Selective Fusion Research Articles

IAmyP: A Multi-view Learning for Amyloidogenic Hexapeptides Identification Based on Sequence Least Squares Programming.

The development of peptide drug is hindered by the risk of amyloidogenic aggregation; if peptides tend to aggregate in this manner, they may be unsuitable for drug design. Computational methods aimed at predicting amyloidogenic sequences often face challenges in extracting high-quality features, and their predictive performance can be enchanced. To surmount these challenges, iAmyP was introduced as a specialized computational tool designed for predicting amyloidogenic hexapeptides. Utilizing multi-view learning, iAmyP incorporated sequence, structural, and evolutionary features, performing feature selection and feature fusion through recursive feature elimination and attention mechanisms. This amalgamation of features and subsequent feature selection and fusion lead to optimal performance facilitated by an optimization algorithm based on sequence least squares programming. Notably, iAmyP exhibited robust generalization for peptides with lengths of 7-10 amino acids. The role of hydrophobic amino acids in the aggregation process is critical, and a thorough analysis have significantly enhanced our insight into their significance in amyloidogenic hexapeptides. This tool represented an advancement in the development of peptide therapeutics by providing an understanding of amyloidogenic aggregation, establishing itself as a valuable framework for assessing amyloidogenic sequences. The data and code can be freely accessed at https://github.com/xialab-ahu/iAmyP .

Selective Deposition and Fusion of AISI 316L: An Additive Manufacturing Process for Space Environments via Direct Ink Writing and Laser Processing

Abstract Unlocking the potential of additive manufacturing (AM) for space exploration hinges on overcoming key challenges, notably the ability to manufacture or repair parts on-site during exploration missions with consideration of quality, feedstock utilization, and challenges involved in microgravity environments. While there are multiple efforts to investigate the use of existing metal AM processes such as powder bed fusion (PBF), directed energy deposition (DED), and filament-based material extrusion, each process comes with a different set of challenges in space environments. Here, we introduce a new AM method that integrates the benefits of Direct Ink Writing (DIW) to selectively deposit metallic pastes with laser-based processing to locally de-bind and subsequently melt and fuse metal powder, layer by layer, enabling the manufacturing of AISI 316L samples with densities exceeding 99.0%. The impact of process parameters on single-track dimensions, surface morphology, and porosity was characterized. The efficacy of laser de-binding was assessed via secondary-ion mass spectrometry, permitting the carbon content to be estimated at 0.0152%, which is safely below the acceptable limit (0.03 wt%) for AISI 316L.

SRPNet: stroke risk prediction based on two-level feature selection and deep fusion network

BackgroundStroke is one of the major chronic non-communicable diseases (NCDs) with high morbidity, disability and mortality. The key to preventing stroke lies in controlling risk factors. However, screening risk factors and quantifying stroke risk levels remain challenging.MethodsA novel prediction model for stroke risk based on two-level feature selection and deep fusion network (SRPNet) is proposed to solve the problem mentioned above. First, the two-level feature selection method is used to screen comprehensive features related to stroke risk, enabling accurate identification of significant risk factors while eliminating redundant information. Next, the deep fusion network integrating Transformer and fully connected neural network (FCN) is utilized to establish the risk prediction model SRPNet for stroke patients.ResultsWe evaluate the performance of the SRPNet using screening data from the China Stroke Data Center (CSDC), and further validate its effectiveness with census data on stroke collected in affiliated hospital of Jining Medical University. The experimental results demonstrate that the SRPNet model selects features closely related to stroke and achieves superior risk prediction performance over benchmark methods.ConclusionsSRPNet can rapidly identify high-quality stroke risk factors, improve the accuracy of stroke prediction, and provide a powerful tool for clinical diagnosis.

Multi-scale contextual learning for medical image segmentation via dual distillation.

