Jaccard Index Research Articles

Tobacco crop rotation enhances the stability and complexity of microbial networks.

The effects of continuous cropping and rotation cropping, two important tobacco cultivation practices, on soil microbial communities at different stages remain unclear. Different planting patterns have been shown to influence soil physical and chemical properties, which in turn can affect the composition and diversity of soil microbial communities. In order to investigate the impact of different planting methods on soil microbial community structure, we selected two representative planting methods: continuous cropping (tobacco) and rotational cropping (tobacco-maize). These methods were chosen as the focal points of our research to explore the potential effects on soil microbial communities. High-throughput sequencing technology was employed to investigate the structure of soil microbial communities, as well as their relationships with soil environmental factors, by utilizing the 16S rRNA, ITS, and 18S genes. Furthermore, the interaction among microorganisms was explored through the application of the Random Matrix Theory (RMT) molecular ecological network approach. There was no significant difference in α diversity, but significant difference in β diversity based on Jaccard distance test. Compared to continuous cropping, crop rotation significantly increased the abundance of beneficial prokaryotes Verrucomicrobia and Rhodanobacter. These findings indicate that crop rotation promotes the enrichment of Verrucomicrobia and Rhodanobacter in the soil microbial community. AP and NH4-N had a greater effect on the community structure of prokaryotes and fungi in tobacco soil, while only AP had a greater effect on the community structure of protist. Molecular ecological network analysis showed that the network robustness and Cohesion of rotation were significantly higher than that of continuous cropping, indicating that the complexity and stability of molecular ecological networks were higher in the rotational, and the microbial communities cooperated more effectively, and the community structure was more stable. From this point of view, rotational cropping is more conducive to changing the composition of soil microbial community, enhancing the stability of microbial network structure, and enhancing the potential ecological functions in soil.

Composition and Diversity Indices of the Fish Assemblage in the Tidal Mudflats of the Shatt Al-Arab Estuary in the Northwest Arabian Gulf

Ten thousand four hundred and thirty-five fish were collected from January 2021 to February 2022, belonging to 71 species distributed across 39 families and 20 orders. The study included the tidal zone of the Shatt al-Arab estuary in the northwest of the Arabian Gulf. Two methods used to collection the fishes; fixed gill nets and trowel nets. Various diversity indices, relative abundance, and richness were assessed. Several biodiversity indices were used to assess the area biologically and ecologically which Species diversity for Shannon and Simpson, species richness for Margalef and Menhinick, dominance for Berger-Parker, species similarity for Jaccard, numerical similarity for Bray – Curtis, and evenness for Peilou index. Nematalosa nasus had the highest relative abundance, which was recorded at 24.53%. However, the lowest relative abundance was observed for 20 species, with a value of 0.01% with only one individual. The Carangiformes had the highest percentage of 18.31% and the lowest percentage of 1.41 % for many orders only, with one species recorded. The Margalef index had its highest value on June 5.24 and its lowest on December 1.24. The Menhinick index had its highest value of 1.53 in October and its lowest value of 0.71 in November. The Shannon Index reached a peak of 2.69 in January 2021, and fell to 0.98 in September. The Simpson index recorded the highest value of 0.92 in January 2021 and the lowest value of 0.37 in September. The Peilou index reached its highest value of 0.965 in December, while its lowest value was 0.294 in September. The Jaccard index had its highest value of 0.667 between January 2021 and March, while its lowest value was 0.079 between December, June, and July. The highest value for the Bray-Curtis index occurred between July and August, reaching 0.644. In contrast, the lowest value observed was 0.005 between July and December.

