Annotation Model Research Articles

Performance of a Protein Language Model for Variant Annotation in Cardiac Disease.

Genetic testing is a cornerstone in the assessment of many cardiac diseases. However, variants are frequently classified as variants of unknown significance, limiting the utility of testing. Recently, the DeepMind group (Google) developed AlphaMissense, a unique artificial intelligence-based model, based on language model principles, for the prediction of missense variant pathogenicity. We aimed to report on the performance of AlphaMissense, accessed by VarCardio, an open web-based variant annotation engine, in a real-world cardiovascular genetics center. All genetic variants from an inherited arrhythmia program were examined using AlphaMissense via VarCard.io and compared with the ClinVar variant classification system, as well as another variant classification platform (Franklin by Genoox). The mutation reclassification rate and genotype-phenotype concordance were examined for all variants in the study. We included 266 patients with heritable cardiac diseases, harboring 339 missense variants. Of those, 230 (67.8%) were classified by ClinVar as either variants of unknown significance or nonclassified. Using VarCard.io, 198 variants of unknown significance (86.1%, 95% CI, 80.9-90.3) were reclassified to either likely pathogenic or likely benign. The reclassification rate was significantly higher for VarCard.io than for Franklin (86.1% versus 34.8%, P<0.001). Genotype-phenotype concordance was highly aligned using VarCard.io predictions, at 95.9% (95% CI, 92.8-97.9) concordance rate. For 109 variants classified as pathogenic, likely pathogenic, benign, or likely benign by ClinVar, concordance with VarCard.io was high (90.5%). AlphaMissense, accessed via VarCard.io, may be a highly efficient tool for cardiac genetic variant interpretation. The engine's notable performance in assessing variants that are classified as variants of unknown significance in ClinVar demonstrates its potential to enhance cardiac genetic testing.

Knowledge graph construction in hyperbolic space for automatic image annotation

Automatic image annotation (AIA) is a fundamental and challenging task in computer vision. Considering the correlations between tags can lead to more accurate image understanding, benefiting various applications, including image retrieval and visual search. While many attempts have been made to incorporate tag correlations in annotation models, the method of constructing a knowledge graph based on external knowledge sources and hyperbolic space has not been explored. In this paper, we create an attributed knowledge graph based on vocabulary, integrate external knowledge sources such as WordNet, and utilize hyperbolic word embeddings for the tag representations. These embeddings provide a sophisticated tag representation that captures hierarchical and complex correlations more effectively, enhancing the image annotation results. In addition, leveraging external knowledge sources enhances contextuality and significantly enriches existing AIA datasets. We exploit two deep learning-based models, the Relational Graph Convolutional Network (R-GCN) and the Vision Transformer (ViT), to extract the input features. We apply two R-GCN operations to obtain word descriptors and fuse them with the extracted visual features. We evaluate the proposed approach using three public benchmark datasets. Our experimental results demonstrate that the proposed architecture achieves state-of-the-art performance across most metrics on Corel5k, ESP Game, and IAPRTC-12.

BUILDING DOCUMENT READING ASSISTANT FOR THE VISUALLY IMPAIRED

This study introduces a solution that applies document analysis and recognition technologies to enhance document accessibility for individuals with visual impairments. The objective is to develop an algorithm capable of accurately analyzing the content of document components and converting them into voice format. Leveraging the pre-trained YOLOv8 model for document analysis and optical character recognition technology, the image annotation model uses the AIAnytime API and Pix2Tex technology to extract LaTeX code from images, facilitating the conversion of mathematical formulas into spoken words. The research results demonstrate significant progress in effectively supporting document reading, making a meaningful contribution to the field of assistive technology for the visually impaired.

Digital mapping of resected cancer specimens: The visual pathology report

BackgroundThe current standard-of-care pathology report relies only on lengthy written text descriptions without a visual representation of the resected cancer specimen. This study demonstrates the feasibility of incorporating virtual, three-dimensional (3D) visual pathology reports to improve communication of final pathology reporting. Materials and methodsSurgical specimens are 3D scanned and virtually mapped alongside the pathology team to replicate grossing. The 3D specimen maps are incorporated into a hybrid visual pathology report which displays the resected specimen and sampled margins alongside gross measurements, tumor characteristics, and microscopic diagnoses. ResultsVisual pathology reports were created for 10 head and neck cancer cases. Each report concisely communicated information from the final pathology report in a single page and contained significantly fewer words (293.4 words) than standard written pathology reports (850.1 words, p < 0.01). ConclusionsWe establish the feasibility of a novel visual pathology report that includes an annotated visual model of the resected cancer specimen in place of lengthy written text of standard of care head and neck cancer pathology reports.

