal data retrieval (0.06) and between FHIR adaptation and personal data retrieval (0.08), depicting research opportunities to develop unique contributions for both combinations. Our data-driven insights provide a roadmap for future research and innovation.

Dynamics and hazards of pyroclastic avalanches at Etna volcano (Italy)

We present a multidisciplinary research aimed at quantifying the conditional probabilities for hazards associated with pyroclastic avalanches at Etna, which combines physical and numerical modeling of granular avalanches and probabilistic analysis. Pyroclastic avalanches are modeled using the depth-averaged model IMEX-SfloW2D, which is able to simulate the transient propagation and emplacement of granular flows generated by the collapse of a prescribed volume of granular material. Preliminary sensitivity analysis allowed us to identify the main controlling parameters of the dynamics, i.e. the total avalanche mass, the initial position of the collapsing granular mass (and the associated terrain morphology), the initial avalanche velocity, and the two rheological parameters which determine the mechanical properties of the flow. While the first two parameters can be considered as “scenario parameters” in the definition of the hazards, the initial velocity and the rheological parameters need to be calibrated. We therefore adopted a methodology for the statistical calibration of the physical model parameters based on field observations. We used data from the pyroclastic avalanche that occurred on February 10, 2022 at Etna, for which we had an accurate mapping of the deposit and some estimates of the total mass and the initial volume. We then run a preliminary ensemble of numerical simulations, with fixed initial volume and position, to calibrate the other input parameters. Based on the accuracy of the matching of the simulated and observed deposits (measured by the Jaccard Index), we extracted from the simulation ensemble a subsample of equally probable combinations of initial velocities and rheological parameters. We then built an ensemble of model input parameters, with varying (i) avalanche volumes, (ii) initial positions, (iii) velocity, and (iv) rheological coefficients. The initial volume range was chosen within the range of observed pyroclastic avalanches at Etna (i.e., between 0.1 and 3 × 106 m3), using a prescribed probability distribution extracted from the literature data. The initial positions have been chosen on the flanks of the South East Crater of Etna, with homogeneous spatial distribution. The initial velocity and the rheological coefficients were chosen from the subsample created with the calibration. Finally, a semi-automatic procedure (digital workflow) running the Monte Carlo simulation allowed us to produce the first probabilistic map of pyroclastic avalanche invasion at Etna. Such a map, conditional to the occurrence of a pyroclastic avalanche event, can be used to identify the hazardous areas of the volcano and to plan mitigation measures.

Overcoming annotation bottlenecks in underwater fish segmentation: a robust self-supervised learning approach

Accurate fish segmentation in underwater videos is challenging due to low visibility, variable lighting, and dynamic backgrounds, making fully-supervised methods that require manual annotation impractical for many applications. This paper introduces a novel self-supervised learning approach for fish segmentation using Deep Learning. Our model, trained without manual annotation, learns robust and generalizable representations by aligning features across augmented views and enforcing spatial-temporal consistency. We demonstrate its effectiveness on three challenging underwater video datasets: DeepFish, Seagrass, and YouTube-VOS, surpassing existing self-supervised methods and achieving segmentation accuracy comparable to fully-supervised methods without the need for costly annotations. Trained on DeepFish, our model exhibits strong generalization, achieving high segmentation accuracy on the unseen Seagrass and YouTube-VOS datasets. Furthermore, our model is computationally efficient due to its parallel processing and efficient anchor sampling technique, making it suitable for real-time applications and potential deployment on edge devices. We present quantitative results using Jaccard Index and Dice coefficient, as well as qualitative comparisons, showcasing the accuracy, robustness, and efficiency of our approach for advancing underwater video analysis.

Diversity and abundance of butterflies in urban areas of Bankura district, Bankura, West Bengal, India

