Accurate Labeling Research Articles

Genomic interrogation of commercial infant probiotic products for label accuracy.

To evaluate the label accuracy of commercial infant probiotic products and identify potential microbial contamination. DNA was extracted from seventeen infant probiotic products purchased from a large online vendor. Samples underwent 16S ribosomal RNA gene sequencing, QIIME analysis, and bacterial taxonomic classification. Identified bacterial species were compared with product labels for label accuracy and potential contaminants. Additionally, fungal DNA was amplified using qPCR with universal 18S fungal primers, and cultures were performed to assess viability. Over 82% of bacterial DNA extracted from samples corresponded to species listed on product labels. Contaminating bacteria were closely related species. Lot-to-lot variation in bacterial species abundance was greater in multi-strain products compared with single-strain products. Fungal DNA was detected in some samples, but culture results indicated these organisms were not viable. This pilot study highlights that most bacterial species in commercial infant probiotic products are accurately represented on labels. Single-strain products reliably match product labels whereas multi-strain products were more prone to variation of species abundance and species omission. Additionally, fungal contamination of limited samples was not viable. These findings emphasize the need for improved regulatory guidelines and standardization of probiotic products, particularly those marketed for infants. Commercial infant probiotic products were interrogated for label accuracy using 16S ribosomal RNA sequencing. Most bacterial species in commercial infant probiotic products are accurately represented on labels without pathogenic bacterial contamination. Live fungal contamination is absent. To date, no studies have interrogated infant probiotic products on this scale (other studies have been limited to single or dual strain infant probiotics). Since probiotics lack FDA regulation, this pilot study provides important information for pediatricians and consumers, further highlighting the need for improved quality control and standardization of probiotics.

Metabolic Tagging Technology of Exosomes- An Updated Review.

Exosomes are small extracellular vesicles secreted by various cell types, playing a crucial role in intercellular communication by carrying proteins, lipids, and nucleic acids, thus holding significant potential in diagnostics and therapeutics. Accurate labeling of exosomes is vital for studying their biogenesis, trafficking, and functional properties, enabling precise tracking and manipulation. This review examines current labeling techniques, including metabolic glycan labeling, chemical tagging, membrane fluorescent dyes, bio-conjugation, non-covalent labeling, and cell-engineering approaches. Each method is analyzed for its efficiency, specificity, and practicality, with attention to potential artifacts and challenges. Advancements in these techniques are essential for improving our understanding of exosome biology and developing exosome-based diagnostic and therapeutic strategies, providing researchers with valuable insights into state-of-the-art techniques and their applications in exosome research.

Creation and annotation of a code-mixed Kannada english dataset with accurate labels for detecting depression and major depressive disorder categories

Creation and annotation of a code-mixed Kannada english dataset with accurate labels for detecting depression and major depressive disorder categories

Efficient Generative-Adversarial U-Net for Multi-Organ Medical Image Segmentation.

Manual labeling of lesions in medical image analysis presents a significant challenge due to its labor-intensive and inefficient nature, which ultimately strains essential medical resources and impedes the advancement of computer-aided diagnosis. This paper introduces a novel medical image-segmentation framework named Efficient Generative-Adversarial U-Net (EGAUNet), designed to facilitate rapid and accurate multi-organ labeling. To enhance the model's capability to comprehend spatial information, we propose the Global Spatial-Channel Attention Mechanism (GSCA). This mechanism enables the model to concentrate more effectively on regions of interest. Additionally, we have integrated Efficient Mapping Convolutional Blocks (EMCB) into the feature-learning process, allowing for the extraction of multi-scale spatial information and the adjustment of feature map channels through optimized weight values. Moreover, the proposed framework progressively enhances its performance by utilizing a generative-adversarial learning strategy, which contributes to improvements in segmentation accuracy. Consequently, EGAUNet demonstrates exemplary segmentation performance on public multi-organ datasets while maintaining high efficiency. For instance, in evaluations on the CHAOS T2SPIR dataset, EGAUNet achieves approximately 2% higher performance on the Jaccard metric, 1% higher on the Dice metric, and nearly 3% higher on the precision metric in comparison to advanced networks such as Swin-Unet and TransUnet.

Commentary on Borodovsky et al.: Enhancing research on THC quantification-Consumer awareness through accurate labelling.

Commentary on Borodovsky et al.: Enhancing research on THC quantification-Consumer awareness through accurate labelling.