ScGAA: a general gated axial-attention model for accurate cell-type annotation of single-cell RNA-seq data

Single-cell RNA sequencing (scRNA-seq) is a key technology for investigating cell development and analysing cell diversity across various diseases. However, the high dimensionality and extreme sparsity of scRNA-seq data pose great challenges for accurate cell type annotation. To address this, we developed a new cell-type annotation model called scGAA (general gated axial-attention model for accurate cell-type annotation of scRNA-seq). Based on the transformer framework, the model decomposes the traditional self-attention mechanism into horizontal and vertical attention, considerably improving computational efficiency. This axial attention mechanism can process high-dimensional data more efficiently while maintaining reasonable model complexity. Additionally, the gated unit was integrated into the model to enhance the capture of relationships between genes, which is crucial for achieving an accurate cell type annotation. The results revealed that our improved transformer model is a promising tool for practical applications. This theoretical innovation increased the model performance and provided new insights into analytical tools for scRNA-seq data.

Crop Growth Analysis Using Automatic Annotations and Transfer Learning in Multi-Date Aerial Images and Ortho-Mosaics

Growth monitoring of crops is a crucial aspect of precision agriculture, essential for optimal yield prediction and resource allocation. Traditional crop growth monitoring methods are labor-intensive and prone to errors. This study introduces an automated segmentation pipeline utilizing multi-date aerial images and ortho-mosaics to monitor the growth of cauliflower crops (Brassica Oleracea var. Botrytis) using an object-based image analysis approach. The methodology employs YOLOv8, a Grounding Detection Transformer with Improved Denoising Anchor Boxes (DINO), and the Segment Anything Model (SAM) for automatic annotation and segmentation. The YOLOv8 model was trained using aerial image datasets, which then facilitated the training of the Grounded Segment Anything Model framework. This approach generated automatic annotations and segmentation masks, classifying crop rows for temporal monitoring and growth estimation. The study’s findings utilized a multi-modal monitoring approach to highlight the efficiency of this automated system in providing accurate crop growth analysis, promoting informed decision-making in crop management and sustainable agricultural practices. The results indicate consistent and comparable growth patterns between aerial images and ortho-mosaics, with significant periods of rapid expansion and minor fluctuations over time. The results also indicated a correlation between the time and method of observation which paves a future possibility of integration of such techniques aimed at increasing the accuracy in crop growth monitoring based on automatically derived temporal crop row segmentation masks.

CACER: Clinical concept Annotations for Cancer Events and Relations.

Clinical notes contain unstructured representations of patient histories, including the relationships between medical problems and prescription drugs. To investigate the relationship between cancer drugs and their associated symptom burden, we extract structured, semantic representations of medical problem and drug information from the clinical narratives of oncology notes. We present Clinical concept Annotations for Cancer Events and Relations (CACER), a novel corpus with fine-grained annotations for over 48000 medical problems and drug events and 10000 drug-problem and problem-problem relations. Leveraging CACER, we develop and evaluate transformer-based information extraction models such as Bidirectional Encoder Representations from Transformers (BERT), Fine-tuned Language Net Text-To-Text Transfer Transformer (Flan-T5), Large Language Model Meta AI (Llama3), and Generative Pre-trained Transformers-4 (GPT-4) using fine-tuning and in-context learning (ICL). In event extraction, the fine-tuned BERT and Llama3 models achieved the highest performance at 88.2-88.0 F1, which is comparable to the inter-annotator agreement (IAA) of 88.4 F1. In relation extraction, the fine-tuned BERT, Flan-T5, and Llama3 achieved the highest performance at 61.8-65.3 F1. GPT-4 with ICL achieved the worst performance across both tasks. The fine-tuned models significantly outperformed GPT-4 in ICL, highlighting the importance of annotated training data and model optimization. Furthermore, the BERT models performed similarly to Llama3. For our task, large language models offer no performance advantage over the smaller BERT models. We introduce CACER, a novel corpus with fine-grained annotations for medical problems, drugs, and their relationships in clinical narratives of oncology notes. State-of-the-art transformer models achieved performance comparable to IAA for several extraction tasks.

Deep learning-based structure segmentation and intramuscular fat annotation on lumbar magnetic resonance imaging.

