Search Results (4,309)

Acromegaly is a rare chronic endocrine disorder characterized by delayed diagnosis, multisystem comorbidity, and heterogeneous therapeutic response. We aimed to describe the clinical characteristics, tumor profile, treatment patterns, biochemical control, pituitary insufficiencies, and comorbidity burden in an endocrinology tertiary center in northeastern Romania. This observational retrospective study included 87 adult patients admitted for general inpatient evaluation between December 2023 and November 2024, with retrospective data collected from diagnosis and follow-up assessed through the last available hospital visit at St. Spiridon Clinical Emergency Hospital. Clinical, hormonal, imaging, and therapeutic data were analyzed using descriptive statistics and inferential statistical tests. Most patients were diagnosed in middle adulthood, with a female predominance. Macroadenomas and extrasellar extension were common, consistent with advanced tumor stage at presentation. Treatment was predominantly multimodal, with surgery as the main therapeutic intervention and somatostatin receptor ligands as the main medical treatment backbone. Biochemical improvement was observed over time, although complete remission was achieved only in a subset of patients. These findings describe the clinical and therapeutic complexity of acromegaly in a single tertiary-center inpatient cohort and support the need for individualized long-term monitoring. Full article

Accurate, automated delineation of adult diffuse gliomas from multi-parametric magnetic resonance imaging (mpMRI) is central to quantitative neuro-oncology. Volumetric 3D networks dominate the BraTS leaderboard but require expensive GPUs, long training cycles, and provide diminishing returns relative to their compute budget. Slice-wise 2D models, by contrast, discard inter-slice context that is informative for thin tumor rims and small enhancing foci. We introduce GDNet, a 2.5D multimodal MRI segmentation framework for adult glioma evaluated on the BraTS 2024 cohort. GDNet consumes a stack of three adjacent axial slices from the four standard BraTS modalities (T1, T1ce, T2, FLAIR) as a 12-channel input to a compact U-shaped encoder–decoder with Group Normalization and predicts whole tumor (WT), tumor core (TC), and enhancing tumor (ET) masks for the central slice. The training pipeline pairs the 2.5D backbone with: (i) Exponential Moving Average (EMA) of model weights with decay 0.999, (ii) mixed tumor-aware slice sampling (p_tumor = 0.50), (iii) a compound Cross-Entropy + Soft-Dice loss, and (iv) AdamW with warm-up plus cosine annealing under Automatic Mixed Precision. We performed a systematic, step-by-step ablation covering a 2D baseline, EMA + mixed sampling, tumor-centered crop fine-tuning, a GDNet-inspired architectural integration, a region-aware loss, 3-slice and 5-slice 2.5D inputs, and connected-component post-processing, and we report multi-seed results to quantify reproducibility. On the held-out BraTS 2024 test partition, the final 3-slice 2.5D GDNet achieved positive-only Dice scores of 0.791 ± 0.000 (WT), 0.736 ± 0.003 (TC), 0.654 ± 0.004 (ET), and a mean foreground positive-only Dice of 0.820 ± 0.000 across seeds; the all-slice mean foreground Dice exceeded 0.927 ± 0.000. Validation positive-only scores were 0.805 ± 0.002 (WT), 0.757 ± 0.004 (TC), 0.683 ± 0.009 (ET). The inter-seed standard deviation was small for every region (≤0.01 Dice points), indicating low inter-seed variance across the two seeds evaluated; with only two seeds, we regard this as preliminary evidence of training stability rather than a strong reproducibility claim. The ablation isolated EMA + mixed tumor sampling and the 2.5D context window as the dominant sources of improvement; notably, a GDNet-style architectural integration with a region-aware loss did not outperform the simpler 2.5D U-Net on positive-only WT/TC/ET, and light post-processing improved only all-slice Dice. A failure-mode audit found that the residual catastrophic predictions are concentrated on a small minority of diffuse, infiltrative tumors with mass effect. Conclusions: Carefully engineered training strategies, tumor-aware sampling, EMA stabilization, and a modest 2.5D context window recover a substantial fraction of the accuracy of much heavier 3D networks at a fraction of the compute, are reproducible across seeds, and outperform a heavier GDNet-inspired architectural variant on the same data. GDNet is therefore a practical and, pending external validation, potentially clinically deployable framework for multimodal glioma segmentation on workstation-class GPU hardware. Full article