Application of Generative Artificial Intelligence Models for Accurate Prescription Label Identification and Information Retrieval for the Elderly in Northern East of Thailand.

This study introduces a novel AI-driven approach to support elderly patients in Thailand with medication management, focusing on accurate drug label interpretation. Two model architectures were explored: a Two-Stage Optical Character Recognition (OCR) and Large Language Model (LLM) pipeline combining EasyOCR with Qwen2-72b-instruct and a Uni-Stage Visual Question Answering (VQA) model using Qwen2-72b-VL. Both models operated in a zero-shot capacity, utilizing Retrieval-Augmented Generation (RAG) with DrugBank references to ensure contextual relevance and accuracy. Performance was evaluated on a dataset of 100 diverse prescription labels from Thai healthcare facilities, using RAG Assessment (RAGAs) metrics to assess Context Recall, Factual Correctness, Faithfulness, and Semantic Similarity. The Two-Stage model achieved high accuracy (94%) and strong RAGAs scores, particularly in Context Recall (0.88) and Semantic Similarity (0.91), making it well-suited for complex medication instructions. In contrast, the Uni-Stage model delivered faster response times, making it practical for high-volume environments such as pharmacies. This study demonstrates the potential of zero-shot AI models in addressing medication management challenges for the elderly by providing clear, accurate, and contextually relevant label interpretations. The findings underscore the adaptability of AI in healthcare, balancing accuracy and efficiency to meet various real-world needs.

Label credibility correction based on cell morphological differences for cervical cells classification

Cervical cancer is one of the deadliest cancers that pose a significant threat to women’s health. Early detection and treatment are commonly used methods to prevent cervical cancer. The use of pathological image analysis techniques for the automatic interpretation of cervical cells in pathological slides is a prominent area of research in the field of digital medicine. According to The Bethesda System, cervical cytology necessitates further classification of precancerous lesions based on positive interpretations. However, clinical definitions among different categories of lesion are complex and often characterized by fuzzy boundaries. In addition, pathologists can deduce different criteria for judgment based on The Bethesda System, leading to potential confusion during data labeling. Noisy labels due to this reason are a great challenge for supervised learning. To address the problem caused by noisy labels, we propose a method based on label credibility correction for cervical cell images classification network. Firstly, a contrastive learning network is used to extract discriminative features from cell images to obtain more similar intra-class sample features. Subsequently, these features are fed into an unsupervised method for clustering, resulting in unsupervised class labels. Then unsupervised labels are corresponded to the true labels to separate confusable and typical samples. Through a similarity comparison between the cluster samples and the statistical feature centers of each class, the label credibility analysis is carried out to group labels. Finally, a cervical cell images multi-class network is trained using synergistic grouping method. In order to enhance the stability of the classification model, momentum is incorporated into the synergistic grouping loss. Experimental validation is conducted on a dataset comprising approximately 60,000 cells from multiple hospitals, showcasing the effectiveness of our proposed approach. The method achieves 2-class task accuracy of 0.9241 and 5-class task accuracy of 0.8598. Our proposed method achieves better performance than existing classification networks on cervical cancer.

Understanding broader community perspectives on the scientific accuracy and stigma of personality trait labels.

Many labels are used within and across subfields to describe personality disorder (PD) and interpersonally-oriented trait dimensions. For example, "interpersonal disorders" is a suggested alternative label to "personality disorders" in clinical research. Other "dark trait" terms, though not proposed as formal labels for PDs, also are used in different research areas for describing externalizing traits. Terminology changes have been proposed both due to concerns about different descriptors' validity and their usage potentially being stigmatizing. Improving terminology consensus can also unify research and clinical assessment efforts, and we recruited participants from a range of sources who provided their views on terminology used in PD research toward this goal. This included data from 362 undergraduates, 408 adults recruited online, and 161 adults recruited from the community, and we used targeted recruitment strategies to ensure that individuals with a range of mental health histories were represented in our study. All participants completed questionnaires assessing their personalities and symptoms, and a subset of participants also completed structured clinical interviews. Results indicated that traditional "personality disorders" terminology were viewed favorably compared to other terms both in regard to scientific accuracy and stigma. Additionally, "interpersonal disorders" terminology was also viewed more favorably overall than many other terms, whereas "dark trait" terminology was viewed negatively. Participants' characteristics (e.g., personality, age) were mostly unrelated to their terminology views. These results provide insight into how various descriptors are viewed by the broader community and provide a foundation for future research investigating how different terms are perceived across contexts. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

