The effect of the apolipoprotein E ε4 allele and olfactory function on odor identification networks.

INTRODUCTION: The combination of apolipoprotein E ε4 (ApoE ε4) status, odor identification, and odor familiarity predicts conversion to mild cognitive impairment (MCI) and Alzheimer's disease (AD). METHODS: To further understand olfactory disturbances and AD risk, ApoE ε4 carrier (mean age 76.38 ± 5.21) and ε4 non-carrier (mean age 76.8 ± 3.35) adults were given odor familiarity and identification tests and performed an odor identification task during fMRI scanning. Five task-related functional networks were detected using independent components analysis. Main and interaction effects of mean odor familiarity ratings, odor identification scores, and ε4 status on network activation and task-modulation of network functional connectivity (FC) during correct and incorrect odor identification (hits and misses), controlling for age and sex, were explored using multiple linear regression. RESULTS: Findings suggested that sensory-olfactory network activation was positively associated with odor identification scores in ε4 carriers with intact odor familiarity. The FC of sensory-olfactory, multisensory-semantic integration, and occipitoparietal networks was altered in ε4 carriers with poorer odor familiarity and identification. In ε4 carriers with poorer familiarity, connectivity between superior frontal areas and the sensory-olfactory network was negatively associated with odor identification scores. CONCLUSIONS: The results contribute to the clarification of the neurocognitive structure of odor identification processing and suggest that poorer odor familiarity and identification in ε4 carriers may signal multi-network dysfunction. Odor familiarity and identification assessment in ε4 carriers may contribute to the predictive value of risk for MCI and AD due to the breakdown of sensory-cognitive network integration. Additional research on olfactory processing in those at risk for AD is warranted.

Have you been there before? Decoding recognition of spatial scenes from fMRI signals in precuneus.

One potential application of forensic "brain reading" is to test whether a suspect has previously experienced a crime scene. Here, we investigated whether it is possible to decode real life autobiographic exposure to spatial locations using fMRI. In the first session, participants visited four out of eight possible rooms on a university campus. During a subsequent scanning session, subjects passively viewed pictures and videos from these eight possible rooms (four old, four novel) without giving any responses. A multivariate searchlight analysis was employed that trained a classifier to distinguish between "seen" versus "unseen" stimuli from a subset of six rooms. We found that bilateral precuneus encoded information that can be used to distinguish between previously seen and unseen rooms and that also generalized to the two stimuli left out from training. We conclude that activity in bilateral precuneus is associated with the memory of previously visited rooms, irrespective of the identity of the room, thus supporting a parietal contribution to episodic memory for spatial locations. Importantly, we could decode whether a room was visited in real life without the need of explicit judgments about the rooms. This suggests that recognition is an automatic response that can be decoded from fMRI data, thus potentially supporting forensic applications of concealed information tests for crime scene recognition.

Brain areas interconnected to ventral pathway circuits are independently able to induce enhancement in object recognition memory and cause reversal in object recognition memory deficit.

AIMS: Ventral pathway circuits are constituted by the interconnected brain areas that are distributed throughout the brain. These brain circuits are primarily involved in processing of object related information in brain. However, their role in object recognition memory (ORM) enhancement remains unknown. Here, we have studied on the implication of these circuits in ORM enhancement and in reversal of ORM deficit in aging. METHODS: The brain areas interconnected to ventral pathway circuits in rat brain were activated by an expression of a protein called regulator of G-protein signaling 14 of 414 amino acids (RGS14414). RGS14414 is an ORM enhancer and therefore used here as a gain-in-function tool. ORM test and immunohistochemistry, lesions, neuronal arborization, and knockdown studies were performed to uncover the novel function of ventral pathway circuits. RESULTS: An activation of each of the brain areas interconnected to ventral pathway circuits individually induced enhancement in ORM; however, same treatment in brain areas not interconnected to ventral pathway circuits produced no effect. Further study in perirhinal cortex (PRh), area V2 of visual cortex and frontal cortex (FrC), which are brain areas that have been shown to be involved in ORM and are interconnected to ventral pathway circuits, revealed that ORM enhancement seen after the activation of any one of the three brain areas was unaffected by the lesions in other two brain areas either individually in each area or even concurrently in both areas. This ORM enhancement in all three brain areas was associated to increase in structural plasticity of pyramidal neurons where more than 2-fold higher dendritic spines were observed. Additionally, we found that an activation of either PRh, area V2, or FrC not only was adequate but also was sufficient for the reversal of ORM deficit in aging rats, and the blockade of RGS14414 activity led to loss in increase in dendritic spine density and failure in reversal of ORM deficit. CONCLUSIONS: These results suggest that brain areas interconnected to ventral pathway circuits facilitate ORM enhancement by an increase in synaptic connectivity between the local brain area circuits and the passing by ventral pathway circuits and an upregulation in activity of ventral pathway circuits. In addition, the finding of the reversal of ORM deficit through activation of an interconnected brain area might serve as a platform for developing not only therapy against memory deficits but also strategies for other brain diseases in which neuronal circuits are compromised.