Background: Cutaneous squamous cell carcinoma (cSCC) is commonly regarded as a sporadic malignancy primarily driven by ultraviolet exposure. However, the occurrence of multiple, aggressive tumors at a relatively young age suggests the presence of underlying genetic susceptibility. The role of germline variants affecting extracellular matrix organization, pigmentation pathways, and tumor metabolism in aggressive cSCC remains incompletely understood. Case Presentation: We describe a 53-year-old patient with a long-standing history of multiple aggressive cutaneous squamous cell carcinomas involving the scalp and facial regions, characterized by recurrent and multifocal disease. A comprehensive diagnostic approach was undertaken, including histopathological examination, fluorescence confocal microscopy, high-frequency cutaneous ultrasound, and genetic analysis using whole-exome sequencing (WES). Results: Histopathology confirmed high-risk features consistent with aggressive cSCC. Cutaneous ultrasound and fluorescence confocal microscopy provided complementary, non-invasive insights into tumor depth, architecture, and invasive patterns. Whole-exome sequencing identified a heterozygous truncating variant in COL6A3 (NM_004369.4:c.5645C>A, p.Ser1882Ter), classified as likely pathogenic according to ACMG criteria. Additionally, two heterozygous variants of uncertain significance were detected in TYR (NM_000372.5:c.1569C>A, p.Ser523Arg) and FH (NM_000143.4:c.1237-5_1237-4insTCTCCCTCCCTC). Although individually inconclusive, the combined germline genetic background may have contributed to the patient’s aggressive and multifocal cutaneous phenotype. Discussion: This case report supports a potential role of extracellular matrix remodeling, pigmentation-related susceptibility, and metabolic dysregulation in cutaneous carcinogenesis and tumor aggressiveness. This case illustrates how integrating WES with advanced non-invasive imaging techniques can enhance the understanding of biologically aggressive cSCC. Conclusions: This report highlights a unique case of multifocal aggressive cSCC characterized by a distinct germline genetic profile identified by WES and multimodal imaging assessment. Comprehensive molecular and imaging evaluation may be beneficial in selected patients with atypical or aggressive cutaneous squamous cell carcinoma, with implications for personalized surveillance and management. Full article

Background: Lung cancer remains the leading cause of cancer-related mortality worldwide, and early diagnosis and accurate disease stratification are still major clinical challenges. Radiomics has emerged as a quantitative imaging approach that extracts high-dimensional features from radiological imaging, with applications in diagnosis, prognosis, radio genomics, and assessment of treatment response. However, its clinical translation is still limited by methodological heterogeneity and a lack of standardization. Aim: This narrative review synthesizes evidence from systematic reviews and meta-analyses on radiomics in thoracic imaging for lung cancer, focusing on clinical applications, methodological limitations, and translational challenges. Methods: A structured search was conducted in PubMed and Scopus using predefined keywords related to radiomics, lung cancer, and imaging modalities. Only peer-reviewed systematic reviews and meta-analyses published in English were included. In total, 27 studies were selected and synthesized using a structured narrative approach guided by the ANDJ checklist. A differential integrative framework was adopted to connect evidence from systematic reviews and meta-analyses with primary empirical studies and policy documents through an intermediate layer of translational recommendations, ensuring a multi-level and interpretation-driven synthesis. Results: Radiomics demonstrated consistent potential across multiple clinical domains, including lesion classification, histological differentiation, molecular profiling, prognostic stratification, and prediction of treatment response. Machine learning and deep learning approaches frequently improved predictive performance. However, key limitations were identified, including heterogeneity in imaging protocols, lack of external validation, small single-centre datasets, and limited reproducibility of radiomic features. Conclusions: Radiomics in lung cancer imaging shows strong clinical potential but remains constrained by methodological and translational barriers. Future progress will depend on standardization, external validation, multimodal data integration, and improved interpretability, alongside alignment with regulatory and clinical implementation frameworks. Full article

Breast cancer remains one of the most significant challenges in modern oncology, while advances in artificial intelligence (AI) are creating new opportunities to improve diagnosis, prognosis, and treatment personalization. The aim of this review was to summarize current and emerging applications of AI in the comprehensive care of patients with breast cancer. This study was conducted as a structured narrative review with elements of integrative evidence synthesis based on publications retrieved from PubMed/MEDLINE, Scopus, Web of Science, and Embase. The review included studies evaluating machine learning and deep learning approaches, such as support vector machines, random forests, convolutional neural networks, Vision Transformers, foundation models, self-supervised learning, federated learning, and multimodal AI systems. The strongest clinical evidence currently concerns AI-supported mammographic screening, where large prospective and real-world studies suggest improvements in cancer detection and workflow efficiency. Applications involving MRI, ultrasound, histopathology, molecular prediction, treatment-response assessment, and treatment selection have shown promising performance, but most remain investigational because of limited prospective multicenter validation. Emerging approaches integrating imaging, pathological, molecular, and clinical data show considerable potential for precision oncology. AI may also support treatment selection, patient monitoring, and survivorship care. Despite promising results, widespread clinical implementation remains limited by challenges related to data heterogeneity, model interpretability, external validation, and integration into clinical workflows. Further prospective multicenter studies are required to establish the safety, reliability, and clinical utility of AI-driven systems in breast cancer care. Full article