Western Atlantic regionalization based on distributional congruence patterns of Scleractinian corals

Coral reefs represent a source of ecosystem services relevant for the permanence of coastal human communities. Given the importance of zooxanthellate scleractinian corals (ZSC) in these ecosystems, information on the presence and distribution of this group is valuable for studying the biodiversity and establishing management strategies to ensure their permanence over time. Despite being a highly studied biological group in the Western Atlantic, records of ZSC presence are heterogeneous and are distributed mainly in the Caribbean Sea. One way to analyze the distribution of biota at large spatial scales is through biogeographic regionalizations. To identify the distribution patterns of ZSC in the Western Atlantic and propose a biogeographic regionalization based in this biological group, we performed a Parsimony Analysis of Endemism (PAE) at ecoregion scale using Species Distribution Models (SDM) to fill the gaps in data on the distribution of these corals. Through these analyses, we identified the presence of two main biogeographic provinces: Northwestern Atlantic Tropical Province and Tropical Southwestern Atlantic. According to Jaccard Index, this regionalization is highly similar to those performed with multitaxa by other authors, which demonstrates that for these purposes this taxa can be considered as a surrogate group. In terms of management, this study highlights the importance of establishing preservation and conservation measures that address this biological group, in order to guarantee both the permanence of other taxa and the ecosystem goods and services they provide.

A comparative study of Mongolian medicine prescriptions in the Chinese Pharmacopoeia and Tibetan medicinal compounds in the Ministerial Standards

Objective: To analyze the similarity between Mongolian medicine prescriptions included in the 2020 edition of the Pharmacopoeia of the People’s Republic of China (hereinafter referred to as the Chinese Pharmacopoeia) and Tibetan medicinal compounds according to the Drug Standards of the Ministry of Public Health of the People’s Republic of China: Tibetan Medicine (Volume I) (hereinafter referred to as the Ministerial Standards). Methods: A comparative analysis was conducted between Mongolian medicine prescriptions contained in the Chinese Pharmacopoeia and the corresponding Tibetan medicinal compounds included in the Ministerial Standards using Jaccard similarity coefficients and Hellinger distance algorithms across three dimensions: composition, shared drug dosage ratio, and efficacy. Different weights were applied to calculate the comprehensive similarity between the formulas. Simultaneously, we collated and verified the original source literature, and determined the era of composition for each formula. Results: The overall similarity between the Mongolian medicine prescriptions from the Chinese Pharmacopoeia and the Tibetan medicinal compounds from the Ministerial Standards was over 50%. Notably, the composite similarity for the pairings Sanwei Jili San and Sanwei Jili Tangsan, as well as Wuwei Qingzhuo San and Shiliu Anzhi Jinghua San, exceeded 90%. Additionally, six sets of formulas achieved a similarity of over 75%. Literature research has revealed that all 13 sets of formulas first appeared in Tibetan medical literature and were long-standing practices in Tibetan medicine clinics, significantly predating their initial documentation in Mongolian medical literature. Conclusion: The 13 sets of Mongolian medicine prescriptions originated from Tibetan medicinal compounds. Throughout their evolution, the sets adhered to the principles of similar origins and/or properties and effects, leading to the emergence of both commonly used and alternative items. Nevertheless, the efficacy of the formulas has remained highly consistent. The phenomenon of ingredient substitution in the formulas could be attributed to various factors, such as resource availability and cognitive differences among inheritance schools. This study has important implications for further comparative research on content quantification and pharmacological activity. Such research could uncover new medicinal varieties, fostering the sustainable development of Tibetan and Mongolian medicine.

CACSNet for automatic robust classification and segmentation of carotid artery calcification on panoramic radiographs using a cascaded deep learning network

Stroke is one of the major causes of death worldwide, and is closely associated with atherosclerosis of the carotid artery. Panoramic radiographs (PRs) are routinely used in dental practice, and can be used to visualize carotid artery calcification (CAC). The purpose of this study was to automatically and robustly classify and segment CACs with large variations in size, shape, and location, and those overlapping with anatomical structures based on deep learning analysis of PRs. We developed a cascaded deep learning network (CACSNet) consisting of classification and segmentation networks for CACs on PRs. This network was trained on ground truth data accurately determined with reference to CT images using the Tversky loss function with optimized weights by balancing between precision and recall. CACSNet with EfficientNet-B4 achieved an AUC of 0.996, accuracy of 0.985, sensitivity of 0.980, and specificity of 0.988 in classification for normal or abnormal PRs. Segmentation performances for CAC lesions were 0.595 for the Jaccard index, 0.722 for the Dice similarity coefficient, 0.749 for precision, and 0.756 for recall. Our network demonstrated superior classification performance to previous methods based on PRs, and had comparable segmentation performance to studies based on other imaging modalities. Therefore, CACSNet can be used for robust classification and segmentation of CAC lesions that are morphologically variable and overlap with surrounding structures over the entire posterior inferior region of the mandibular angle on PRs.