ObjectiveDuring the present study, the biodiversity profile of butterflies was assessed on the Bankura Christian College campus, a large area with a huge amount of trees, bushes, rain forest, and undisturbed areas within the urban areas of Bankura district of West Bengal, India.MethodologyPresent study was carried out in Bankura Christian College campus, a relatively undisturbed area in Bankura town, of West Bengal, India. The findings presented here were based on seasonal diversity of Butterflies in the college campus. The species dominance of Butterflies was numerically measured on the basis of density, relative density, abundance, relative abundance, frequency, and relative frequency. To analyze the result, single factor ANOVA was performed in between five different seasons. Seasonal variation was analysed using individual rarefaction analysis. The Jaccard similarity index was also used to conduct hierarchical classical clustering.ResultA total of 47 genera and 59 species of butterflies that belongs to five families were recorded. The findings of the present study indicated a high abundance of the butterflies that belongs to Nymphalidae family comprising 13 Genera and 22 species.ConclusionIn the Monsoon season, the highest number of butterflies were recorded in comparison to other seasons. Different types of ecological indices were also calculated from observed species diversity of butterflies. Individual rarefaction analysis revealed that the probability of finding the greatest specimen was in monsoon season. Hierarchical cluster analysis among seasons based on the Jaccard Similarity index using Paired group (UPGMA) shows higher similarity of butterflies in between monsoon and Fall season.

Hybrid deep learning based stroke detection using CT images with routing in an IoT environment

ABSTRACT Stroke remains a leading global health concern and early diagnosis and accurate identification of stroke lesions are essential for improving treatment outcomes and reducing long-term disabilities. Computed Tomography (CT) imaging is widely used in clinical settings for diagnosing stroke, assessing lesion size, and determining the severity. However, the accurate segmentation and early detection of stroke lesions in CT images remain challenging. Thus, a Jaccard_Residual SqueezeNet is proposed for predicting stroke from CT images with the integration of the Internet of Things (IoT). The Jaccard_Residual SqueezeNet is the integration of the Jaccard index in Residual SqueezeNet. Firstly, the brain CT image is routed to the Base Station (BS) using the Fractional Jellyfish Search Pelican Optimization Algorithm (FJSPOA) and preprocessing is accomplished by median filter. Then, the skull segmentation is accomplished by ENet and then feature extraction is done. Lastly, Stroke is detected using the Jaccard_Residual SqueezeNet. The values of throughput, energy, distance, trust, and delay determined in terms of routing are 72.172 Mbps, 0.580J, 22.243 m, 0.915, and 0.083S. Also, the accuracy, sensitivity, precision, and F1-score for stroke detection are 0.902, 0.896, 0.916, and 0.906. These findings suggest that Jaccard_Residual SqueezeNet offers a robust and efficient platform for stroke detection.

A prospective multi-site translational study investigating the association of gut microbiome (GM) diversity with pathological complete response (pCR) after neoadjuvant treatment in early stage rectal and esophageal cancer.

TPS849 Background: The diversity of the GM is defined as the number and relative abundance distribution of distinct types of microorganisms colonizing within the gut. Studies have suggested that dysbiosis of the GM confers a predisposition to certain malignancies and impacts response to immunomodulating therapies. The influence of the GM diversity on the pathological response after neoadjuvant chemotherapy and radiotherapy is unclear. Some studies have suggested that the GM may offer predictive biomarkers for response to chemoradiation in rectal cancer. Other studies in early-stage rectal cancer patients indicated an association between GM diversity and pathological outcomes following neo-adjuvant therapy (NAT). We hypothesize that a more diverse GM constitution at baseline leads to an improved pathological response at the time of definitive surgery. Methods: We designed a cross-institutional translational study investigating the impact of the GM diversity on the efficacy of NAT in GI cancers by assessing its association with pathological response. The study population includes patients with early-stage rectal or esophageal cancer due to commence NAT (including chemotherapy and chemoradiation) who are planned for definitive surgery. Exclusion criteria includes prior allogenic tissue/solid organ transplantation and prior receipt of anti-cancer therapy. The study assessments include fecal sampling of the GM prior to NAT, upon completion and again six months post completion of therapy. Fecal samples are analysed by 16S RNA sequencing. Pathological response will be examined at time of surgery and patients will be classified as responders (complete pathological response) or non-responders. The primary endpoint of the study is to examine the association between the GM diversity and pathological response. Exploratory analysis will include the assessment of the association between cf-DNA and the GM diversity as well as an assessment of the association between cf-DNA at baseline and pCR. Species richness (Alpha Diversity) will be analysed using the Shannon diversity index and Jaccard similarity index will be used to calculate beta diversity. Following planned study recruitment, classification and clustering analysis will be performed with Principal Component Analysis (PCA) and Random Forest analysis. Logistic regression analysis adjusting for potential confounding factors will be employed to assess the primary endpoint of the association between GM and complete pCR in the final statistical analysis. Adjusted odds ratios (OR) and 95% confidence intervals will be presented. This trial accrued 11 patients between May 2023 and Sept 2024. Out of the 11 patients enrolled, 9 patients have undergone their planned surgery. We are expecting to have 30 patients accrued prior to Jan 2025.