An Aptamer-Initiated Catalytic hairpin assembly fluorescent biosensor for simultaneous detection of major seafood allergens in food system

Seafood allergy represents a critical public health challenge, necessitating the implementation of sensitive, precise, and quantitative detection methods for seafood allergens to guarantee accurate food labeling. Parvalbumin (PV) and tropomyosin (TM) are recognized as the primary allergens in fish and shellfish. Herein, we designed a fluorescence sensing system for the simultaneous detection of PV and TM that employs aptamer-initiated catalytic hairpin assembly (CHA). In this strategy, we utilized four hairpin dimers (H1, H2, H3, and H4) as foundational elements to create the sensing scaffolds. The CHA-based fluorescent aptasensor demonstrated an extensive linear range, ranging from 0.001 μg/mL to 100 μg/mL, with detection limits of 0.29 μg/mL for PV and 0.27 μg/mL for TM. This biosensor demonstrates exceptional selectivity and stability, in addition to good accuracy. Importantly, this biosensor has been effectively utilized to quantify PV and TM in four different commercial food products, achieving satisfactory accuracy. With these advantages, our proposed CHA-based aptasensor can offer a universal tool for sensitive and precise detection of food allergens in commercial foods throughout the supply chain.

Fast online feature selection in streaming data

The challenge of getting big amounts of high-quality labeled data is compounded by the fact that data labeling is often subjective and requires significant human effort. In many cases, the quality of the labeled data depends entirely on the expertise and experience of human annotators, making it challenging to ensure labeling accuracy in large and dynamic datasets. Moreover, there may be a significant delay between the arrival of a new instance and its manual labeling. This paper explores the use of fully unsupervised feature selection algorithms in non-stationary data streams, where the importance of features may change over time. We introduce a novel feature selection algorithm called Online Fast FEa-ture SELection-OFFESEL, which calculates the feature importance scores in each incoming window based on their mean normalized values and without using any class labels. We evaluate OFFESEL on 17 benchmark data streams, both stationary and non-stationary, using popular online classifiers like PerceptronMask, VFDT, Online Boosting, and Linear SVM. We compare OFFESEL to several other feature selection algorithms, including state-of-the-art supervised ones like FIRES and ABFS, as well as popular unsupervised ones like MCFS, LS, and Max Variance, which we adapted to data streams. Our results indicate that OFFESEL outperforms all supervised and unsupervised feature selection algorithms in terms of classification accuracy. Specifically, OFFESEL preserves the accuracy level of the supervised FIRES algorithm, which proved more accurate than ABFS in our experiments, while maintaining the accuracy level achieved by the unsupervised Max Variance algorithm. Moreover, OFFESEL requires even less computation time than Max Variance and shows high stability on stationary datasets. Overall, our study demonstrates the potential benefits of using unlabeled data for feature ranking and selection in dynamic data streams.

Cells Grouping Detection and Confusing Labels Correction on Cervical Pathology Images.

Cervical cancer is one of the most prevalent cancers among women, posing a significant threat to their health. Early screening can detect cervical precancerous lesions in a timely manner, thereby enabling the prevention or treatment of the disease. The use of pathological image analysis technology to automatically interpret cells in pathological slices is a hot topic in digital medicine research, as it can reduce the substantial effort required from pathologists to identify cells and can improve diagnostic efficiency and accuracy. Therefore, we propose a cervical cell detection network based on collecting prior knowledge and correcting confusing labels, called PGCC-Net. Specifically, we utilize clinical prior knowledge to break down the detection task into multiple sub-tasks for cell grouping detection, aiming to more effectively learn the specific structure of cells. Subsequently, we merge region proposals from grouping detection to achieve refined detection. In addition, according to the Bethesda system, clinical definitions among various categories of abnormal cervical cells are complex, and their boundaries are ambiguous. Differences in assessment criteria among pathologists result in ambiguously labeled cells, which poses a significant challenge for deep learning networks. To address this issue, we perform a labels correction module with feature similarity by constructing feature centers for typical cells in each category. Then, cells that are easily confused are mapped with these feature centers in order to update cells' annotations. Accurate cell labeling greatly aids the classification head of the detection network. We conducted experimental validation on a public dataset of 7410 images and a private dataset of 13,526 images. The results indicate that our model outperforms the state-of-the-art cervical cell detection methods.