Lumbar disc herniation (LDH) is a prevalent cause of low back pain. LDH patients commonly experience paraspinal muscle atrophy and fatty infiltration (FI), which further exacerbates the symptoms of low back pain. Magnetic resonance imaging (MRI) is crucial for assessing paraspinal muscle condition. Our study aims to develop a dual-model for automated muscle segmentation and FI annotation on MRI, assisting clinicians evaluate LDH conditions comprehensively. The study retrospectively collected data diagnosed with LDH from December 2020 to May 2022. The dataset was split into a 7:3 ratio for training and testing, with an external test set prepared to validate model generalizability. The model's performance was evaluated using average precision (AP), recall and F1 score. The consistency was assessed using the Dice similarity coefficient (DSC) and Cohen's Kappa. The mean absolute percentage error (MAPE) was calculated to assess the error of the model measurements of relative cross-sectional area (rCSA) and FI. Calculate the MAPE of FI measured by threshold algorithms to compare with the model. A total of 417 patients being evaluated, comprising 216 males and 201 females, with a mean age of 49 ± 15 years. In the internal test set, the muscle segmentation model achieved an overall DSC of 0.92 ± 0.10, recall of 92.60%, and AP of 0.98. The fat annotation model attained a recall of 91.30%, F1 Score of 0.82, and Cohen's Kappa of 0.76. However, there was a decrease on the external test set. For rCSA measurements, except for longissimus (10.89%), the MAPE of other muscles was less than 10%. When comparing the errors of FI for each paraspinal muscle, the MAPE of the model was lower than that of the threshold algorithm. The models demonstrate outstanding performance, with lower error in FI measurement compared to thresholding algorithms.

A generative hierarchical topic model for image annotation

A redesigned hierarchical model for image annotation is presented in this study, which can also be simplified to image classification. Based on some previous models for textual analysis, such as LDA and the hierarchical clustering model by Hoffman et al. in 1988, the author applies similar ideas to image data, extracting image features as words and setting the latent variables of topics. Compared with the popular annotation methods which are deep-learning-based, the model put forward in this document details the generating process of images by adding the layer of topics. Besides, compared to other hierarchical models such as the multi-modal hierarchical model by Barnard and Forsyth in 2001, the model in this paper simplifies it to a single-modal one and simultaneously achieves similar effects. We collect images under different human activities, such as riding horses, playing the guitar, etc., mainly from Google search engine and the result proves that our model works better than most previous probabilistic models.

ScTab: Scaling cross-tissue single-cell annotation models

Identifying cellular identities is a key use case in single-cell transcriptomics. While machine learning has been leveraged to automate cell annotation predictions for some time, there has been little progress in scaling neural networks to large data sets and in constructing models that generalize well across diverse tissues. Here, we propose scTab, an automated cell type prediction model specific to tabular data, and train it using a novel data augmentation scheme across a large corpus of single-cell RNA-seq observations (22.2 million cells). In this context, we show that cross-tissue annotation requires nonlinear models and that the performance of scTab scales both in terms of training dataset size and model size. Additionally, we show that the proposed data augmentation schema improves model generalization. In summary, we introduce a de novo cell type prediction model for single-cell RNA-seq data that can be trained across a large-scale collection of curated datasets and demonstrate the benefits of using deep learning methods in this paradigm.

A Hierarchical Multi-Task Learning Framework for Semantic Annotation in Tabular Data.

To optimize the utilization and analysis of tables, it is essential to recognize and understand their semantics comprehensively. This requirement is especially critical given that many tables lack explicit annotations, necessitating the identification of column types and inter-column relationships. Such identification can significantly augment data quality, streamline data integration, and support data analysis and mining. Current table annotation models often address each subtask independently, which may result in the neglect of constraints and contextual information, causing relational ambiguities and inference errors. To address this issue, we propose a unified multi-task learning framework capable of concurrently handling multiple tasks within a single model, including column named entity recognition, column type identification, and inter-column relationship detection. By integrating these tasks, the framework exploits their interrelations, facilitating the exchange of shallow features and the sharing of representations. Their cooperation enables each task to leverage insights from the others, thereby improving the performance of individual subtasks and enhancing the model's overall generalization capabilities. Notably, our model is designed to employ only the internal information of tabular data, avoiding reliance on external context or knowledge graphs. This design ensures robust performance even with limited input information. Extensive experiments demonstrate the superior performance of our model across various tasks, validating the effectiveness of unified multi-task learning framework in the recognition and comprehension of table semantics.

An Ensemble Spectral Prediction (ESP) model for metabolite annotation.