Recently, many studies have explored fusing features extracted from Convolutional neural networks (CNNs) and transformers to integrate multi-scale representations for better performance in medical image segmentation tasks. Although these hybrid models have achieved better results than previous CNN-based and transformer-based methods, they suffer from high computation and spacecomplexities. The purpose of this research is to address the prohibitive computation and space complexities of hybrid models, which limit their application in clinical practice where computational resources are usuallyconstrained. We propose a novel model equipped with a dual distillation scheme to sufficiently harness the complementary advantages of CNNs and transformers without compromising model efficiency. We further propose a multi-scale prior-knowledge distillation (MPD) module to effectively distill multi-scale knowledge from features extracted from transformers. In addition, to cooperate with the knowledge distillation scheme, we also propose an efficient and robust Selective Fusion module in the studentnetwork. We extensively evaluate the proposed model against fourteen different network frameworks on two representative datasets: SipakMed and ISIC 2017. In the SipakMed dataset, 3037 Pap smear images are used for training and 1012 for testing. In the ISIC 2017 dataset, 2000 dermoscopic images are used for training, 150 for validation, and 600 for testing. Experimental results demonstrate that our method not only surpasses existing methods by a considerable margin with respect to the evaluation metrics of mean Intersection over Union, mean Dice coefficient, mean average symmetric surface distance, but also requires fewer computational resources in terms of model parameters and floating-point operations persecond. Comprehensive comparisons in terms of segmentation accuracy and computational complexity unequivocally confirm that our method effectively and efficiently integrates the advantages of both CNNs and transformers, showing its suitability and significance for clinicalapplications.

RUSNet: Robust fish segmentation in underwater videos based on adaptive selection of optical flow

Fish segmentation in underwater videos can be used to accurately determine the silhouette size of fish objects, which provides key information for fish population monitoring and fishery resources survey. Some researchers have utilized underwater optical flow to improve the fish segmentation accuracy of underwater videos. However, the underwater optical flow is not evaluated and screen in existing works, and its predictions are easily disturbed by motion of non-fish. Therefore, in this paper, by analyzing underwater optical flow data, we propose a robust underwater segmentation network, RUSNet, with adaptive screening and fusion of input information. First, to enhance the robustness of the segmentation model to low-quality optical flow inputs, a global optical flow quality evaluation module is proposed for evaluating and aligning the underwater optical flow. Second, a decoder is designed by roughly localizing the fish object and then applying the proposed multidimension attention (MDA) module to iteratively recover the rough localization map from the spatial and edge dimensions of the fish. Finally, a multioutput selective fusion method is proposed in the testing stage, in which the mean absolute error (MAE) of the prediction using a single input is compared with that obtained using multisource input. Then, the information with the highest confidence is selected for predictive fusion, which facilitates the acquisition of the ultimate underwater fish segmentation results. To verify the effectiveness of the proposed model, we trained and evaluated it using a publicly available joint underwater video dataset and a separate DeepFish public dataset. Compared with the advanced underwater fish segmentation model, the proposed model has greater robustness to low-quality background optical flow in the DeepFish dataset, with the mean pixel accuracy (mPA) and mean intersection over union (mIoU) values reaching 98.77% and 97.65%, respectively. On the joint dataset, the mPA and mIoU of the proposed model are 92.61% and 90.12%, respectively, which are 0.72% and 1.21% higher than those of the advanced underwater video object segmentation model MSGNet. The results indicate that the proposed model can adaptively select the input and accurately segment fish in complex underwater scenes, which provides an effective solution for investigating fishery resources.

A multilevel cooperative attention network of precise quantitative analysis for predicting ractopamine concentration via adaptive weighted feature selection and multichannel feature fusion

Surface enhanced Raman spectroscopy (SERS) holds great potential due to its rapid detection and high sensitivity. However, issues such as signal noise, fluctuations, and spectral shifts can negatively impact its performance in detecting ractopamine in pork. Hierarchical Gradient Aware Spectral Network (HGASNet) was proposed to address these issues. The key innovations are the Spectral Gradient Weighted Attention (SGWA) and Multi-Channel Peak Attention Mechanism (MCPAM) modules within HGASNet. The SGWA module dynamically adjusts feature weights, enhancing sensitivity to critical Raman spectral shifts. Meanwhile, MCPAM leverages multi-channel attention to better capture long range dependencies and fuse global and local information. Additionally, HGASNet's hierarchical structure incrementally extracts and integrates features at various levels, enabling the model to focus on global features in the higher layers while preserving fine-grained local details in the lower layers. Experimental results show HGASNet outperforming existing approaches, achieving R2 of 0.9972, RMSRE of 0.0413, RMSLE of 0.0548, sMAPE of 6.47 %, and MAPE of 6.72 %.