Effects of two different social exclusion paradigms on ambiguous facial emotion recognition.

Social exclusion is an emotionally painful experience that leads to various alterations in socio-emotional processing. The perceptual and emotional consequences that may arise from experiencing social exclusion can vary depending on the paradigm used to manipulate it. Exclusion paradigms can vary in terms of the severity and duration of the leading exclusion experience, thereby classifying it as either a short-term or long-term experience. The present study aimed to study the impact of exclusion on socio-emotional processing using different paradigms that caused experiencing short-term and imagining long-term exclusion. Ambiguous facial emotions were used as socio-emotional cues. In study 1, the Ostracism Online paradigm was used to manipulate short-term exclusion. In study 2, a new sample of participants imagined long-term exclusion through the future life alone paradigm. Participants of both studies then completed a facial emotion recognition task consisting of morphed ambiguous facial emotions. By means of Point of Subjective Equivalence analyses, our results indicate that the experience of short-term exclusion hinders recognising happy facial expressions. In contrast, imagining long-term exclusion causes difficulties in recognising sad facial expressions. These findings extend the current literature, suggesting that not all social exclusion paradigms affect socio-emotional processing similarly.

Interaction of high-fat diet and brain trauma alters adipose tissue macrophages and brain microglia associated with exacerbated cognitive dysfunction.

Obesity increases the morbidity and mortality of traumatic brain injury (TBI). Detailed analyses of transcriptomic changes in the brain and adipose tissue were performed to elucidate the interactive effects between high-fat diet-induced obesity (DIO) and TBI. Adult male mice were fed a high-fat diet (HFD) for 12 weeks prior to experimental TBI and continuing after injury. High-throughput transcriptomic analysis using Nanostring panels of the total visceral adipose tissue (VAT) and cellular components in the brain, followed by unsupervised clustering, principal component analysis, and IPA pathway analysis were used to determine shifts in gene expression patterns and molecular pathway activity. Cellular populations in the cortex and hippocampus, as well as in VAT, during the chronic phase after combined TBI-HFD showed amplification of central and peripheral microglia/macrophage responses, including superadditive changes in selected gene expression signatures and pathways. Furthermore, combined TBI and HFD caused additive dysfunction in Y-Maze, Novel Object Recognition (NOR), and Morris water maze (MWM) cognitive function tests. These novel data suggest that HFD-induced obesity and TBI can independently prime and support the development of altered states in brain microglia and VAT, including the disease-associated microglia/macrophage (DAM) phenotype observed in neurodegenerative disorders. The interaction between HFD and TBI promotes a shift toward chronic reactive microglia/macrophage transcriptomic signatures and associated pro-inflammatory disease-altered states that may, in part, underlie the exacerbation of cognitive deficits. Thus, targeting of HFD-induced reactive cellular phenotypes, including in peripheral adipose tissue immune cell populations, may serve to reduce microglial maladaptive states after TBI, attenuating post-traumatic neurodegeneration and neurological dysfunction.

A deep learning-based dynamic deformable adaptive framework for locating the root region of the dynamic flames.