Artificial Versus Human Intelligence in the Diagnostic Approach of Ophthalmic Case Scenarios: A Qualitative Evaluation of Performance and Consistency.

To evaluate the efficiency of three artificial intelligence (AI) chatbots (ChatGPT-3.5 (OpenAI, San Francisco, California, United States), Bing Copilot (Microsoft Corporation, Redmond, Washington, United States), Google Gemini (Google LLC, Mountain View, California, United States)) in assisting the ophthalmologist in the diagnostic approach and management of challenging ophthalmic cases and compare their performance with that of a practicing human ophthalmic specialist. The secondary aim was to assess the short- and medium-term consistency of ChatGPT's responses. Eleven ophthalmic case scenarios of variable complexity were presented to the AI chatbots and to an ophthalmic specialist in a stepwise fashion. Advice regarding the initial differential diagnosis, the final diagnosis, further investigation, and management was asked for. One month later, the same process was repeated twice on the same day for ChatGPT only. The individual diagnostic performance of all three AI chatbots was inferior to that of the ophthalmic specialist; however, they provided useful complementary input in the diagnostic algorithm. This was especially true for ChatGPT and Bing Copilot. ChatGPT exhibited reasonable short- and medium-term consistency, with the mean Jaccard similarity coefficient of responses varying between 0.58 and 0.76. AI chatbots may act as useful assisting tools in the diagnosis and management of challenging ophthalmic cases; however, their responses should be scrutinized for potential inaccuracies, and by no means can they replace consultation with an ophthalmic specialist.

Cross-Social-Network User Identification Based on Bidirectional GCN and MNF-UI Models

Due to the distinct functionalities of various social network platforms, users often register accounts on different platforms, posing significant challenges for unified user management. However, current multi-social-network user identification algorithms heavily rely on user attributes and cannot perform user identification across multiple social networks. To address these issues, this paper proposes two identity recognition models. The first model is a cross-social-network user identification model based on bidirectional GCN. It calculates user intimacy using the Jaccard similarity coefficient and constructs an adjacency matrix to accurately represent user relationships in the social network. It then extracts cross-social-network user information to accomplish user identification tasks. The second model is the multi-network feature user identification (MNF-UI) model, which introduces the concept of network feature vectors. It effectively maps the structural features of different social networks and performs user identification based on the common features of seed nodes in the cross-network environment. Experimental results demonstrate that the bidirectional GCN model significantly outperforms baseline algorithms in cross-social-network user identification tasks. The MNF-UI (multi-network feature user identification) model can operate in situations with two or more networks with inconsistent structures, resulting in improved identification accuracy. These two user identification algorithms provide technical and theoretical support for in-depth research on social network information integration and network security maintenance.

Ant Communities (Hymenoptera: Formicidae) in a Protected Urban Park in Northeast Mexico

Introduction: The creation of protected natural areas within urban environments has played an essential role in providing small reservoirs of native vegetation and refuges for endemic species. Objective: To analyze, through ant communities, the influence of urbanization in a protected natural area (PNA) located in an urban zone. Methods: From April to July 2022, using bait traps and pitfall traps along 100 m transects, the presence of ant species was recorded in the El Refugio State Park Natural Area in Tamaulipas, Mexico. Alpha diversity was assessed using Hill numbers, while beta diversity was evaluated using the Jaccard index. Additionally, indicator species were identified through the indicator value (IndVal). Multiple correspondence analysis was conducted to associate the presence of species with environmental variables and vegetation types, and contingency tables were created to relate ant presence to vegetation cover size. Results: Species richness and diversity decreased in sites with low vegetation cover, such as in Tamaulipan thornscrub; moreover, secondary vegetation retained greater species richness. Indicator species were found in secondary vegetation and grassland areas, mainly composed of species that tolerate high temperatures and generalists such as Forelius mccooki and Tetramorium spinosum, while lowland jungle species were the only indicator species. The presence of these species is primarily associated with temperature, humidity, and vegetation cover. A significant relationship between the presence of ant species and vegetation cover size in El Refugio State Park was evident. Conclusion: The structure and composition of ant communities vary according to the type of vegetation structure, particularly in areas that have undergone greater anthropogenic damage. Therefore, the conservation of landscapes, as well as monitoring of these areas, should be considered in future research. Keywords: green areas; formicids; conservation; indicator species; community ecology.