Long Short Observation-driven Prediction Network for pedestrian crossing intention prediction with momentary observation

Pedestrian crossing intention prediction aims to predict whether the pedestrian will cross the road, which is crucial for the decision-making of intelligent vehicles and ensuring traffic safety. Existing methods just rely on long-term observation and rarely consider it challenging to obtain sufficiently long and precise observation in real-world scenarios. Focus on momentary observation, which only contains two frames of the preceding and current time, we propose a novel Long Short Observation-driven Prediction Network (LSOP-Net). LSOP-Net comprises two critical components, the Momentary Observation feature Extraction Module (MOE-Module) and the Multimodal Long-Short-term feature Fusion Module (MLSFusion). Utilizing a hybrid training strategy and an external long-term feature pool, the MOE-Module is proposed to extract features with long-term patterns from momentary observations, which effectively mitigates feature deficiency arising from momentary observations. Based on a feature selection fusion mechanism, the MLSFusion is proposed to explicitly model the importance relationship between various modalities’ long-short-term features and the output, which adaptively fuses the long-short-term features from various modalities. Experimental results on the JAAD and PIE datasets demonstrate that our approach achieves superior performance in pedestrian crossing intention prediction with momentary observation.

TCDDU-Net: combining transformer and convolutional dual-path decoding U-Net for retinal vessel segmentation

Accurate segmentation of retinal blood vessels is crucial for enhancing diagnostic efficiency and preventing disease progression. However, the small size and complex structure of retinal blood vessels, coupled with low contrast in corresponding fundus images, pose significant challenges for this task. We propose a novel approach for retinal vessel segmentation, which combines the transformer and convolutional dual-path decoding U-Net (TCDDU-Net). We propose the selective dense connection swin transformer block, which converts the input feature map into patches, introduces MLPs to generate probabilities, and performs selective fusion at different stages. This structure forms a dense connection framework, enabling the capture of long-distance dependencies and effective fusion of features across different stages. The subsequent stage involves the design of the background decoder, which utilizes deformable convolution to learn the background information of retinal vessels by treating them as segmentation objects. This is then combined with the foreground decoder to form a dual-path decoding U-Net. Finally, the foreground segmentation results and the processed background segmentation results are fused to obtain the final retinal vessel segmentation map. To evaluate the effectiveness of our method, we performed experiments on the DRIVE, STARE, and CHASE datasets for retinal vessel segmentation. Experimental results show that the segmentation accuracies of our algorithms are 96.98, 97.40, and 97.23, and the AUC metrics are 98.68, 98.56, and 98.50, respectively.In addition, we evaluated our methods using F1 score, specificity, and sensitivity metrics. Through a comparative analysis, we found that our proposed TCDDU-Net method effectively improves retinal vessel segmentation performance and achieves impressive results on multiple datasets compared to existing methods.

Membrane Fusion Mediated by Cationic Helical Peptide L-MMBen through Phosphatidylglycerol Recruitment.

Membrane fusion is the basis for many biological processes, which holds promise in biomedical applications including the creation of engineered hybrid cells and cell membrane functionalization. Extensive research efforts, including investigations into DNA zippers and carbon nanotubes, have been dedicated to the development of membrane fusion strategies inspired by natural SNARE proteins; nevertheless, achieving a delicate balance between membrane selectivity and high fusion efficiency through precise molecular engineering remains unclear. In our recent study, we successfully designed L-MMBen, a cationic helical antimicrobial peptide that exhibits remarkable antimicrobial efficacy while demonstrating moderate cytotoxicity. In this work, we demonstrate the effective and selective induction of fusion between phosphatidylglycerol (PG)-containing membranes by L-MMBen. By combining biophysical assays at the single-vesicle level with computer simulations at the molecular level, we discovered that L-MMBen can stably adsorb onto the surface of PG-containing membranes, leading to the formation of stalk structures between vesicles and ultimately resulting in membrane fusion. Furthermore, the occurrence of fusion is attributed to the unique ability of L-MMBen to recruit PG lipids and bridge adjacent vesicles. In contrast, its nonhelical counterpart DL-MMBen was found to lack this capability despite possessing an identical positive charge. These findings present an alternative molecule for achieving selective membrane fusion and provide insights for designing helical peptides with diverse applications.

Is It Enough to Stop Distal Fusion at L3 in Mild to Moderate Lenke 5C Adolescent Idiopathic Scoliosis Patients?