Multilevel perception boundary-guided network for breast lesion segmentation in ultrasound images.

Automatic segmentation of breast tumors from the ultrasound images is essential for the subsequent clinical diagnosis and treatment plan. Although the existing deep learning-based methods have achieved considerable progress in automatic segmentation of breast tumors, their performance on tumors with similar intensity to the normal tissues is still not satisfactory, especially for the tumorboundaries. To accurately segment the non-enhanced lesions with more accurate boundaries, a novel multilevel perception boundary-guided network (PBNet) is proposed to segment breast tumors from ultrasoundimages. PBNet consists of a multilevel global perception module (MGPM) and a boundary guided module (BGM). MGPM models long-range spatial dependencies by fusing both intra- and inter-level semantic information to enhance tumor recognition. In BGM, the tumor boundaries are extracted from the high-level semantic maps using the dilation and erosion effects of max pooling; such boundaries are then used to guide the fusion of low- and high-level features. Additionally, a multi-level boundary-enhanced segmentation (BS) loss is introduced to improve boundary segmentation performance. To evaluate the effectiveness of the proposed method, we compared it with state-of-the-art methods on two datasets, one publicly available datasets BUSI containing 780 images and one in-house dataset containing 995 images. To verify the robustness of each method, a 5-fold cross-validation method was used to train and test the models. Dice score (Dice), Jaccard coefficients (Jac), Hausdorff Distance (HD), Sensitivity (Sen), and specificity(Spe) were used to evaluate the segmentation performance quantitatively. The Wilcoxon test and Benjamini-Hochberg false discovery rate (FDR) multi-comparison correction were then performed to assess whether the proposed method presents statistically significant performance ( ) difference comparing with existing methods. In addition, to comprehensively demonstrate the difference between different methods, the Cohen's d effect size and compound p-value (c-Pvalue) obtained with Fisher's method were alsocalculated. On the BUSI dataset, the mean Dice and Sen of PBNet was increased by 0.93% ( ) and 1.42% ( ), respectively, comparing against the corresponding suboptimal methods. On the in-house dataset, PBNet improved Dice, Jac and Spe by approximately 0.86% ( ), 1.42% ( ), and 0.1%, respectively, and reduced HD by 1.7% ( ) compared to the sub-optimal model. Comprehensively, in terms of all the evaluation metics, the performance of the proposed method significantly (c-Pvalue ) outperformed the others but the effect size was smaller than 0.2. Ablation results confirmed that MGPM is effective in distinguishing non-enhanced tumors, while BGM and BS loss are beneficial for refining tumor segmentationcontours. The proposed PBNet allows us to segment the non-enhanced breast lesions from ultrasound images with more accurate boundaries, which provides a valuable means for the subsequent clinicalapplications.

P-1980. Decoding the Complexity of Post COVID Conditions: Transcriptomic Analyses of Machine-Learning-Based Chronic Symptom Phenotypes