Traditional optical flame detectors (OFDs) in flame detection are susceptible to environmental interference, which will inevitably cause detection errors and miscalculations when confronted with a complex environment. The conventional deep learning-based models can mitigate the interference of complex environments by flame image feature extraction, which significantly improves the precision of flame recognition. However, these models focus on identifying the general profile of the static flame, but neglect to effectively locate the source of the dynamic flame. Therefore, this paper proposes a novel dynamic flame detection method named Dynamic Deformable Adaptive Framework (DDAF) for locating the flame root region dynamically. Specifically, to address limitations in flame feature extraction of existing detection models, the Deformable Convolution Network v2 (DCNv2) is introduced for more flexible adaptation to the deformations and scale variations of target objects. The Context Augmentation Module (CAM) is used to convey flame features into Dynamic Head (DH) to feature extraction from different aspects. Subsequently, the Layer-Adaptive Magnitude-based Pruning (LAMP) where the connection with the smallest LAMP score is pruned sequentially is employed to further enhance the speed of model detection. More importantly, both the coarse- and fine-grained location techniques are designed in the Inductive Modeling (IM) to accurately delineate the flame root region for effective fire control. Additionally, the Temporal Consistency-based Detection (TCD) contributes to improving the robustness of model detection by leveraging the temporal information presented in consecutive frames of a video sequence. Compared with the classical deep learning method, the experimental results on the custom flame dataset demonstrate that the AP0.5 value is improved by 4.4%, while parameters and FLOPs are reduced by 25.3% and 25.9%, respectively. The framework of this research extends applicability to a variety of flame detection scenarios, including industrial safety and combustion process control.

Does culture moderate the encoding and recognition of negative cues? Evidence from an eye-tracking study.

Cross-cultural research has elucidated many important differences between people from Western European and East Asian cultural backgrounds regarding how each group encodes and consolidates the contents of complex visual stimuli. While Western European groups typically demonstrate a perceptual bias towards centralised information, East Asian groups favour a perceptual bias towards background information. However, this research has largely focused on the perception of neutral cues and thus questions remain regarding cultural group differences in both the perception and recognition of negative, emotionally significant cues. The present study therefore compared Western European (n = 42) and East Asian (n = 40) participants on a free-viewing task and a subsequent memory task utilising negative and neutral social cues. Attentional deployment to the centralised versus background components of negative and neutral social cues was indexed via eye-tracking, and memory was assessed with a cued-recognition task two days later. While both groups demonstrated an attentional bias towards the centralised components of the neutral cues, only the Western European group demonstrated this bias in the case of the negative cues. There were no significant differences observed between Western European and East Asian groups in terms of memory accuracy, although the Western European group was unexpectedly less sensitive to the centralised components of the negative cues. These findings suggest that culture modulates low-level attentional deployment to negative information, however not higher-level recognition after a temporal interval. This paper is, to our knowledge, the first to concurrently consider the effect of culture on both attentional outcomes and memory for both negative and neutral cues.

Customizing spatial remapping of letters to aid reading in the presence of a simulated central field loss.

Reading is a primary concern of patients with central field loss (CFL) because it is typically performed with foveal vision. Spatial remapping offers one potential avenue to aid in reading; it entails shifting occluded letters to retinal areas where vision is functional. Here, we introduce a method of creating and testing different remapping strategies-ways to remap text-customized for CFL of different shapes. By simulating CFL in typically-sighted individuals, we tested the customization hypothesis-that the benefits of different remapping strategies will depend on the properties of the CFL. That is, remapping strategies will aid reading differentially in the presence of differently shaped CFL. In Experiment 1, letter recognition in the presence of differently shaped CFL was assessed in and around central vision. Using these letter recognition "maps" different spatial remappings were created and tested in Experiment 2 using a word recognition task. Results showed that the horizontal gap remapping, which did not remap any letters vertically, resulted in the best word recognition. Results were also consistent with the customization hypothesis; the benefits of different remappings on word recognition depended on the different CFL shapes. Although the horizontal gap remapping resulted in very good word recognition, tailoring remapping strategies to the shape of patients' CFL may aid reading with the wide range of sizes and shapes encountered by patients with CFL.

The Role of Attention in Category Representation.

Numerous studies have found that selective attention affects category learning. However, previous research did not distinguish between the contribution of focusing and filtering components of selective attention. This study addresses this issue by examining how components of selective attention affect category representation. Participants first learned a rule-plus-similarity category structure, and then were presented with category priming followed by categorization and recognition tests. Additionally, to evaluate the involvement of focusing and filtering, we fit models with different attentional mechanisms to the data. In Experiment 1, participants received rule-based category training, with specific emphasis on a single deterministic feature (D feature). Experiment 2 added a recognition test to examine participants' memory for features. Both experiments indicated that participants categorized items based solely on the D feature, showed greater memory for the D feature, were primed exclusively by the D feature without interference from probabilistic features (P features), and were better fit by models with focusing and at least one type of filtering mechanism. The results indicated that selective attention distorted category representation by highlighting the D feature and attenuating P features. To examine whether the distorted representation was specific to rule-based training, Experiment 3 introduced training, emphasizing all features. Under such training, participants were no longer primed by the D feature, they remembered all features well, and they were better fit by the model assuming only focusing but no filtering process. The results coupled with modeling provide novel evidence that while both focusing and filtering contribute to category representation, filtering can also result in representational distortion.