A key challenge in metabolomics is annotating measured spectra from a biological sample with chemical identities. Currently, only a small fraction of measurements can be assigned identities. Two complementary computational approaches have emerged to address the annotation problem: mapping candidate molecules to spectra, and mapping query spectra to molecular candidates. In essence, the candidate molecule with the spectrum that best explains the query spectrum is recommended as the target molecule. Despite candidate ranking being fundamental in both approaches, limited prior works incorporated rank learning tasks in determining the target molecule. We propose a novel machine learning model, Ensemble Spectral Prediction (ESP), for metabolite annotation. ESP takes advantage of prior neural network-based annotation models that utilize multilayer perceptron (MLP) networks and Graph Neural Networks (GNNs). Based on the ranking results of the MLP- and GNN-based models, ESP learns a weighting for the outputs of MLP and GNN spectral predictors to generate a spectral prediction for a query molecule. Importantly, training data is stratified by molecular formula to provide candidate sets during model training. Further, baseline MLP and GNN models are enhanced by considering peak dependencies through label mixing and multi-tasking on spectral topic distributions. When trained on the NIST 2020 dataset and evaluated on the relevant candidate sets from PubChem, ESP improves average rank by 23.7% and 37.2% over the MLP and GNN baselines, respectively, demonstrating performance gain over state-of-the-art neural network approaches. However, MLP approaches remain strong contenders when considering top five ranks. Importantly, we show that annotation performance is dependent on the training dataset, the number of molecules in the candidate set and candidate similarity to the target molecule. The ESP code, a trained model, and a Jupyter notebook that guide users on using the ESP tool is available at https://github.com/HassounLab/ESP.

The Effect of Annotation Quality on Wear Semantic Segmentation by CNN.

In this work, we investigate the impact of annotation quality and domain expertise on the performance of Convolutional Neural Networks (CNNs) for semantic segmentation of wear on titanium nitride (TiN) and titanium carbonitride (TiCN) coated end mills. Using an innovative measurement system and customized CNN architecture, we found that domain expertise significantly affects model performance. Annotator 1 achieved maximum mIoU scores of 0.8153 for abnormal wear and 0.7120 for normal wear on TiN datasets, whereas Annotator 3 with the lowest expertise achieved significantly lower scores. Sensitivity to annotation inconsistencies and model hyperparameters were examined, revealing that models for TiCN datasets showed a higher coefficient of variation (CV) of 16.32% compared to 8.6% for TiN due to the subtle wear characteristics, highlighting the need for optimized annotation policies and high-quality images to improve wear segmentation.

Advancing common bean (Phaseolus vulgaris L.) disease detection with YOLO driven deep learning to enhance agricultural AI

Common beans (CB), a vital source for high protein content, plays a crucial role in ensuring both nutrition and economic stability in diverse communities, particularly in Africa and Latin America. However, CB cultivation poses a significant threat to diseases that can drastically reduce yield and quality. Detecting these diseases solely based on visual symptoms is challenging, due to the variability across different pathogens and similar symptoms caused by distinct pathogens, further complicating the detection process. Traditional methods relying solely on farmers’ ability to detect diseases is inadequate, and while engaging expert pathologists and advanced laboratories is necessary, it can also be resource intensive. To address this challenge, we present a AI-driven system for rapid and cost-effective CB disease detection, leveraging state-of-the-art deep learning and object detection technologies. We utilized an extensive image dataset collected from disease hotspots in Africa and Colombia, focusing on five major diseases: Angular Leaf Spot (ALS), Common Bacterial Blight (CBB), Common Bean Mosaic Virus (CBMV), Bean Rust, and Anthracnose, covering both leaf and pod samples in real-field settings. However, pod images are only available for Angular Leaf Spot disease. The study employed data augmentation techniques and annotation at both whole and micro levels for comprehensive analysis. To train the model, we utilized three advanced YOLO architectures: YOLOv7, YOLOv8, and YOLO-NAS. Particularly for whole leaf annotations, the YOLO-NAS model achieves the highest mAP value of up to 97.9% and a recall of 98.8%, indicating superior detection accuracy. In contrast, for whole pod disease detection, YOLOv7 and YOLOv8 outperformed YOLO-NAS, with mAP values exceeding 95% and 93% recall. However, micro annotation consistently yields lower performance than whole annotation across all disease classes and plant parts, as examined by all YOLO models, highlighting an unexpected discrepancy in detection accuracy. Furthermore, we successfully deployed YOLO-NAS annotation models into an Android app, validating their effectiveness on unseen data from disease hotspots with high classification accuracy (90%). This accomplishment showcases the integration of deep learning into our production pipeline, a process known as DLOps. This innovative approach significantly reduces diagnosis time, enabling farmers to take prompt management interventions. The potential benefits extend beyond rapid diagnosis serving as an early warning system to enhance common bean productivity and quality.