Abstract Background Post-COVID conditions (PCC, ‘Long COVID’) remain a military health concern, and symptom heterogeneity limits understanding of PCC pathogenesis. To address this, we used machine learning to define PCC phenotypes in Military Health System (MHS) beneficiaries. Here, we extend these findings to identify early transcriptomic profiles which may predict the development of these distinct symptom-based PCC phenotypes. (A) Machine learning techniques identified three groups of participants based on chronic symptom-clustering. Radial charts display the distribution of post-COVID condition (PCC) symptoms. Each plot’s axes represent different symptoms, with the length of each spoke indicating symptom prevalence. Shorter spokes indicate lower prevalence, while longer spokes indicate higher prevalence. (B) Clinical and demographic characteristics of 154 military health system beneficiaries by identified symptom-clusters. (C) Schematic representation of pathway enrichment analysis using the differentially expressed signature and pathway databases. (D-E) Jaccard similarity analysis identifies the similarity of leading-edge pathways within each symptom-cluster phenotype, with a result close to 1 indicating high similarity and close to 0 indicating significant differences. (F) Network analysis depicts pathway interactions between each symptom-cluster phenotype, with larger nodes representing clusters and smaller nodes indicating individual pathways. Grey lines indicate associations between clusters and pathways, revealing both overlapping and distinct pathway sets. Upregulated and downregulated pathways are highlighted in red and blue, respectively. (G) Non-overlapping leading-edge pathways are identified, with the size of each pathway node corresponding to the normalized enrichment score (NES), with an upward arrow and red color indicates upregulation, while a downward arrow and blue color denotes downregulation. Methods The EPICC cohort study evaluates SARS-CoV-2 infection in MHS beneficiaries. PCC phenotypes were identified using symptom data; transcriptome profiling was performed on early post-infection blood samples from participants who did and did not develop 3-month symptom PCC phenotypes. Participants with pre-COVID comorbidities were excluded to delineate gene responses for PCC from pre-existing illnesses. Comparison of transcriptomic signatures associated with symptom clusters involved utilizing Reactome, KEGG, BioCarta, and Hallmark pathway enrichment analysis. Hypergeometric p-values and the Jaccard coefficient (JC) assess pathway distinctiveness between symptom clusters and those with no PCC. Results We analyzed 154 participants with no reported comorbidities (Fig. 1), who completed a symptom survey 3-months after a SARS-CoV-2 positive test, and who had an early blood sample (median: 36 days post-infection). Clustering analysis stratified 3 distinct PCC symptom clusters: Sensory, Fatigue/difficulty thinking, and Difficulty breathing/exercise intolerance. Transcriptomic analysis noted post-infection signatures which predicted the development of each cluster (JC: 0.18–0.28). Leading-edge pathways included cytokine signaling in the Sensory cluster (Normalized Enrichment Score, NES=16.16, P < 0.001), nervous system development in the Fatigue/difficulty thinking cluster (NES=10.23, P < 0.001), and MYC Targets V1 in the Difficulty thinking/exercise intolerance cluster (NES=8.89, P < 0.001). Conclusion Distinct symptom-based PCC phenotypes are associated with different early post-infection gene expression pathways. These findings inform mechanistic hypotheses and the development of PCC prediction tools. These methods may be applied for prediction of post-acute sequelae in other pathogens. Disclosures Timothy Burgess, MD, MPH, AstraZeneca: The IDCRP and HJF were funded to conduct an unrelated phase III COVID-19 monoclonal antibody immunoprophylaxis trial as part of US Govt COVID Response Simon Pollett, MBBS, AstraZeneca: The IDCRP and HJF were funded to conduct an unrelated phase III COVID-19 monoclonal antibody immunoprophylaxis trial as part of US Govt COVID Response

Multi-scale channel attention U-Net: a novel framework for automated gallbladder segmentation in medical imaging

ObjectivesTo develop a novel automatic delineation model, the Multi-Scale Channel Attention U-Net (MCAU-Net) model, for gallbladder segmentation on CT images of patients with liver cancer.MethodsWe retrospectively collected the CT images from 120 patients with liver cancer, based on which ground truth was manually delineated by physicians. The images and ground truth constitute a dataset, which was proportionally divided into a training set (54%), a validation set (6%), and a test set (40%). Data augmentation was performed on the training set. Our proposed MCAU-Net model was employed for gallbladder segmentation and its performance was evaluated using Dice Similarity Coefficient (DSC), Jaccard Similarity Coefficient (JSC), Positive Predictive Value (PPV), Sensitivity (SE), Hausdorff Distance (HD), Relative Volume Difference (RVD), and Volumetric Overlap Error (VOE) metrics.ResultsOn the test set, MCAU-Net achieved DSC, JSC, PPV, SE, HD, RVD, and VOE values of 0.85 ± 0.22, 0.79 ± 0.23, 0.92 ± 0.14, 0.84 ± 0.23, 2.75 ± 0.98, 0.18 ± 0.48, and 0.22 ± 0.42, respectively. Compared to the control models, U-Net, SEU-Net and TransUNet, the MCAU-Net improved DSC 0.06, 0.04 and 0.06, JSC by 0.09, 0.06 and 0.09, PPV by 0.08, 0.08 and 0.05, SE by 0.05,0.05 and 0.07, and reduced HD by 0.45, 0.28 and 0.41, RVD by 0.07, 0.03 and 0.07, VOE by 0.04, 0.02 and 0.08 respectively. Qualitative results revealed that MCAU-Net produced smoother and more accurate boundaries, closer to the expert delineation, with less over-segmentation and under-segmentation and improved robustness.ConclusionsThe MCAU-Net model significantly improves gallbladder segmentation on CT images. It satisfies clinical requirements and enhances the efficiency of physicians, particularly in segmenting complex anatomical structures.