Randomness impacts the building of specific priors, visual exploration, and perception in object recognition.

Recognising objects is a vital skill on which humans heavily rely to respond quickly and adaptively to their environment. Yet, we lack a full understanding of the role visual information sampling plays in this process, and its relation to the individual's priors. To bridge this gap, the eye-movements of 18 adult participants were recorded during a free-viewing object-recognition task using Dots stimuli1. Participants viewed the stimuli in one of three orders: from most visible to least (Descending), least visible to most (Ascending), or in a randomised order (Random). This dictated the strength of their priors along the experiment. Visibility order influenced the participants' recognition performance and visual exploration. In addition, we found that while orders allowing for stronger priors generally led participants to visually sample more informative locations, this was not the case of Random participants. Indeed, they appeared to behave naïvely, and their use of specific object-related priors was fully impaired, while they maintained the ability to use general, task-related priors to guide their exploration. These findings have important implications for our understanding of perception, which appears to be influenced by complex cognitive processes, even at the basic level of visual sampling during object recognition.

Pattern Mining-Based Pig Behavior Analysis for Health and Welfare Monitoring.

The increasing popularity of pigs has prompted farmers to increase pig production to meet the growing demand. However, while the number of pigs is increasing, that of farm workers has been declining, making it challenging to perform various farm tasks, the most important among them being managing the pigs' health and welfare. This study proposes a pattern mining-based pig behavior analysis system to provide visualized information and behavioral patterns, assisting farmers in effectively monitoring and assessing pigs' health and welfare. The system consists of four modules: (1) data acquisition module for collecting pigs video; (2) detection and tracking module for localizing and uniquely identifying pigs, using tracking information to crop pig images; (3) pig behavior recognition module for recognizing pig behaviors from sequences of cropped images; and (4) pig behavior analysis module for providing visualized information and behavioral patterns to effectively help farmers understand and manage pigs. In the second module, we utilize ByteTrack, which comprises YOLOx as the detector and the BYTE algorithm as the tracker, while MnasNet and LSTM serve as appearance features and temporal information extractors in the third module. The experimental results show that the system achieved a multi-object tracking accuracy of 0.971 for tracking and an F1 score of 0.931 for behavior recognition, while also highlighting the effectiveness of visualization and pattern mining in helping farmers comprehend and manage pigs' health and welfare.

The EEG-Based Fusion Entropy-Featured Identification of Isometric Contraction Forces under the Same Action.

This study explores the important role of assessing force levels in accurately controlling upper limb movements in human-computer interfaces. It uses a new method that combines entropy to improve the recognition of force levels. This research aims to differentiate between different levels of isometric contraction forces using electroencephalogram (EEG) signal analysis. It integrates eight different entropy measures: power spectrum entropy (PSE), singular spectrum entropy (SSE), logarithmic energy entropy (LEE), approximation entropy (AE), sample entropy (SE), fuzzy entropy (FE), alignment entropy (PE), and envelope entropy (EE). The findings emphasize two important advances: first, including a wide range of entropy features significantly improves classification efficiency; second, the fusion entropy method shows exceptional accuracy in classifying isometric contraction forces. It achieves an accuracy rate of 91.73% in distinguishing between 15% and 60% maximum voluntary contraction (MVC) forces, along with 69.59% accuracy in identifying variations across 15%, 30%, 45%, and 60% MVC. These results illuminate the efficacy of employing fusion entropy in EEG signal analysis for isometric contraction detection, heralding new opportunities for advancing motor control and facilitating fine motor movements through sophisticated human-computer interface technologies.

The Monitoring and Cell Imaging of Fe3+ Using a Chromone-Based Fluorescence Probe.

A new structurally simple fluorescent CP probe based on chromone was designed and synthesized, and its structure was fully characterized using various analytical techniques. The CP probe displays a high selectivity and sensitivity for sensing Fe3+ with a "turn-off" fluorescence response over other metal ions in a DMSO/H2O (4:1, v/v) solution. The experiment results show that the CP probe is stable over a wide pH range of 2.0-12.0. The detection limit for Fe3+ was calculated to be 0.044 µmol•L-1. The molar ratio method indicated that the binding mode between the CP probe and Fe3+ is a 1:1 complex formation. HR-MS and density functional theory (DFT) calculations were also performed to further confirm the recognition mechanism. Both fluorescence imaging experiments and the MTT assay demonstrated that the CP probe was suitable for detecting intracellular Fe3+ and no significant cytotoxicity in living cells.