Multi-label Adversarial Attack with New Measures and Self-paced Constraint Weighting.

An adversarial attack is typically implemented by solving a constrained optimization problem. In top-k adversarial attacks implementation for multi-label learning, the attack failure degree (AFD) and attack cost (AC) of a possible attack are major concerns. According to our experimental and theoretical analysis, existing methods are negatively impacted by the coarse measures for AFD/AC and the indiscriminate treatment for all constraints, particularly when there is no ideal solution. Hence, this study first develops a refined measure based on the Jaccard index appropriate for AFD and AC, distinguishing the failure degrees/costs of two possible attacks better than the existing indicator function-based scheme. Furthermore, we formulate novel optimization problems with the least constraint violation via new measures for AFD and AC, and theoretically demonstrate the effectiveness of weighting slack variables for constraints. Finally, a self-paced weighting strategy is proposed to assign different priorities to various constraints during optimization, resulting in larger attack gains compared to previous indiscriminate schemes. Meanwhile, our method avoids fluctuations during optimization, especially in the presence of highly conflicting constraints. Extensive experiments on four benchmark datasets validate the effectiveness of our method across different evaluation metrics.

Metagenomic Characterization and Comparative Analysis of Removable Denture-Wearing and Non-Denture-Wearing Individuals in Healthy and Diseased Periodontal Conditions.

Removable denture wearers are at an increased risk of developing periodontal diseases due to biofilm deposition and microbial colonization on the denture surface. This study aimed to characterize and compare the metagenomic composition of saliva in denture wearers with different periodontal statuses. Twenty-four community-dwelling elders were recruited and grouped into denture wearers with active periodontitis (APD), non-denture wearers with active periodontitis (APXD), denture wearers with stable periodontal health conditions (SPCD), and non-denture wearers with stable periodontal health conditions (SPCXD). Saliva samples were collected and underwent Type IIB restriction-site-associated DNA for microbiome (2bRAD-M) metagenomic sequencing to characterize the species-resolved microbial composition. Alpha diversity analysis based on the Shannon index revealed no significant difference between groups. Beta diversity analysis using the Jaccard distance matrix was nearly significantly different between denture-wearing and non-denture-wearing groups (p = 0.075). Some respiratory pathogens, including Streptococcus agalactiae and Streptococcus pneumoniae, were detected as the top 30 species in saliva samples. Additionally, LEfSe analysis revealed a substantial presence of pathogenic bacteria in denture groups. In the cohort of saliva samples collected from community-dwelling elders, a remarkable abundance of certain opportunistic pathogens was detected in the microbial community.

A Comparative Study of Segmentation Algorithms for Intracerebral Hemorrhage Detection

Segmentation in the medical field has special importance. One of the purposes of segmentation is to visualize the area affected by the disease after disease detection in any organ. In recent years, efficient studies have been carried out for this purpose with deep learning models. In this study, three segmentation algorithms were compared for the detection of hemorrhage in brain parenchyma. These algorithms are the most familiar: U-net, LinkNet, and FPN algorithms. For the background of these algorithms, five backbones consisting of deep learning models were used. These backbones are Resnet34, ResNet50, ResNet169, EfficientNetB0, and EfficientNet B1. An original dataset was created for the study. The dataset in the study was verified by experts. In the study, the Dice coefficient and Jaccard index, which are the most common metrics in the medical field, were chosen as evaluation metrics. Considering the performance results of the algorithms, the FPN architecture with a 0.9495 Dice coefficient value for the training data and LinkNet with a 0.9244 Dice coefficient for the test data gave the best results. In addition, EfficientNetB1 provided the best results among the backbones used. When the results obtained were examined, better segmentation performance was obtained than in existing studies.