Deep Learning-Powered CT-Less Multitracer Organ Segmentation From PET Images: A Solution for Unreliable CT Segmentation in PET/CT Imaging.

The common approach for organ segmentation in hybrid imaging relies on coregistered CT (CTAC) images. This method, however, presents several limitations in real clinical workflows where mismatch between PET and CT images are very common. Moreover, low-dose CTAC images have poor quality, thus challenging the segmentation task. Recent advances in CT-less PET imaging further highlight the necessity for an effective PET organ segmentation pipeline that does not rely on CT images. Therefore, the goal of this study was to develop a CT-less multitracer PET segmentation framework. We collected 2062 PET/CT images from multiple scanners. The patients were injected with either 18F-FDG (1487) or 68Ga-PSMA (575). PET/CT images with any kind of mismatch between PET and CT images were detected through visual assessment and excluded from our study. Multiple organs were delineated on CT components using previously trained in-house developed nnU-Net models. The segmentation masks were resampled to coregistered PET images and used to train 4 different deep learning models using different images as input, including noncorrected PET (PET-NC) and attenuation and scatter-corrected PET (PET-ASC) for 18F-FDG (tasks 1 and 2, respectively using 22 organs) and PET-NC and PET-ASC for 68Ga tracers (tasks 3 and 4, respectively, using 15 organs). The models' performance was evaluated in terms of Dice coefficient, Jaccard index, and segment volume difference. The average Dice coefficient over all organs was 0.81 ± 0.15, 0.82 ± 0.14, 0.77 ± 0.17, and 0.79 ± 0.16 for tasks 1, 2, 3, and 4, respectively. PET-ASC models outperformed PET-NC models (P < 0.05) for most of organs. The highest Dice values were achieved for the brain (0.93 to 0.96 in all 4 tasks), whereas the lowest values were achieved for small organs, such as the adrenal glands. The trained models showed robust performance on dynamic noisy images as well. Deep learning models allow high-performance multiorgan segmentation for 2 popular PET tracers without the use of CT information. These models may tackle the limitations of using CT segmentation in PET/CT image quantification, kinetic modeling, radiomics analysis, dosimetry, or any other tasks that require organ segmentation masks.

Clinical significance criteria in the ICSD and DSM sleep disorder classifications: a content overlap analysis using the Jaccard index.

Both the International Classification of Sleep Disorders (ICSD) and the sleep-wake disorders section of the Diagnostic and Statistical Manual of Mental Disorders (DSM) emphasize the importance of clinical judgment in distinguishing the normal from the pathological in sleep medicine. The fourth edition of the DSM (DSM-IV, 1994) introduced the clinical significance criterion (CSC) to standardize this judgment and enhance diagnostic reliability. This review conducts a theoretical and historical content analysis of CSC presence, frequency, and formulation in the diagnostic criteria of sleep disorders. Three steps were undertaken: i) selecting main sleep disorder classifications and main disorders within them, ii) systematically extracting CSC-related content, and iii) analyzing the evolution of CSC presence through quantification (Jaccard index) and visualization (radar plots). The fifth edition, text revision of the DSM (DSM-5-TR) includes CSC in five of nine main sleep disorders, compared to two in the third edition, text revision of the ICSD (ICSD-3-TR) (circadian rhythm sleep-wake disorder and nightmare disorder). The overlap between DSM-5-TR and ICSD-3-TR is moderate (Jaccard index = 0.40). The overlap between DSM versions is higher (0.53). Conversely, ICSD revisions exhibit minimal similarity (0.16). Radar plots reveal a gradual increase in CSC use within ICSD versions. These results highlight the variable application of CSC in sleep disorder classifications. Universal CSC inclusion may not be essential, but systematic discussion of its potential use can help refine diagnostic criteria. This refinement is important for accurately diagnosing sleep disorders and better differentiating the normal from the pathological, a major challenge in sleep medicine.