There has been no definitive conclusion on the selection of the lowest instrumented vertebra (LIV) in Lenke 5C adolescent idiopathic scoliosis (AIS) patients. The purpose of this study was to evaluate whether it is enough to stop distal fusion at L3 in mild to moderate Lenke 5C AIS patients with posterior selective lumbar fusion, Ponte osteotomies and segmental direct vertebra rotation and to analyze the risk factors for postoperative complications in patients selecting L3 as the LIV. A retrospective review was conducted on 106 Lenke 5C AIS patients who underwent corrective surgery in our institution from 2010 to 2021, with a minimum 2-year follow-up. The LIV was L3 or L4. According to the LIV, patients were initially divided into Group I (the LIV was L3) and Group II (the LIV was L4). Then, Group I was further divided into a complication group and a non-complication group. Demographics, radiological parameters, postoperative complications, and clinical outcomes were recorded. Univariate analysis and multivariate logistic analysis were used to identify the risk factors for postoperative complications in patients with L3 as the LIV. There were no significant differences in the demographics, radiological parameters, postoperative complications, or clinical outcomes between Group I and Group II (p > 0.05), and the outcomes were satisfactory in both groups. The main postoperative complications were distal adding-on (11 cases), coronal imbalance (16 cases), proximal junctional kyphosis (2 cases), and internal fixation failure (4 cases). Logistic regression analysis revealed that age and postoperative C7-CSVL were independent predictors of postoperative complications when selecting L3 as the LIV. Terminating the distal fusion level at L3 was practical for mild to moderate Lenke 5C AIS patients. For patients selecting L3 as the LIV, younger patients should be cautious, and maintaining postoperative coronal balance is necessary for avoiding postoperative complications.

Multi-Source Information-Based Bearing Fault Diagnosis Using Multi-Branch Selective Fusion Deep Residual Network.

Rolling bearing is the core component of industrial machines, but it is difficult for common single signal source-based fault diagnosis methods to ensure reliable results since sensor signals are vulnerable to the pollution of background noises and the attenuation of transmitted information. Recently, multi-source information-based fault diagnosis methods have become popular, but the information redundancy between multiple signals is a tough problem that will negatively impact the representational capacity of deep learning algorithms and the precision of fault diagnosis methods. Besides that, the characteristics of various signals are actually different, but this problem was usually omitted by researchers, and it has potential to further improve the diagnosing performance by adaptively adjusting the feature extraction process for every input signal source. Aimed at solving the above problems, a novel model for bearing fault diagnosis called multi-branch selective fusion deep residual network is proposed in this paper. The model adopts a multi-branch structure design to enable every input signal source to have a unique feature processing channel, avoiding the information of multiple signal sources blindly coupled by convolution kernels. And in each branch, different convolution kernel sizes are assigned according to the characteristics of every input signal, fully digging the precious fault components on respective information sources. Lastly, the dropout technique is used to randomly throw out some activated neurons, alleviating the redundancy and enhancing the quality of the multiscale features extracted from different signals. The proposed method was experimentally compared with other intelligent methods on two authoritative public bearing datasets, and the experimental results prove the feasibility and superiority of the proposed model.

Selective and multi-scale fusion Mamba for medical image segmentation

Given the high variability in the morphology and size of lesion areas in medical images, accurate medical image segmentation requires both precise positioning of global contours and careful processing of local boundaries. This emphasizes the importance of fusing multi-scale local and global features. However, existing CNN and Transformer-based models are often limited by high parameter counts and complex calculations, making it difficult to efficiently integrate these features. To address this challenge, we proposed two innovative optimization architectures: Selective Fusion Mamba (SF-Mamba) and Multi-Scale Fusion Mamba (MF-Mamba). SF-Mamba can flexibly and dynamically adjust the fusion strategy of local and global features according to the characteristics of the lesions, effectively handling segmentation tasks with variable morphology. MF-Mamba enhances the model’s segmentation ability for lesions of different sizes by capturing global information across scales. Based on these two structures, we constructed a lightweight SMM-UNet model, which not only significantly reduces the computational burden (with only 0.038M parameters) but also demonstrates excellent generalization ability and can efficiently adapt to various types of medical images. Extensive tests on the ISIC2017, ISIC2018, and BUSI public datasets show that SMM-UNet achieves excellent segmentation performance with an extremely low parameter cost.