Background/Objectives: Androgenetic alopecia (AGA) responds only partially to pharmacologic monotherapy. Combination procedural regimens incorporating platelet-rich plasma (PRP), stromal vascular fraction (SVF), and botulinum toxin (BTX) have been reported, but objective quantitative trichoscopic data on multimodal single-session protocols are limited. We retrospectively quantified the trichoscopic response to a four-component single-session scalp procedure used in routine clinical practice. Methods: Fifty-one consecutive AGA patients underwent a single-session procedure combining partial temporalis muscle resection with silicone implantation, negative-pressure scalp stimulation, and BTX, PRP, and SVF injections; 28 completed ≥ 4-month follow-up. Standardized 60× videodermoscopy at five predefined scalp locations was archived for paired quantitative analysis in six patients (30 location pairs, of which 28 were analyzable after excluding two pairs for motion artifact), with one additional patient imaged at four years. Six trichoscopic outcomes were derived by automated image analysis (Otsu thresholding, skeletonization, distance-transform shaft thickness); the primary analysis was performed at the patient level (n = 6) and a supporting analysis at the panel level (n = 28), each using paired Student’s t-tests. Results: In the primary patient-level analysis (n = 6 patients), five of six trichoscopic outcomes improved significantly at 3–4-month follow-up, each with a large effect size: median shaft thickness +54% (p = 0.025), terminal-hair proportion +52% (p = 0.028), vellus-hair proportion −33% (p = 0.011), diameter heterogeneity −14% (p = 0.017), and mean shaft thickness +33% (p = 0.029); hair coverage increased but did not reach statistical significance (+11%, p = 0.125). The supporting panel-level analysis (n = 28 paired panels) was concordant in direction and significant for all six metrics. In a single illustrative case followed for four years (n = 1; exploratory), mean shaft thickness gain (+41%, p = 0.039) and vellus reduction (−36%, p = 0.025) were sustained, while the transient coverage gain at 3–4 months (+38%, p = 0.007) partially receded. Conclusions: In this preliminary case series, the integrative procedure was associated with quantifiable trichoscopic re-thickening rather than gross densification, with sustained shaft-caliber gain at four years in the long-term case. Causal attribution to any single component is not possible from this single-arm design; prospective controlled trials are required. Full article

Railway scene understanding is critical for ensuring train operational safety and advancing intelligent railway systems. Existing railway vision methods mainly focus on perception and classification, while lacking regulation-guided semantic reasoning capabilities in complex environments. To address these limitations, this paper proposes a retrieval-augmented generation (RAG)-enhanced vision–language framework for railway signal cognition and safety reasoning. The proposed method integrates railway signal perception, regulatory knowledge retrieval, and multi-modal reasoning to improve factual consistency, reasoning reliability, and operational interpretability. In addition, a dedicated railway signal dataset comprising 500 standardized railway scene images with structured QA annotations is constructed to support regulation-oriented multi-modal recognition evaluation. Experimental results show that the proposed framework improves reasoning accuracy from 28.40% to 67.20% with an average end-to-end inference latency of 11.31 s per sample, and the inference speed can be further improved by adjusting experimental configurations to trade off between efficiency and accuracy, demonstrating the potential of RAG-enhanced architectures as a foundational step toward reliable multi-modal cognition in intelligent railway systems. Full article