3D Segmentation of Whole Lung and Metastatic Lung Nodules Using Adaptive Region Growing and Shape-based Morphology.

Early diagnosis of primary and metastatic lung nodules is critical for effective therapeutic planning. Manual delineation of lung nodules is not time-efficient and is prone to human error as well as interobserver and intraobserver variability. This study aimed to address the unmet need for an open-source computer-aided detection (CAD) system for 3D segmentation of lung and metastatic lung nodules along with radiomic feature extraction. The proposed adaptive region-growing-based lung nodule segmentation (RGLNS) tool was developed in-house, requiring only manual input to select a seed point within the nodule on computed tomography (CT) images. A total of 230 CT scans from 100 patients with sarcomas were screened. Lung nodules were present in 200 CT scans, which were further analyzed. The accuracy of the lung and nodule segmentation was evaluated qualitatively using a 5-point Likert scale (uninterpretable: 1; poor: 2; fair: 3; good: 4; excellent: 5) and quantitatively using the Dice coefficient and Jaccard index. A total of 200 CT scans comprising 12,000 CT slices were analyzed, among which 786 lung nodules were identified. Quantitative lung segmentation accuracies (n=2400 slices) yielded a Dice coefficient of 0.92±0.06 and a Jaccard index of 0.85±0.05. Qualitative scores (n=9600 slices) for lung boundary correction (4.56±1.18) and inclusion of pulmonary vessels (4.75±0.72) were rated as good to excellent. Quantitative nodule segmentation (n=142 nodules) accuracies were as follows: dice coefficient=0.92±0.03, 0.88±0.04, 0.86±0.03, 0.85±0.03, 084±0.04 and Jaccard index=0.84±0.03, 0.81±0.04, 0.78±0.04, 0.78±0.02, 0.76±0.04 for solitary (n=73), juxtapleural (n=32), juxtavascular (n=28), fissure-attached (n=6), and ground-glass (n=6) nodules, respectively. Qualitative scores (n=644 nodules) for nodule-boundary were good to excellent [solitary (n=342): 4.97±0.15; juxtapleural (n=155): 4.45±0.60; juxtavascular (n=127): 4.40±0.65; fissure-attached (n=9): 4.40±0.70; ground-glass (n=11): 4.25±0.75] and for exclusion of pulmonary vessels/pleura from nodules were good [juxtapleural (n=155): 4.10±0.66; juxtavascular (n=127): 4.08±0.64; fissure-attached (n=9): 4.30±0.67]. The proposed semiautomated CAD system, RGLNS, requiring minimal manual input, demonstrated robust, and promising segmentation results for the whole lung and various types of metastatic lung nodules.

Sentiment Analysis of Movie and News: A Review on Unbalanced Data Using Deep Learning with Abstract Feature Extraction

This study assessed a sentiment analysis model using a BERT-based deep learning approach, combined with a feedforward neural network (FFNN), designed to handle unbalanced datasets. The model was evaluated on the Rotten Tomatoes movie review dataset and the 20 Newsgroups dataset. Our method achieved an outstanding performance, with perfect accuracy of 100% and minimal loss (0.001 training and 0.010 testing loss) on the Rotten Tomatoes dataset and similar results on the 20 Newsgroups dataset. Key metrics, such as the Jaccard index, Hamming loss, Cohen’s kappa, mean squared error, and Matthews correlation coefficient (MCC), demonstrated its effectiveness, particularly in managing class imbalance and capturing nuanced sentiment variations. Comparative analysis showed the proposed model to outperform existing methods, highlighting its robustness and potential application in sentiment classification tasks across various textual data domains.

Analiza zgodności danych opisowych ewidencji gruntów i budynku z księgami wieczystymi na wybranym przykładzie

There are currently two basic public registers concerning real estate in Poland: the Land and Building Register (EGiB) and the Land and Mortgage Register (KW). The descriptive data in the Land and Building Register includes information on registered plots of land, buildings and premises, while the Land and Mortgage Register presents information on the legal status of the property. These systems operate independently, as they are maintained by different public administration bodies. However, the data contained in the Land and Building Register form the basis for the designation of the property in the Land and Mortgage Register, and thus both registers are interconnected and should contain consistent information. It is crucial that the information repetitively reported in both the EGiB and KW are identical. Consistency of data attribute values contained in both registers provides technical and organisational opportunities for harmonisation and integration of data from both sets. It is also an important factor for conducting investment activities on real estate. Previous analyses by researchers indicate that data on plots in the real estate registers are not sufficiently consistent. The aim of the article is to review the current situation with regard to the consistency of the basic data on land plots contained in the Land and Building Register and Land and Mortgage Registers on a selected example. The analysis by the comparative method was supplemented with the determination of the Jaccard index and the Russel-Rao coefficient. These are indices allowing to determine the degree of similarity between sets containing the same attributes.