Is Boundary Annotation Necessary? Evaluating Boundary-Free Approaches to Improve Clinical Named Entity Annotation Efficiency: Case Study.

Named entity recognition (NER) is a fundamental task in natural language processing. However, it is typically preceded by named entity annotation, which poses several challenges, especially in the clinical domain. For instance, determining entity boundaries is one of the most common sources of disagreements between annotators due to questions such as whether modifiers or peripheral words should be annotated. If unresolved, these can induce inconsistency in the produced corpora, yet, on the other hand, strict guidelines or adjudication sessions can further prolong an already slow and convoluted process. The aim of this study is to address these challenges by evaluating 2 novel annotation methodologies, lenient span and point annotation, aiming to mitigate the difficulty of precisely determining entity boundaries. We evaluate their effects through an annotation case study on a Japanese medical case report data set. We compare annotation time, annotator agreement, and the quality of the produced labeling and assess the impact on the performance of an NER system trained on the annotated corpus. We saw significant improvements in the labeling process efficiency, with up to a 25% reduction in overall annotation time and even a 10% improvement in annotator agreement compared to the traditional boundary-strict approach. However, even the best-achieved NER model presented some drop in performance compared to the traditional annotation methodology. Our findings demonstrate a balance between annotation speed and model performance. Although disregarding boundary information affects model performance to some extent, this is counterbalanced by significant reductions in the annotator's workload and notable improvements in the speed of the annotation process. These benefits may prove valuable in various applications, offering an attractive compromise for developers and researchers.

Interpretable AI analysis of chaos systems distribution in time series data from industrial robotics

In this study, the generalizability and distributivity of three different chaotic systems within an industrial robotics time series dataset are explored using an annotated artificial intelligence algorithm. A time series dataset derived from industrial robotics processes was constructed and transformed into the Runge-Kutta system, comprising fourth-order differential equations for normalization. Among the processed data, variables related to x-y-z positions underwent chaotic transformations through Lorenz, Chen, and Rossler chaos systems. The x variable and angle variables from the transformed x-y-z data were inputted into the InterpretML model, an annotated artificial intelligence model, to elucidate the effects of angle variables on the x position variable. As a result of this analysis, InterpretML Local analysis revealed a sensitivity of 0.05 for the Rossler chaos system, 0.15 for Chen, and 0.25 for Lorenz. Furthermore, global analysis indicated precision rates of 0.17 for Rossler, 0.255 for Chen, and 0.35 for Lorenz chaos systems. These sensitivity results suggest that the Rossler chaos system consistently provides more accurate results in both InterpretML local and global analyses compared to other chaotic systems. This study contributes significantly to the literature by analyzing the distributive and generalization properties of chaos systems and enhancing understanding of these systems.

Assessing citation integrity in biomedical publications: corpus annotation and NLP models.

Citations have a fundamental role in scholarly communication and assessment. Citation accuracy and transparency is crucial for the integrity of scientific evidence. In this work, we focus on quotation errors, errors in citation content that can distort the scientific evidence and that are hard to detect for humans. We construct a corpus and propose natural language processing (NLP) methods to identify such errors in biomedical publications. We manually annotated 100 highly-cited biomedical publications (reference articles) and citations to them. The annotation involved labeling citation context in the citing article, relevant evidence sentences in the reference article, and the accuracy of the citation. A total of 3063 citation instances were annotated (39.18% with accuracy errors). For NLP, we combined a sentence retriever with a fine-tuned claim verification model to label citations as ACCURATE, NOT_ACCURATE, or IRRELEVANT. We also explored few-shot in-context learning with generative large language models. The best performing model-which uses citation sentences as citation context, the BM25 model with MonoT5 reranker for retrieving top-20 sentences, and a fine-tuned MultiVerS model for accuracy label classification-yielded 0.59 micro-F1 and 0.52 macro-F1 score. GPT-4 in-context learning performed better in identifying accurate citations, but it lagged for erroneous citations (0.65 micro-F1, 0.45 macro-F1). Citation quotation errors are often subtle, and it is currently challenging for NLP models to identify erroneous citations. With further improvements, the models could serve to improve citation quality and accuracy. We make the corpus and the best-performing NLP model publicly available at https://github.com/ScienceNLP-Lab/Citation-Integrity/.