Epidermolysis bullosa (EB) comprises a group of rare inherited genodermatoses characterized by fragility and blistering of the skin and mucous membranes, chronic wounding, and significant morbidity including increased risk of squamous cell carcinoma in severe subtypes. Key unmet priorities include reducing diagnostic latency, establishing objective wound monitoring, enabling early detection of malignant transformation within chronic ulcerations, and developing therapies that durably modify disease progression. Artificial intelligence (AI) encompassing machine learning (ML), and deep learning (DL) is increasingly integrated into EB research and clinical practice to address these unmet needs. This structured narrative review synthesises current evidence on AI applications in EB spanning genetic diagnostics, wound assessment, inflammatory endotyping, drug repurposing, and emerging therapeutic technologies, and integrates evidence from registered clinical trials. In genomics, DL-based splicing prediction models and variant prioritisation frameworks accelerate pathogenic variant detection and reduce diagnostic latency. In wound care, convolutional neural networks-based platforms enable automated lesion segmentation and remote monitoring, while multimodal AI models predict healing trajectories and support stratification of wounds by chronicity. Computational transcriptomic analyses have identified candidate repurposing agents by reversing pathogenic gene expression signatures in EB tissue. Emerging convergence of AI with biosensors-integrated wound dressings and three-dimensional bioprinting of genetically corrected skin substitutes represents a transformative future direction. Translational barriers include limited EB-specific training datasets, algorithmic bias across diverse skin phototypes, the interpretability deficit of DL systems, and evolving regulatory frameworks for AI as a medical device. Expansion of internationally interoperable EB disease registries with standardised wound imaging protocols is identified as the single most impactful intervention to accelerate AI adoption. A minimum endpoint set for AI-assisted EB wound assessment, incorporating wound area trajectory, wound type classification, tissue composition, and paired patient-reported pain and itch scores, is proposed to standardise outcome reporting across future studies. Full article

The worldwide spread of semiconductor devices has driven a surge in electronic waste (e-waste), which reached 62 million metric tons in 2022 and is projected to exceed 80 million metric tons by 2030. E-waste contains hazardous substances such as cadmium and mercury, yet also represents a $57 billion annual opportunity through the recovery of valuable and critical raw materials (CRMs). However, formal recycling rates remain stagnant at 22.3%, largely due to limitations of current automated sorting methods. These systems primarily rely on visible-light (RGB) imaging, which lacks the spectral resolution needed to distinguish chemically similar polymers, complex metal alloys, and composite substrates on printed circuit boards (PCBs). This paper presents a multidisciplinary synthesis of AI-driven detection and classification for e-waste, bridging materials science and computer vision through three interconnected themes. 1. Material and Economic Context: The toxicological risks and economic drivers of semiconductor recycling are characterized, framing fine-grained material identification as essential for a circular economy. 2. Multispectral Sensing & Fusion: Sensing modalities such as near-infrared (NIR), hyperspectral imaging (HSI), and X-ray fluorescence (XRF) are assessed, and sensor fusion strategies, including early, late, and intermediate fusion, are reviewed for high-throughput industrial settings. 3. Deep Learning Benchmarking: 11 publicly available PCB datasets are analyzed, and the YOLO series (YOLOv3–YOLOv12) is compared with leading non-YOLO detectors, including Faster R-CNN, RT-DETR-L, and RetinaNet. The results show that while YOLOv9s achieves a peak mAP@0.5 of 56.5% and YOLOv11s offers an optimal industrial profile (37.2% mAP@0.5:0.95 at 115 ms edge inference), all RGB-based models fail to detect visually ambiguous surface-mount devices (SMDs), with mAP values below 12%. This confirms a performance ceiling for purely visual systems. The review concludes that transitioning from RGB-centric to multispectral fusion architectures is the primary research frontier and proposes a roadmap for standardized multimodal datasets and edge-deployable fusion models to enable next-generation, high-recovery automated recycling. Full article

Chronic epididymitis and orchitis represent significant yet frequently under-recognized contributors to male infertility, particularly among men of reproductive age. These conditions arise from persistent inflammatory or immunological processes affecting the epididymis and testis, leading to impaired spermatogenesis, altered sperm maturation and possible obstruction of the male reproductive tract. Infectious aetiologies, especially those linked to sexually transmitted pathogens and uropathogens, remain predominant; however, non-infectious mechanisms, including autoimmune activation, post-vasectomy changes and idiopathic inflammation, also play critical roles. The persistent inflammatory milieu induces cytokine release, oxidative stress and structural tissue remodelling, ultimately compromising the functional and immune-privileged microenvironment necessary for optimal sperm production and transport. Diagnostic evaluation requires a multimodal approach incorporating clinical examination, microbiological testing, semen analysis and scrotal ultrasonography, with advanced imaging and molecular assays reserved for complex or equivocal cases. Management is individualized and may involve antimicrobial therapy, anti-inflammatory treatment, immunomodulation or microsurgical intervention in cases of ductal obstruction. Assisted reproductive technologies provide additional options when natural conception is not feasible. Despite increased recognition of their impact, chronic epididymitis and orchitis remain insufficiently studied, with gaps in standardized definitions, biomarker validation and long-term outcome data. This review provides a focused synthesis and phenotype-driven clinical framework for chronic epididymitis and orchitis through a fertility-preservation lens, bridging urological and andrological perspectives and integrating evidence on subclinical inflammation, contemporary diagnostic biomarkers and a staged therapeutic pathway. Full article