Applications of deep learning in Alzheimer’s disease: a systematic literature review of current trends, methodologies, challenges, innovations, and future directions

Alzheimer’s Disease (AD) constitutes a significant global health issue. In the next 40 years, it is expected to affect 106 million people. Although more and more people are getting AD, there are still no effective drugs to treat it. Insightful information about how important it is to find and treat AD quickly. Recently, Deep Learning (DL) techniques have been used more and more to diagnose AD. They claim better accuracy in drug reuse, medication recognition, and labeling. This essay meticulously examines the works that have talked about using DL with Alzheimer’s disease. Some of the methods are Natural Language Processing (NLP), drug reuse, classification, and identification. Concerning these methods, we examine their pros and cons, paying special attention to how easily they can be explained, how safe they are, and how they can be used in medical situations. One important finding is that Convolutional Neural Networks (CNNs) are most often used for AD research and Python is most often used for DL issues. Some security problems, like data protection and model stability, are not looked at enough in the present research, according to us. This study thoroughly examines present methods and also points out areas that need more work, like better data integration and AI systems that can be explained. The findings should help guide more research and speed up the creation of DL-based AD identification tools in the future.

Pioneering automation in agricultural subsidy processing through deep learning for computer vision

ABSTRACT Validating farmers’ eligibility for subsidies under the European Union’s Common Agricultural Policy (CAP) requires substantial resources. An automated solution can minimize human errors and save time. This study leverages deep learning to automate the validation process using ground-level landscape images (GLLP), an approach not previously explored. We conducted two rigorous experiments to identify the optimal model architecture and assess the impact of data cleaning, accurate labelling, reduced feature ambiguity, and stricter photo collection protocols on performance. Our results reveal that while the ResNet-18 architecture achieved a recall of 0.61 on the original dataset, distinguishing between eligible and ineligible grasslands was challenging. By addressing image quality and enforcing standardized data collection, accuracy improved to 0.89 and recall to 0.94. Crucially, our research emphasizes the development and validation of effective data collection strategies, ensuring uniformity and high-quality data essential for accurate model performance. Our findings affirm the potential of deep learning to automate subsidy eligibility classification, contingent on resolving challenges related to the quality and standardization of GLLP data.

Label Accuracy and Quality of Select Weight-Loss Dietary Supplements Sold on or near US Military Bases.

Background/Objectives: Sixty-eight percent of service members are living with overweight or obesity, some who may not consult a healthcare provider when they decide to lose weight. Instead, they often turn to weight-loss dietary supplements for self-care solutions. The purpose of this case series study was to examine the label accuracy and quality of select weight-loss dietary supplements sold on or near US military bases across the country. Methods: Weight-loss dietary supplements (n = 44) were selected and purchased in GNCs, Exchanges, and Shoppettes across 12 military installations from March 2023 to January 2024. Liquid chromatography-mass spectrometry was used to verify the label accuracy according to the Supplement Facts listed ingredients and whether they contained any ingredients prohibited for use in the military. Product label claims were analyzed using the Operation Supplement Safety (OPSS) Risk Assessment Scorecard. Results: Thirty-six products (82%) had inaccurate labels. Twenty-seven (61%) had ingredients listed on the label not detected through analysis. Sixteen products (36%) contained hidden ingredients. The four products purchased within one mile from the base listed multiple prohibited ingredients on the labels, with all detected. Forty (91%) products scored a "no-go" according to the OPSS Scorecard and none contained a third-party certification seal on the label. Multiple stimulants were included in products such that the product safety was unknown. Conclusions: The majority of weight-loss dietary supplements analyzed in this case series study had inaccurate labels and were considered a "no-go" according to the Scorecard. Service members should only have access to safe, high-quality dietary supplement products. OPSS is collaborating with the Department of Defense stakeholders to determine the most effective ways for service members to have access to third-party certified products on all military establishments.

18F-FDG metabolic abnormalities precede perfusion and atrophic changes in diagnosis of early frontotemporal dementia: study from a tertiary care university hospital.