Recent Global Trends and Hotspots in Occipitocervical Fusion: A Bibliometric Analysis and Visualization Study

Occipitocervical arthrodesis has a variety of indications to treat craniocervical and atlantoaxial pathologies for which a selective cervical fusion would not provide sufficient stability. Over time, the indications for occipitocervical fusions (OCF) have evolved, as new technologies and surgical techniques were developed. In this bibliometric analysis, we aim to explore the progression of OCF literature over time, analyzing the trends in publications and citations, publishing countries and authors, keywords and topics. The Web of Science database was used for data retrieval on July 3rd, 2024, with the search "occipitocervical fusion" OR "occipito-cervical fusion" OR "occipitocervical arthrodesis" OR "occipital cervical fusion" OR "occipital cervical arthrodesis" OR ("OCF" AND "spine surgery"). Excel was used to create the citation analysis and publication trend figures, along with the publishing countries and author analysis. The bibliometric software VosViewer was used to generate the keyword co-occurrence network visualizations. Overall, 762 articles were extracted. The number of pertinent publications and citations increased until 2020 before beginning to decrease. We found that Ehlers Danlos syndrome (EDS) has become a more prevalent topic, as the association between EDS and craniocervical instability has received further scrutiny. "Dysphagia" continues to be a commonly cited topic, while, conversely, rheumatoid arthritis has decreased in publication frequency, possibly related to advances in medical management and surgical techniques. Overall, the United States of America, China, and Japan are the top publishing countries. This analysis of OCF literature provides a helpful overview of emerging trends and clinician concerns, especially as seen through the perspective of time.

Excessive posterior placement of upper instrumented vertebra relative to lower instrumented vertebra as a predictor of proximal junction kyphosis after selective spinal fusion for adolescent idiopathic scoliosis Lenke type 5C curves.

We defined sagittal S-line tilt (SSLT) as the tilt of the line connecting the upper instrumented vertebra and the lower instrumented vertebra. This study aimed to: (1) examine the correlation between SSLT and proximal junctional angle (PJA) change values, and (2) determine the cut-off value of SSLT with respect to proximal junctional kyphosis (PJK) occurrence. Eighty-six consecutive patients (81 female and 5 male; mean age: 15.8years) with Lenke 5C AIS who underwent posterior selective spinal fusion. Pearson's correlation coefficients were used to examine the relationship between preoperative SSLT and changes in PJA from preoperative to 2years postoperative. The impact of SSLT on PJK at 2years after surgery was assessed using a receiver operating characteristic (ROC) curve. We observed a moderate positive correlation between preoperative SSLT and change in PJA (R = 0.541, P < 0.001). We identified 18 patients (21%) with PJK at 2years postoperative. Mean preoperative SSLT in the PJK group and the non-PJK group differed significantly at 23.3 ± 4.1° and 16.1 ± 5.0°, respectively (P < 0.001). The cut-off value of preoperative SSLT for PJK at 2years postoperative was 18° in ROC curve analysis, with a sensitivity of 94%, specificity of 68%, and area under the ROC curve of 0.868. In selective lumbar fusion for AIS Lenke type 5C curves, preoperative SSLT was significantly correlated with PJA change from preoperative to 2years postoperative. SSLT was a predictor of PJK occurrence, with a cut-off value of 18°.

Efficient Selective Image Fusion: A PCB Diagnosis Approach and Implementation

Efficient Selective Image Fusion: A PCB Diagnosis Approach and Implementation

Environment Reconstruction Based on Multi-User Selection and Multi-Modal Fusion in ISAC

Environment Reconstruction Based on Multi-User Selection and Multi-Modal Fusion in ISAC

Postoperative Coronal Imbalance in Lenke 5C Adolescent Idiopathic Scoliosis: Evolution, Risk Factors, and Clinical Implications.

To explore the changes in coronal imbalance (CIB) in Lenke 5C adolescent idiopathic scoliosis (AIS) after posterior selective fusion surgery and determine their implications for surgical decision-making. One hundred twenty patients were categorized according to the preoperative coronal pattern (type A, coronal balance distance [CBD]<20 mm; type B, CBD≥20 mm and coronal C7 plumbline [C7PL] shifted to the concave side of the curve; type C, CBD≥20 mm and C7PL shifted to the convex side of the curve). CIB group (CIB+) was defined as having a CBD≥20 mm at the 2-year follow-up. Compared to type A patients, the prevalence of postoperative CIB was higher in type C patients both immediately postoperative (22% vs. 38%, p<0.05) and at the final follow-up (5% vs. 29%, p<0.05), whereas type A patients showed a greater improvement in CBD (9 of 12 vs. 6 of 24, p<0.05) at the final follow-up. The majority of patients in all groups had recovered to type A at the final follow-up (96 of 120). The proximal Cobb-1 strategy reduced the incidence of postoperative CIB (1 of 38) at the 2-year follow-up, especially in preoperative type C patients. Multivariate logistic regression analysis revealed that type C and overcorrection of the thoracolumbar curve were risk factors for CIB at the 2-year follow-up (p=0.007 and p=0.026, respectively). Patients with type C CIB in AIS exhibited unsatisfactory restoration, with 29% of them exhibiting CIB at the final follow-up. The selective fusion strategy of proximal Cobb-1 may reduce the risk of postoperative CIB especially when the preoperative coronal pattern is type C.