Optic Disc Segmentation in Human Retina Images Using a Meta Heuristic Optimization Method and Disease Diagnosis with Deep Learning

Glaucoma is a common eye disease that damages the optic nerve and leads to loss of vision. The disease shows few symptoms in the early stages, making its identification a complex task. To overcome the challenges associated with this task, this study aimed to tackle the localization and segmentation of the optic disc, as well as the classification of glaucoma. For the optic disc segmentation, we propose a novel metaheuristic approach called Grey Wolf Optimization (GWO). Two different approaches are used for glaucoma classification: a one-stage approach, in which the whole image without cropping is used for classification, and a two-stage approach. In the two-stage approach, the optic disc region is detected using the You Only Look Once (YOLO) detection algorithm. Once the optic disc region of interest (ROI) is identified, glaucoma classification is performed using pre-trained convolutional neural networks (CNNs) and vision transformation techniques. In addition, both the one-stage and the two-stage approaches are applied in combination with the pre-trained CNN using the Random Forest algorithm. In segmentation, GWO achieved an average sensitivity of 96.04%, a specificity of 99.58%, an accuracy of 99.39%, a DICE coefficient of 94.15%, and a Jaccard index of 90.4% on the Drishti-GS dataset. For classification, the proposed method achieved remarkable results with a high-test accuracy of 100% and 88.18% for hold-out validation and three-fold cross-validation for the Drishti-GS dataset, and 96.15% and 93.84% for ORIGA with hold-out and five-fold cross-validation, respectively. Comparing the results with previous studies, the proposed CNN model outperforms them. In addition, the use of the Swin transformer shows its effectiveness in classifying glaucoma in different subsets of the data.

Technical-scientific production and knowledge networks about medicinal plants and herbal medicines in the Amazon.

This paper explores the role of Brazilian research institutions in the global and national context of study of medicinal plants. Most of these plants have ethnopharmacological use and herbal medicines related to the Amazon. It highlights Brazil's position in scientific production and the importance of Amazonian resources in developing phytomedicines. The study aims to provide an overview of the technical-scientific production of medicinal plants and herbal medicines related to the Amazon, focusing on scientific impact, collaboration, Technology Readiness Level (TRL) of scientific production, and innovation system maturity. The study employs a comprehensive methodological approach, including data collection from Scopus covering the period from 2002 to 2022. The data was cleaned and analyzed using bibliometric and network analysis techniques. Advanced natural language processing techniques, such as Latent Dirichlet Allocation and Jaccard distance measure, were used for TRL classification. The findings reveal a predominant contribution from Brazilian institutions and authors, with 1,850 publications analyzed. Key areas identified include Pharmacology, Toxicology, Pharmaceuticals, Medicine, and Biochemistry. The study also uncovers various collaborative networks and technological maturity levels, with a significant focus on early-stage development phases. The research concludes that Brazilian institutions, particularly those in the Amazon region, play a significant role in the scientific exploration and development of medicinal plants and herbal medicines. Despite this, countries like the USA were proportionally more productive in clinical trial research. The study underscores the potential of Brazil's rich biodiversity and traditional knowledge in the pharmaceutical industry, particularly for neglected diseases. It suggests the need for stronger research systems and international collaboration to leverage these resources for global health benefits.

Altered glia-neuron communication in Alzheimer’s Disease affects WNT, p53, and NFkB Signaling determined by snRNA-seq