Species composition of vascular plants of Western Karatau gorges (Mangyshlak)

Analysis of flora and identification of important plant areas in Kazakhstan is an important task for the conservation of biological diversity. On the territory of Mangistau region an important object for identification is the vegetation of the Western Karatau ridge, where a large number of species grow, including species with conservation status. Analysis of species composition of vascular plants of 8 gorges (Samal, Akmysh, Kogez, Botakan, Zhemsemsay, Karasay, Kendirli, Shybyksay) allowed to determine 60 species from 53 genera and 26 families in the summer flora. The dominant families are Asteraceae, Fabaceae, Brassicaceae, Poaceae, Polygonaceae, Chenopodiaceae, Caryophyllaceae, Lamiaceae, and Rosaceae. By life forms, perennials and small perennials dominate, by ecological groups — xerophytes and xeromesophytes. The similarity between the floras of the gorges was established on the basis of Jaccard's coefficient, which varied from 0.26 to 0.58. The minimum similarity was observed between Zhemsemsay and Shybyksay gorges, the maximum — between Samal and Kogez gorges. Growth of 7 species classified as rare and endangered plants of Mangistau region, as well as 3 species listed in the Red Book of Kazakhstan was noted. On the basis of the performed research recommendations were given for inclusion of the gorges of the Western Karatau Ridge in the list of important plant areas.

LD-SMOTE: A Novel Local Density Estimation-Based Oversampling Method for Imbalanced Datasets

Imbalanced data have become an essential stumbling block in the field of machine learning. In this paper, a novel oversampling method based on local density estimation, namely LD-SMOTE, is presented to address constraints of the popular rebalance technique SMOTE. LD-SMOTE initiates with k-means clustering to quantificationally measure the classification contribution of each feature. Subsequently, a novel distance metric grounded in Jaccard similarity is defined, which accentuates the features that are more intricately linked to the minority class. Utilizing this metric, we estimate the local density with a Gaussian-like function to control the quantity of synthetic samples around every minority sample, thus simulating the distribution of the minority class. Additionally, the generation of synthetic samples occurs within a triangular region constructed by this minority sample and its two chosen neighbors in LD-SMOTE, instead of on the line connecting the minority sample and one of its neighbors. Experimental comparisons between LD-SMOTE and 16 existing resampling methods on 19 datasets reveal a significant average increase in LD-SMOTE with 6.4% in accuracy, 4.4% in the F-measure, 5.4% in the G-mean, and 4.0% in AUC. This result indicates that LD-SMOTE can be an alternative oversampling method for imbalanced datasets.

Fingerprint resemblance amongst siblings and non-siblings - Forensic and anthropological implications.

Fingerprints hold a crucial position in forensic science, primarily for identification purposes, with thumbprints emerging as the most frequently recovered evidence at crime scenes. The main purpose of the present study was to conduct a comparative analysis of thumbprints of siblings and non-sibling pairs to investigate resemblance and similarities in fingerprint patterns. Thumbprints of 144 individuals (72 siblings and 72 non-siblings) from 36 families were obtained. Jaccard Similarity Index was employed to analyse the similarity between the fingerprint patterns of siblings and non-siblings. The strength of association between the variables is evaluated by chi-square and Cramer's V tests. The results indicated a significant similarity (69.44%) in thumbprint patterns among sibling pairs, demonstrating congruence in their right thumbprints, while non-sibling pairs exhibited a mere 30.56% similarity in the right thumbprints. Statistical analyses revealed a significant association (p = 0.0002) between sibling pairs, indicating a strong correlation in right-hand thumbprint patterns. The study bears substantial implications across various fields, including anthropology, population genetics, medical genetics and forensic science. Moreover, it may facilitate the examination of population similarities and variations and may aid in enhancing forensic investigations by identifying familial resemblance and similarity.