ROCOv2: Radiology Objects in COntext Version 2, an Updated Multimodal Image Dataset

Automated medical image analysis systems often require large amounts of training data with high quality labels, which are difficult and time consuming to generate. This paper introduces Radiology Object in COntext version 2 (ROCOv2), a multimodal dataset consisting of radiological images and associated medical concepts and captions extracted from the PMC Open Access subset. It is an updated version of the ROCO dataset published in 2018, and adds 35,705 new images added to PMC since 2018. It further provides manually curated concepts for imaging modalities with additional anatomical and directional concepts for X-rays. The dataset consists of 79,789 images and has been used, with minor modifications, in the concept detection and caption prediction tasks of ImageCLEFmedical Caption 2023. The dataset is suitable for training image annotation models based on image-caption pairs, or for multi-label image classification using Unified Medical Language System (UMLS) concepts provided with each image. In addition, it can serve for pre-training of medical domain models, and evaluation of deep learning models for multi-task learning.

Prediction of Plant Ubiquitylation Proteins and Sites by Fusing Multiple Features

Introduction: Protein ubiquitylation is an important post-translational modification (PTM), which is considered to be one of the most important processes regulating cell function and various diseases. Therefore, accurate prediction of ubiquitylation proteins and their PTM sites is of great significance for the study of basic biological processes and the development of related drugs. Researchers have developed some large-scale computational methods to predict ubiquitylation sites, but there is still much room for improvement. Much of the research related to ubiquitylation is cross-species while the life pattern is diversified, and the prediction method always shows its specificity in practical application. This study just aims at the issue of plants and has constructed computational methods for identifying ubiquitylation protein and ubiquitylation sites. Method: In this work, we constructed two predictive models to identify plant ubiquitylation proteins and sites. First, in the ubiquitylation proteins prediction model, in order to better reflect protein sequence information and obtain better prediction results, the KNN scoring matrix model based on functional domain Gene Ontology (GO) annotation and word embedding model, i.e. Skip-Gram and Continuous Bag of Words (CBOW), are used to extract the features, and the light gradient boosting machine (LGBM) is selected as the ubiquitylation proteins prediction engine. Results: As a result, accuracy (ACC), Precision, recall rate (Recall), F1_score and AUC are respectively 85.12%, 80.96%, 72.80%, 76.37% and 0.9193 in the 10-fold cross-validations on independent dataset. In the ubiquitylation sites prediction model, Skip-Gram, CBOW and enhanced amino acid composition (EAAC) feature extraction codes were used to extract protein sequence fragment features, and the predicted results on training and independent test data have also achieved good performance. Conclusion: In a word, the comparison results demonstrate that our models have a decided advantage in predicting ubiquitylation proteins and sites, and it may provide useful insights for studying the mechanisms and modulation of ubiquitination pathways

Stance Detection with Collaborative Role-Infused LLM-Based Agents

Stance detection automatically detects the stance in a text towards a target, vital for content analysis in web and social media research. Despite their promising capabilities, LLMs encounter challenges when directly applied to stance detection. First, stance detection demands multi-aspect knowledge, from deciphering event-related terminologies to understanding the expression styles in social media platforms. Second, stance detection requires advanced reasoning to infer authors' implicit viewpoints, as stances are often subtly embedded rather than overtly stated in the text. To address these challenges, we design a three-stage framework COLA (short for Collaborative rOle-infused LLM-based Agents) in which LLMs are designated distinct roles, creating a collaborative system where each role contributes uniquely. Initially, in the multidimensional text analysis stage, we configure the LLMs to act as a linguistic expert, a domain specialist, and a social media veteran to get a multifaceted analysis of texts, thus overcoming the first challenge. Next, in the reasoning-enhanced debating stage, for each potential stance, we designate a specific LLM-based agent to advocate for it, guiding the LLM to detect logical connections between text features and stance, tackling the second challenge. Finally, in the stance conclusion stage, a final decision maker agent consolidates prior insights to determine the stance. Our approach avoids extra annotated data and model training and is highly usable. We achieve state-of-the-art performance across multiple datasets. Ablation studies validate the effectiveness of each role design in handling stance detection. Further experiments have demonstrated the explainability and the versatility of our approach. Our approach excels in usability, accuracy, effectiveness, explainability and versatility, highlighting its value.