BackgroundAlzheimer’s disease is the most common cause of dementia and is characterized by amyloid-β plaques, tau neurofibrillary tangles, and neuronal loss. Although neuronal loss is a primary hallmark of Alzheimer’s disease, it is known that non-neuronal cell populations are ultimately responsible for maintaining brain homeostasis and neuronal health through neuron-glia and glial cell crosstalk. Many signaling pathways have been proposed to be dysregulated in Alzheimer’s disease, including WNT, TGFβ, p53, mTOR, NFkB, and Pi3k/Akt signaling. Here, we predict altered cell-cell communication between glia and neurons.MethodsUsing public snRNA-sequencing data generated from postmortem human prefrontal cortex, we predicted altered cell-cell communication between glia (astrocytes, microglia, oligodendrocytes, and oligodendrocyte progenitor cells) and neurons (excitatory and inhibitory). We confirmed interactions in a second and third independent orthogonal dataset. We determined cell-type-specificity using Jaccard Similarity Index and investigated the downstream effects of altered interactions in inhibitory neurons through gene expression and transcription factor activity analyses of signaling mediators. Finally, we determined changes in pathway activity in inhibitory neurons.ResultsCell-cell communication between glia and neurons is altered in Alzheimer’s disease in a cell-type-specific manner. As expected, ligands are more cell-type-specific than receptors and targets. We identified ligand-receptor pairs in three independent datasets and found involvement of the Alzheimer’s disease risk genes APP and APOE across datasets. Most of the signaling mediators of these interactions were not significantly differentially expressed, however, the mediators that are also transcription factors had differential activity between AD and control. Namely, MYC and TP53, which are associated with WNT and p53 signaling, respectively, had decreased TF activity in Alzheimer’s disease, along with decreased WNT and p53 pathway activity in inhibitory neurons. Additionally, inhibitory neurons had both increased NFkB signaling pathway activity and increased TF activity of NFIL3, an NFkB signaling-associated transcription factor.ConclusionsCell-cell communication between glia and neurons in Alzheimer’s disease is altered in a cell-type-specific manner involving Alzheimer’s disease risk genes. Signaling mediators had altered transcription factor activity suggesting altered glia-neuron interactions may dysregulate signaling pathways including WNT, p53, and NFkB in inhibitory neurons.

Artificial Intelligence in Early Detection of Cervical Intraepithelial Neoplasia

Artificial Intelligence (AI) is a quickly evolving field of technology used to develop intelligent machines capable of performing tasks such as problem solving, decision making , perception, language processing, and learning. This paper explores the application of AI in the field of gynecological oncology, specifically in the diagnosis of cervical cancer. The paper proposes a hybrid AI model that uses a Gaussian mixture model and a deep learning model to segment and classifies colposcope images. The model performed with satisfactory segmentation metrics of sensitivity, specificity, dice index, and Jaccard index of 0.976, 0.989, 0.954, and 0.856, respectively. This model aims to accurately classify cancer and non-cancer cases from a colposcope image. The results showed that this method could effectively segment the colposcopy images and extract the cervix region. This can be a valuable tool for automated cancer diagnosis and can help improve the diagnosis's accuracy.

Deep Learning-Based Barley Disease Quantification for Sustainable Crop Production.

Net blotch disease caused by Drechslera teres is a major fungal disease that affects barley (Hordeum vulgare) plants and can result in significant crop losses. In this study, we developed a deep learning model to quantify net blotch disease symptoms on different days postinfection on seedling leaves using Cascade R-CNN (region-based convolutional neural network) and U-Net (a convolutional neural network) architectures. We used a dataset of barley leaf images with annotations of net blotch disease to train and evaluate the model. The model achieved an accuracy of 95% for Cascade R-CNN in net blotch disease detection and a Jaccard index score of 0.99, indicating high accuracy in disease quantification and location. The combination of Cascade R-CNN and U-Net architectures improved the detection of small and irregularly shaped lesions in the images at 4 days postinfection, leading to better disease quantification. To validate the model developed, we compared the results obtained by automated measurement with a classical method (necrosis diameter measurement) and a pathogen detection by real-time PCR. The proposed deep learning model could be used in automated systems for disease quantification and to screen the efficacy of potential biocontrol agents to protect against disease.

A novel CNN-based image segmentation pipeline for individualized feline spinal cord stimulation modeling