A Resource-Efficient Multi-Entropy Fusion Method and Its Application for EEG-Based Emotion Recognition.

Emotion recognition is an advanced technology for understanding human behavior and psychological states, with extensive applications for mental health monitoring, human-computer interaction, and affective computing. Based on electroencephalography (EEG), the biomedical signals naturally generated by the brain, this work proposes a resource-efficient multi-entropy fusion method for classifying emotional states. First, Discrete Wavelet Transform (DWT) is applied to extract five brain rhythms, i.e., delta, theta, alpha, beta, and gamma, from EEG signals, followed by the acquisition of multi-entropy features, including Spectral Entropy (PSDE), Singular Spectrum Entropy (SSE), Sample Entropy (SE), Fuzzy Entropy (FE), Approximation Entropy (AE), and Permutation Entropy (PE). Then, such entropies are fused into a matrix to represent complex and dynamic characteristics of EEG, denoted as the Brain Rhythm Entropy Matrix (BREM). Next, Dynamic Time Warping (DTW), Mutual Information (MI), the Spearman Correlation Coefficient (SCC), and the Jaccard Similarity Coefficient (JSC) are applied to measure the similarity between the unknown testing BREM data and positive/negative emotional samples for classification. Experiments were conducted using the DEAP dataset, aiming to find a suitable scheme regarding similarity measures, time windows, and input numbers of channel data. The results reveal that DTW yields the best performance in similarity measures with a 5 s window. In addition, the single-channel input mode outperforms the single-region mode. The proposed method achieves 84.62% and 82.48% accuracy in arousal and valence classification tasks, respectively, indicating its effectiveness in reducing data dimensionality and computational complexity while maintaining an accuracy of over 80%. Such performances are remarkable when considering limited data resources as a concern, which opens possibilities for an innovative entropy fusion method that can help to design portable EEG-based emotion-aware devices for daily usage.

DYNAMIC OBSERVATION OF UPPER AIRWAYS COLONIZATION IN A CYSTIC FIBROSIS PATIENT INFECTED WITH ROSEOMONAS AEROFRIGIDENS UNDER TARGETED THERAPY

Abstract Cystic fibrosis (CF) is a disease characterized by a varying microbiological diversity, including bacteria such as Pseudomonas aeruginosa, Burkholderia cepacia complex, Achromobacter xylosoxidans/ruhlandii. Additionally, there are microorganisms with unknown clinical significance, including Roseomonas aerofrigidens. The main goal of the study is to evaluate the prevalence of such microorganisms among CF patients undergoing regular examinations at the microbiological laboratory of the Samara State Medical University, Ministry of Health of the Russian Federation. An analysis of 12,094 clinical material cultures was carried out, with identification done by using MALDI-ToF mass spectrometry. A composite correlation index was calculated. The Jaccard similarity coefficient was used to assess the nature of symbiotic or antagonistic interactions. A total of 20 strains of Roseomonas spp. bacteria was isolated. Four strains were obtained from the mucous membrane of the posterior pharyngeal wall, while the remaining strains – from nasal lavage fluid. The patient with F508del/F508del genetics had been receiving dual targeted therapy since October 2022. Before treatment, no clinically significant Gram-negative species or Roseomonas aerofrigidens strains were noted in the patient’s anamnesis. However, during observation, Roseomonas aerofrigidens cultures were obtained from nasal lavage fluid at the third, fifth, sixth, eighth, ninth, eleventh, and eighteenth months of observation, with no concurrent bacterial strain growth in sputum samples. The study of the protein profiles demonstrated a high degree of kinship among the isolates, with a correlation index exceeding 0.8, confirming the hypothesis on long-term colonization of the paranasal sinuses by same Roseomonas aerofrigidens strain. While assessing the consistency criterion, it was found that among 30 microbial species isolated from nasal lavage fluid over 18-month observation, two microorganisms were classified as representatives of the constant microbiota, four (including Roseomonas aerofrigidens) as additional microbiota. The remaining species belong to random microbiota. Based on the Jaccard coefficient, it was found that four pairs of microorganisms are characterized by a synergistic nature of the relationship. These observations demonstrate a case of multi-month colonization with a unique strain of Roseomonas aerofrigidens in a patient with cystic fibrosis receiving targeted therapy.