Diagnosis of early onset dementia is critical for initiating management. Although structural MRI is an established procedure for dementia evaluation, early cases may be missed. Neurodegenerative diseases lead to reductions in glucose consumption and grey matter volume loss. Our primary aim was to establish whether metabolic changes precede perfusion abnormalities in early cases of dementia especially, frontotemporal dementia (FTD). Secondly to study if cerebral atrophy using Pasquier visual rating scales can be used reliably to correlate with hypometabolism in this group of patients. A total of 56 patients (M:F = 39:17) with memory loss as per the DSM-5 diagnostic criteria were clinically and neurologically examined and referred for 18F-fluorodeoxyglucose (FDG) PET brain imaging. A few patients who had a prior (recent, <1 week) MR brain underwent brain 18F-FDG PET-CT, and all others were considered for simultaneous 18F-FDG PETMR imaging of brain. T2-weighted images were used to report Pasquier rating scales in all our patients as per recommendation. Cognitive assessments were analysed along with neuroimaging findings. Highest diagnostic performance was obtained with 18F-FDG PET for identifying early FTD in our series of patients. Sensitivity, specificity and accuracy of FDG and arterial spin labeling (ASL) using simultaneous PETMR were found to be 96.34%/90.1%/89 : 53.57%/62.12%/78, respectively. Cerebral atrophy rated using Pasquier visual scales showed the lowest diagnostic performance. Our study showed that the earliest phase of cognitive decline was found to be associated with specific patterns of hypometabolism, even in the absence of atrophy, which are currently considered diagnostic biomarkers. Metabolic derangements precede perfusion changes and cerebral atrophy in the setting of early dementia. Simultaneous FDG PETMR is being recommended as the investigation of choice for the evaluation of early FTD. This pilot study shows that FDG PET outperforms cognitive assessments and anatomical imaging modalities in early dementia evaluation. Although ASL can detect perfusion deficits in dementia, compared with FDG PET, its sensitivity and specificity for discerning early onset dementia from controls remain inferior to FDG PET. Pasquier scales, although easy to implement, cannot replace FDG PET metabolic findings, which start very early in the neuronal disease process.

Improved self-training-based distant label denoising method for cybersecurity entity extractions.

The task of named entity recognition (NER) plays a crucial role in extracting cybersecurity-related information. Existing approaches for cybersecurity entity extraction predominantly rely on manual labelling data, resulting in labour-intensive processes due to the lack of a cybersecurity-specific corpus. In this paper, we propose an improved self-training-based distant label denoising method for cybersecurity entity extraction. Firstly, we create two domain dictionaries of cybersecurity. Then, an algorithm that combines reverse maximum matching and part-of-speech tagging restrictions is proposed, for generating distant labels for the cybersecurity domain corpus. Lastly, we propose a high-confidence text selection method and an improved self-training algorithm that incorporates a teacher-student model and weight update constraints, for exploring the true labels of low-confidence text using a model trained on high-confidence text, thereby reducing the noise in the distant annotation data. Experimental results demonstrate that the cybersecurity distantly-labelled data we obtained is of high quality. Additionally, the proposed constrained self-training algorithm effectively improves the F1 score of several state-of-the-art NER models on this dataset, yielding a 3.5% improvement for the Vendor class and a 3.35% improvement for the Product class.

Human activity recognition utilizing optimized attention induced Multihead Convolutional Neural Network with Mobile Net V1 from Mobile health data.

Human Activity Recognition (HAR) systems are designed to continuously monitor human behaviour, mainly in the areas of entertainment and surveillance in intelligent home environments. In this manuscript, Human Activity Recognition utilizing optimized Attention Induced Multi head Convolutional Neural Network with Mobile Net V1 from Mobile Health Data (HAR-AMCNN-MNV1) is proposed. The input data is collected through MHEALTH and UCI HAR datasets. Neural Spectrospatial Filtering (NSF) is used for avoiding accurate labelling and reduces errors. Afterwards, Variational Density Peak Clustering Algorithm (VDPCA) is used for segmenting the data. Feature Extraction and Classification is done by Attention Induced Multi head Convolutional Neural Network with Mobile Net V1 (AMCNN-MNV1). AMCNN is used for extracting Hand-crafted features. AMCNN-MNV1 effectively classifies the human activities as Sitting and relaxing (Sit), Climbing stairs (CS), Walking (Walk), Standing still (Std), Waist bends forward (WBF), Frontal elevation of arms (FEA), Jogging (Jog), Knees bending (crouching) (KB), Cycling (Cycl), Lying down (Lay), Jump front & back (JFB) and Running (Run). Siberian Tiger Optimization Algorithm (STOA) is proposed to optimize the weight parameter of AMCNN-MNV1 classifier. The proposed method attains 21.19%, 23.45%, and 21.76% higher accuracy, 31.15%, 24.65% and 22.72% higher precision; 21.15%, 20.18%, and 21.28% higher recall evaluated to the existing methods.