[18F]-Radiolabelled Nanoplatforms: A Critical Review of Their Intrinsic Characteristics, Radiolabelling Methods, and Purification Techniques.

A wide range of nano-objects is found in many applications of our everyday life. Recognition of their peculiar properties and ease of functionalization has prompted their engineering into multifunctional platforms that are supposed to afford efficient tools for the development of biomedical applications. However, bridging the gap between bench to bedside cannot be expected without a good knowledge of their behaviour in vivo, which can be obtained through non-invasive imaging techniques, such as positron emission tomography (PET). Their radiolabelling with [18F]-fluorine, a technique already well established and widely used routinely for PET imaging, with [18F]-FDG for example, and in preclinical investigation using [18F]-radiolabelled biological macromolecules, has, therefore, been developed. In this context, this review highlights the various nano-objects studied so far, the reasons behind their radiolabelling, and main in vitro and/or in vivo results obtained thereof. Then, the methods developed to introduce the radioelement are presented. Detailed indications on the chemical steps involved are provided, and the stability of the radiolabelling is discussed. Emphasis is then made on the techniques used to purify and analyse the radiolabelled nano-objects, a point that is rarely discussed despite its technical relevance and importance for accurate imaging. The pros and cons of the different methods developed are finally discussed from which future work can develop.

Beyond visual integration: sensitivity of the temporal-parietal junction for objects, places, and faces.

One important role of the TPJ is the contribution to perception of the global gist in hierarchically organized stimuli where individual elements create a global visual percept. However, the link between clinical findings in simultanagnosia and neuroimaging in healthy subjects is missing for real-world global stimuli, like visual scenes. It is well-known that hierarchical, global stimuli activate TPJ regions and that simultanagnosia patients show deficits during the recognition of hierarchical stimuli and real-world visual scenes. However, the role of the TPJ in real-world scene processing is entirely unexplored. In the present study, we first localized TPJ regions significantly responding to the global gist of hierarchical stimuli and then investigated the responses to visual scenes, as well as single objects and faces as control stimuli. All three stimulus classes evoked significantly positive univariate responses in the previously localized TPJ regions. In a multivariate analysis, we were able to demonstrate that voxel patterns of the TPJ were classified significantly above chance level for all three stimulus classes. These results demonstrate a significant involvement of the TPJ in processing of complex visual stimuli that is not restricted to visual scenes and that the TPJ is sensitive to different classes of visual stimuli with a specific signature of neuronal activations.

An analysis of information segregation in parallel streams of a multi-stream convolutional neural network.

Visual information is processed in hierarchically organized parallel streams in the primate brain. In the present study, information segregation in parallel streams was examined by constructing a convolutional neural network with parallel architecture in all of the convolutional layers. Although filter weights for convolution were initially set to random values, color information was segregated from shape information in most model instances after training. Deletion of the color-related stream decreased recognition accuracy of animate images, whereas deletion of the shape-related stream decreased recognition accuracy of both animate and inanimate images. The results suggest that properties of filters and functions of a stream are spontaneously segregated in parallel streams of neural networks.

Legal actors' and laypersons' utility judgments of eyewitness lineup procedures and outcomes.

OBJECTIVE: Recent attempts to model the relative performances of eyewitness lineup procedures necessarily include theoretical assumptions about the various costs/benefits, or utilities, of different identification outcomes. We collected data to incorporate empirically derived utilities into such modeling as well as data on various stakeholders' views of lineup procedures as tertiary objectives. HYPOTHESES: This research was exploratory; therefore, we did not have a priori hypotheses. METHOD: We surveyed judges' (n = 70), prosecutors' (n = 28), police officers' (n = 82), and laypersons' (n = 191) opinions about eyewitness identification procedures and the utilities of outcomes of eyewitness identification procedures. We incorporated the utility judgments into models comparing the desirability of various lineup reforms and compared policy preferences between our samples. RESULTS: All samples frequently mentioned estimator and system variables in open-ended evaluations of lineup procedures, but legal samples mentioned system variables more often than did laypersons. Reflector variables (e.g., confidence) were mentioned less often across the board, as was the scientific basis/standardization of identification policy (especially among laypersons). Utility judgments of various identification outcomes indicated that judges adopt values more closely aligned with normative legal ethics (i.e., the Blackstone ratio), whereas other stakeholders (especially laypersons) depart significantly from those standards. Utility models indicated general agreement among samples in lineup procedure preferences, which varied as a function of culprit-presence base rates. CONCLUSION: Although legal stakeholders vary in how they value eyewitness identification outcomes, their values imply relatively consistent policy preferences that sometimes depart from scientific recommendations. Nonetheless, all samples expressed support for using scientific research to inform legal policy regarding eyewitness evidence. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Plant disease recognition using residual convolutional enlightened Swin transformer networks.