Objective. Spinal cord stimulation (SCS) is a well-established treatment for managing certain chronic pain conditions. More recently, it has also garnered attention as a means of modulating neural activity to restore lost autonomic or sensory-motor function. Personalized modeling and treatment planning are critical aspects of safe and effective SCS (Rowald and Amft 2022 Front. Neurorobotics 16 983072, Wagner et al 2018 Nature 563 65–71). However, the generation of spine models at the required level of detail and accuracy requires time and labor intensive manual image segmentation by human experts. This study aims to develop a maximally automated segmentation routine capable of producing high-quality anatomical models, even with limited data, to facilitate safe and effective personalized SCS treatment planning. Approach. We developed an automated image segmentation and model generation pipeline based on a novel convolutional neural network (CNN) architecture trained on feline spinal cord magnetic resonance imaging data. The pipeline includes steps for image preprocessing, data augmentation, transfer learning, and cleanup. To assess the relative importance of each step in the pipeline and our choice of CNN architecture, we systematically dropped steps or substituted architectures, quantifying the downstream effects in terms of tissue segmentation quality (Jaccard index and Hausdorff distance) and predicted nerve recruitment (estimated axonal depolarization). Main results. The leave-one-out analysis demonstrated that each pipeline step contributed a small but measurable increment to mean segmentation quality. Surprisingly, minor differences in segmentation accuracy translated to significant deviations (ranging between 4% and 13% for each pipeline step) in predicted nerve recruitment, highlighting the importance of careful workflow design. Additionally, transfer learning techniques enhanced segmentation metric consistency and allowed generalization to a completely different spine region with minimal additional training data. Significance. To our knowledge, this work is the first to assess the downstream impacts of segmentation quality differences on neurostimulation predictions. It highlights the role of each step in the pipeline and paves the way towards fully automated, personalized SCS treatment planning in clinical settings.

Transcriptomic analysis of cAMP/PKA/CREB signaling in invasive lobular breast cancer.

1109 Background: Invasive lobular breast cancer (ILC) is the most common special type of breast cancer and has unique clinicopathological and molecular hallmarks that differentiate it from the more common invasive carcinoma – no special type (NST). Despite these differences, ILC and NST are treated as a single entity and there is a lack of ILC-targeted therapies. To fill this gap, we sought to identify novel molecular alterations in ILC that could be exploited for targeted therapies. Methods: Differential gene expression and Geneset Enrichment and Variation analyses were performed on RNA-seq data from three large public breast cancer databases – the Sweden Cancerome Analysis Network-Breast (SCAN-B; luminal A ILC N=263, luminal A NST N=1162), The Cancer Genome Atlas (TCGA; luminal A ILC N=157, luminal A NST N=307) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC; luminal A ILC N=65, luminal A NST N=533). Pathways enriched in overlapping differentially expressed genes from these datasets were clustered using Jaccard similarity to identify pathways enriched in ILC. The cAMP/PKA/CREB signaling was studied in ILC, ILC-like and NST cell lines and patient-derived organoids (PDOs) using forskolin, an activator of the pathway. Results: Clinicopathological features of patients with ILC and NST in SCAN-B were similar to prior population-based studies. There was a consistent pattern of up-regulation of cAMP/PKA/CREB related signaling in ILC compared to NST in SCAN-B, TCGA and METABRIC. Treatment with forskolin resulted in a greater increase in phospho-CREB in ILC cell lines and organoids than NST. CRISPR deletion of CDH1 in NST cell lines did not alter response of cells to forskolin as measured by phospho-CREB. Forskolin treatment caused growth inhibition in ILC and NST, with ILC cell lines being more sensitive to forskolin-mediated growth inhibition. Conclusions: In three separate datasets, cAMP/PKA/CREB signaling was identified to be higher in ILC than NST. This in silico finding was validated in cell line and organoid models. Loss of CDH1 was not sufficient to mediate this phenotype. Future studies should investigate the mechanisms for differential cAMP/PKA/CREB signaling and the potential for therapeutic targeting in patients with ILC.

Prospective evaluation of the breast microbiome and tumor microenvironment-related biomarkers of response to neoadjuvant systemic therapy in triple negative breast cancer.