Semi‐Supervised Medical Image Segmentation Based on Feature Similarity and Multi‐Level Information Fusion Consistency

ABSTRACTSemantic segmentation is a key task in computer vision, with medical image segmentation as a prominent downstream application that has seen significant advancements in recent years. However, the challenge of requiring extensive annotations in medical image segmentation remains exceedingly difficult. In addressing this issue, semi‐supervised semantic segmentation has emerged as a new approach to mitigate annotation burdens. Nonetheless, existing methods in semi‐supervised medical image segmentation still face challenges in fully exploiting unlabeled data and efficiently integrating labeled and unlabeled data. Therefore, this paper proposes a novel network model—feature similarity multilevel information fusion network (FSMIFNet). First, the feature similarity module is introduced to harness deep feature similarity among unlabeled images, predicting true label constraints and guiding segmentation features with deep feature relationships. This approach fully exploits deep feature information from unlabeled data. Second, the multilevel information fusion framework integrates labeled and unlabeled data to enhance segmentation quality in unlabeled images, ensuring consistency between original and feature maps for comprehensive optimization of detail and global information. In the ACDC dataset, our method achieves an mDice of 0.684 with 5% labeled data, 0.873 with 10%, 0.884 with 20%, and 0.897 with 50%. Experimental results demonstrate the effectiveness of FSMIFNet in semi‐supervised semantic segmentation of medical images, outperforming existing methods on public benchmark datasets. The code and models are available at https://github.com/liujiayin12/FSMIFNet.git.

Analysis Of Machine Learning Algorithms For Image Classification

This research aims to use a subset of the CIFAR100 dataset to develop and present an empirical examination of the performance of powerful Convolutional Neural Networks (CNN) and Support Vector Machines (SVM) for object detection. Image datasets, CIFAR10, CIFAR 100, and MINIST image datasets are all common benchmark datasets for testing performance. This research examines the well-known dataset CIFAR 100. Both classifiers' important attributes are combined in the constructed models. The suggested model's algorithm is trained and tested using images from the CIFAR-100 dataset. The CIFAR 100 is made up of 100 different classes, each with 600 photos. CNN and SVM's receptive fields aid in automatically extracting the most identifiable features from these images. The experimental results confirm the effectiveness of the implementation by reaching a maximum recognition accuracy of 0.4919 percent. Results were obtained as maximum accuracy of 0.4817% For the CNN on Coarse Super Label and maximum accuracy of 0.3696% For the CNN on Fine Super Label, while SVM on Coarse Super Label resulted in an accuracy of 0.0795% and SVM on Fine super Label resulted in 0.0285%. According to the results, CNN was able to exceed the benchmark of 39.43% and 24.49%, but SVM was unable to do so.

Trace Copilot: Automatically Locating Cryptographic Operations in Side-Channel Traces by Firmware Binary Instrumenting

A common assumption in side-channel analysis is that the attacker knows the cryptographic algorithm implementation of the victim. However, many labsetting studies implicitly extend this assumption to the knowledge of the source code, by inserting triggers to measure, locate or align the Cryptographic Operations (CO) in the trace. For real-world attacks, the source code is typically unavailable, which poses a challenge for locating the COs thus reducing the effectiveness of many methods. In contrast, obtaining the (partial) binary firmware is more prevalent in practical attacks on embedded devices. While binary code theoretically encapsulates necessary information for side-channel attacks on software-implemented cryptographic algorithms, there is no systematic study on leveraging this information to facilitate side-channel analysis. This paper introduces a novel and general framework that utilizes binary information for the automated locating of COs on side-channel traces. We first present a mechanism that maps the execution flow of binary instructions onto the corresponding side-channel trace through a tailored static binary instrumentation process, thereby transforming the challenge of locating COs into one of tracing cryptographic code execution within the binary. For the latter, we propose a method to retrieve binary instruction addresses that are equivalent to the segmenting boundaries of the COs within side-channel traces. By identifying the mapping points of these instructions on the trace, we can obtain accurate segmentation labeling for the sidechannel data. Further, by employing the well-labeled side-channel segments obtained on a profiling device, we can readily identify the locations of COs within traces collected from un-controllable target devices. We evaluate our approach on various devices and cryptographic software, including a real-world secure boot program. The results demonstrate the effectiveness of our method, which can automatically locate typical COs, such as AES or ECDSA, in raw traces using only the binary firmware and a profiling device. Comparison experiments indicate that our method outperforms existing techniques in handling noisy or jittery traces and scales better to complex COs. Performance evaluation confirms that the runtime and storage overheads of the proposed approach are practical for real-world deployment.