Agriculture plays a pivotal role in the economic development of a nation, but, growth of agriculture is affected badly by the many factors one such is plant diseases. Early stage prediction of these disease is crucial role for global health and even for game changers the farmer's life. Recently, adoption of modern technologies, such as the Internet of Things (IoT) and deep learning concepts has given the brighter light of inventing the intelligent machines to predict the plant diseases before it is deep-rooted in the farmlands. But, precise prediction of plant diseases is a complex job due to the presence of noise, changes in the intensities, similar resemblance between healthy and diseased plants and finally dimension of plant leaves. To tackle this problem, high-accurate and intelligently tuned deep learning algorithms are mandatorily needed. In this research article, novel ensemble of Swin transformers and residual convolutional networks are proposed. Swin transformers (ST) are hierarchical structures with linearly scalable computing complexity that offer performance and flexibility at various scales. In order to extract the best deep key-point features, the Swin transformers and residual networks has been combined, followed by Feed forward networks for better prediction. Extended experimentation is conducted using Plant Village Kaggle datasets, and performance metrics, including accuracy, precision, recall, specificity, and F1-rating, are evaluated and analysed. Existing structure along with FCN-8s, CED-Net, SegNet, DeepLabv3, Dense nets, and Central nets are used to demonstrate the superiority of the suggested version. The experimental results show that in terms of accuracy, precision, recall, and F1-rating, the introduced version shown better performances than the other state-of-art hybrid learning models.

Pupil diameter as an indicator of sound pair familiarity after statistically structured auditory sequence.

Inspired by recent findings in the visual domain, we investigated whether the stimulus-evoked pupil dilation reflects temporal statistical regularities in sequences of auditory stimuli. We conducted two preregistered pupillometry experiments (experiment 1, n = 30, 21 females; experiment 2, n = 31, 22 females). In both experiments, human participants listened to sequences of spoken vowels in two conditions. In the first condition, the stimuli were presented in a random order and, in the second condition, the same stimuli were presented in a sequence structured in pairs. The second experiment replicated the first experiment with a modified timing and number of stimuli presented and without participants being informed about any sequence structure. The sound-evoked pupil dilation during a subsequent familiarity task indicated that participants learned the auditory vowel pairs of the structured condition. However, pupil diameter during the structured sequence did not differ according to the statistical regularity of the pair structure. This contrasts with similar visual studies, emphasizing the susceptibility of pupil effects during statistically structured sequences to experimental design settings in the auditory domain. In sum, our findings suggest that pupil diameter may serve as an indicator of sound pair familiarity but does not invariably respond to task-irrelevant transition probabilities of auditory sequences.

Detection and recognition of metal surface corrosion based on CBG-YOLOv5s.

The automatic detection of the degree of surface corrosion on metal structures is of significant importance for assessing structural damage and safety. To effectively identify the corrosion status on the surface of coastal metal facilities, this study proposed a CBG-YOLOv5s model for metal surface corrosion detection, based on the YOLOv5s model. Firstly, we integrated the Convolutional Block Attention Module (CBAM) into the C3 module and developed the C3CBAM module. This module effectively enhanced the channel and spatial attention capabilities of the feature map, thereby improving the feature representation. Second, we introduced a multi-scale feature fusion concept in the feature fusion part of the model and added a small target detection layer to improve small target detection. Finally, we designed a lighter C3Ghost module, which reduced the number of parameters and the computational load of the model, thereby improving the running speed of the model. In addition, to verify the effectiveness of our method, we constructed a dataset containing 6000 typical images of metal surface corrosion and conducted extensive experiments on this dataset. The results showed that compared to the YOLOv5s model and several other commonly used object detection models, our method achieved superior performance in terms of detection accuracy and speed.