The Impact of the Lowest Instrumented Vertebra on the Correction of the Minor Curve During Selective Fusion in Patients With Adolescent Idiopathic Scoliosis.

Retrospective single-center data analysis. The aim of this investigation was to give advises for choosing the LIV in selective fusion to reach the best correction of the minor curve and sagittal profile. Scoliotic curves can be classified as structural or nonstructural. If selective fusion is performed, the nonstructural curves are not instrumented. The choice of the lowest instrumented vertebra (LIV) and the impact of different levels of the LIV on the correction of the minor curve in the frontal profile and on the sagittal balance is under debate. Forty-seven consecutive patients treated by posterior instrumented fusion were included in this retrospective investigation. Impact of the level of the LIV with regard to distance to end vertebra (EV), to the stable vertebra (StV), to the sagittal infliction point (IP), and to the apex of the lumbar lordosis on the correction of the minor curve was analyzed. Distance of LIV to EV was significant with regard to correction of the minor curve if it was more than 5 levels (P<0.001). Distance of LIV to StV was significant with regard to correction of the minor curve if it was more than 4 levels (P<0.01). Distance of LIV to IP was significant with regard to correction of the minor curve if it was more than 2 levels (P<0.01). Choosing a LIV that was more than 2 levels higher or lower than the sagittal infliction point showed a significantly higher correction of the minor curve. We therefore recommend to keep that distance when LIV is chosen.

GLFuse: A Global and Local Four-Branch Feature Extraction Network for Infrared and Visible Image Fusion

Infrared and visible image fusion integrates complementary information from different modalities into a single image, providing sufficient imaging information for scene interpretation and downstream target recognition tasks. However, existing fusion methods often focus only on highlighting salient targets or preserving scene details, failing to effectively combine entire features from different modalities during the fusion process, resulting in underutilized features and poor overall fusion effects. To address these challenges, a global and local four-branch feature extraction image fusion network (GLFuse) is proposed. On one hand, the Super Token Transformer (STT) block, which is capable of rapidly sampling and predicting super tokens, is utilized to capture global features in the scene. On the other hand, a Detail Extraction Block (DEB) is developed to extract local features in the scene. Additionally, two feature fusion modules, namely the Attention-based Feature Selection Fusion Module (ASFM) and the Dual Attention Fusion Module (DAFM), are designed to facilitate selective fusion of features from different modalities. Of more importance, the various perceptual information of feature maps learned from different modality images at the different layers of a network is investigated to design a perceptual loss function to better restore scene detail information and highlight salient targets by treating the perceptual information separately. Extensive experiments confirm that GLFuse exhibits excellent performance in both subjective and objective evaluations. It deserves note that GLFuse effectively improves downstream target detection performance on a unified benchmark.

Early diagnosis of Cladosporium fulvum in greenhouse tomato plants based on visible/near-infrared (VIS/NIR) and near-infrared (NIR) data fusion

Plant diseases can inflict varying degrees of damage on agricultural production. Therefore, identifying a rapid, non-destructive early diagnostic method is crucial for safeguarding plants. Cladosporium fulvum (C. fulvum) is one of the major diseases in tomato growth. This work presents a method of data fusion using two hyperspectral imaging systems of visible/near-infrared (VIS/NIR) and near-infrared (NIR) spectroscopy for the early diagnosis of C. fulvum in greenhouse tomatoes. First, hyperspectral images of samples at health and different times of infection were collected. The average spectral data of the image regions of interest were extracted and preprocessed for subsequent spectral datasets. Then different classification models were established for VIS/NIR and NIR data, optimized through various variable selection and data fusion methods. The principal component analysis-radial basis function neural network (PCA-RBF) model established using low-level data fusion achieved optimal results, achieving accuracies of 100% and 99.3% for calibration and prediction, respectively. Moreover, both the macro-averaged F1 (Macro-F1) values reached 1, and the geometric mean (G-mean) values reached 1 and 1, respectively. The results indicated that it was feasible to establish a PCA-RBF model by using the hyperspectral technique with low-level data fusion for the early detection of C. fulvum in greenhouse tomatoes.