TPS615 Background: Predictive biomarkers of response to combination chemotherapy and immune-checkpoint inhibitors are urgently needed to tailor treatment recommendations for patients with early-stage triple negative breast cancer (eTNBC). Our group has demonstrated that tumour-associated microbiota in primary breast tumours represent promising and novel candidate biomarkers for patients with breast cancer. We aim to prospectively interrogate the breast cancer microbiome and tumour microenvironment (TME) of eTNBC treated with neoadjuvant chemo-immunotherapy, and correlate with clinical outcome. Methods: Trial design: This prospective translational biomarker study is enrolling patients with eTNBC suitable for standard neoadjuvant systemic therapy followed by definitive surgery. The primary objective is to evaluate the change in breast cancer microbiome composition pre- and post-therapy. Secondary objectives include correlation of the breast microbiome with pathologic complete response (pCR) and survival outcomes, tumor infiltrating lymphocytes (TILs), PDL-1 and immune subset analysis, extracellular vesicles analysis and analysis of gut microbiome. Tumor tissue samples will be collected at baseline (mandatory research biopsy) and at the time of surgery. Microbiome evaluation will be conducted using metagenomic sequencing to assess changes in the relative abundance of microbial taxa. Blood and stool samples will be collected and analysed at baseline, on-treatment, at the time of surgery and post-operatively for secondary endpoints. Statistical analysis: Hierarchical clustering of samples based on relative abundance of the operational taxonomic units (OTUs) with sample composition at family and genus level will be performed. Alpha Diversity (Shannon Diversity Index) and Beta Diversity (Jaccard Similarity Index) will be analysed. Principal component analysis (PCA) and Random Forest analysis will be performed for classification and clustering analysis. Comparison of taxa or functions between clinical cohorts will be performed (two tailed Z test) and corrected using the false discovery rate to determine Q-values. Thirty patients will be recruited over 18-24 months. Current status: This University College Cork sponsored trial received ethics approval from the CREC in January 2024 and recruitment is ongoing. Those interested can contact uccctg@ucc.ie. This trial will provide a deeper insight into the potential role of the local breast microbiome as a predictive biomarker for response to neoadjuvant therapy in patients with eTNBC. The results will guide further studies exploring optimal immunomodulatory combinations for the treatment of TNBC and may provide evidence regarding future therapeutic targeting of the breast cancer microbiome.

An analysis of ChatGPT recommendations for the diagnosis and treatment of cervical radiculopathy

OBJECTIVE The objective of this study was to assess the safety and accuracy of ChatGPT recommendations in comparison to the evidence-based guidelines from the North American Spine Society (NASS) for the diagnosis and treatment of cervical radiculopathy. METHODS ChatGPT was prompted with questions from the 2011 NASS clinical guidelines for cervical radiculopathy and evaluated for concordance. Selected key phrases within the NASS guidelines were identified. Completeness was measured as the number of overlapping key phrases between ChatGPT responses and NASS guidelines divided by the total number of key phrases. A senior spine surgeon evaluated the ChatGPT responses for safety and accuracy. ChatGPT responses were further evaluated on their readability, similarity, and consistency. Flesch Reading Ease scores and Flesch-Kincaid reading levels were measured to assess readability. The Jaccard Similarity Index was used to assess agreement between ChatGPT responses and NASS clinical guidelines. RESULTS A total of 100 key phrases were identified across 14 NASS clinical guidelines. The mean completeness of ChatGPT-4 was 46%. ChatGPT-3.5 yielded a completeness of 34%. ChatGPT-4 outperformed ChatGPT-3.5 by a margin of 12%. ChatGPT-4.0 outputs had a mean Flesch reading score of 15.24, which is very difficult to read, requiring a college graduate education to understand. ChatGPT-3.5 outputs had a lower mean Flesch reading score of 8.73, indicating that they are even more difficult to read and require a professional education level to do so. However, both versions of ChatGPT were more accessible than NASS guidelines, which had a mean Flesch reading score of 4.58. Furthermore, with NASS guidelines as a reference, ChatGPT-3.5 registered a mean ± SD Jaccard Similarity Index score of 0.20 ± 0.078 while ChatGPT-4 had a mean of 0.18 ± 0.068. Based on physician evaluation, outputs from ChatGPT-3.5 and ChatGPT-4.0 were safe 100% of the time. Thirteen of 14 (92.8%) ChatGPT-3.5 responses and 14 of 14 (100%) ChatGPT-4.0 responses were in agreement with current best clinical practices for cervical radiculopathy according to a senior spine surgeon. CONCLUSIONS ChatGPT models were able to provide safe and accurate but incomplete responses to NASS clinical guideline questions about cervical radiculopathy. Although the authors’ results suggest that improvements are required before ChatGPT can be reliably deployed in a clinical setting, future versions of the LLM hold promise as an updated reference for guidelines on cervical radiculopathy. Future versions must prioritize accessibility and comprehensibility for a diverse audience.