Urban renewal increasingly adopts Nature-Based Solutions (NBSs) to address environmental challenges and enhance social well-being. However, it remains unclear whether and to what extent NBSs contribute to cultural ecosystem services (CESs), which reflect people’s perceptions, values, and experiences of urban nature. This study develops an integrated framework combining text and image mining of social media data to evaluate the CES outcomes of NBS in regenerated urban districts in Chengdu, China. The comment data were analyzed for CES using Jieba word segmentation and dictionary matching, while images were categorized into NBS types by manual classification. By integrating these multimodal data, the framework effectively clarifies the relationship between NBSs and CESs from the perspective of public perception. Results indicate that recreation and leisure, inspiration, and spiritual values are the most prominent aspects of public perception, with linear green infrastructure and pocket parks being the most frequently identified NBS types. Correspondence analysis further reveals significant associations between specific NBS interventions and CES categories. By integrating textual and visual data, this study offers a practical and real-time approach for capturing public perceptions of CESs and provides actionable insights for the design and management of NBS-driven urban regeneration. Full article

Background/Objectives: The utility of optical coherence tomography angiography (OCTA) for neovascular age-related macular degeneration (nAMD) remains unclear. The current study investigated the choroidal neovascularization (CNV) detection rate by OCTA in comparison with standard fluorescein angiography (FA) and spectral-domain optical coherence tomography (SD-OCT). Methods: Subjects underwent multimodal imaging, including FA, SD-OCT, and OCTA imaging, which were compared. In patients with unilateral nAMD, the contralateral eye with dry AMD (n = 39) was included to determine imaging modality sensitivity and specificity. Eyes with inaccurate automated segmentation from retinal distortion were manually resegmented. Results: The diagnostic performance for nAMD was 86% sensitivity and 100% specificity by OCT (AUC: 0.93; 95% CI 0.87–0.99; p < 0.001); 82% sensitivity and 100% specificity by FA (AUC: 0.91; 95% CI 0.84–0.98; p < 0.001); and 68% sensitivity and 100% specificity by automatically segmented OCTA (AUC: 0.84; 95% CI 0.76–0.93; p < 0.001). OCTA diagnostic accuracy improved following manual resegmentation to 88% sensitivity and 100% specificity (AUC: 0.94; 95% CI 0.89–1.0; p < 0.001). Diagnostic accuracy of OCT combined with manually resegmented OCTA (AUC: 1.0; 95% CI 1.0–1.0; p < 0.001) was greater than that of OCT or FA combined (AUC: 0.96; 95% CI 0.92–1.0; p < 0.001) but both were very accurate. Conclusions: Manual segmentation of the OCTA images can help identify CNV in eyes otherwise undetected by automated segmentation algorithms due to errors in segmentation of retinal layers. Eyes with substantial elevation in one or more layers of the retina were most likely to benefit from resegmentation. Full article

Background/Objectives: Novel vision-language models (VLMs) can integrate patient textual data with image data to support medical diagnosis. Recent studies reported conflicting results regarding the performance of multimodal VLMs compared to other models and physician performance. This systematic review aims to assess the diagnostic performance of multimodal VLMs integrating both patient textual and image data across diverse real-world hospital settings. Methods: We performed comprehensive searches of eight resources, including Embase, MEDLINE, and SCOPUS, on 17 December 2025. Eligible studies reporting diagnostic performance of VLMs integrating both image and patient history textual data from real-world adult patients compared to that of other models and physicians were included. The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The Prediction model study Risk Of Bias Assessment Tool + AI (PROBAST + AI) was used to assess the quality and risk of bias. The study protocol was registered in the PROSPERO database (CRD420251244054). This review received no external funding. Results: We screened 11,026 records, of which 18 studies met the inclusion criteria. Six studies comparing multimodal and unimodal models demonstrated the consistent superiority of the multimodal models. Four studies evaluating VLM accuracy as standalone agents compared with physician performance reported conflicting evidence. One study assessing VLMs as a clinical copilot demonstrated higher accuracy from the group of physicians using VLM assistance. A meta-analysis could not be performed due to the heterogeneity across study populations and outcomes. The majority of the studies were assessed as having a high risk of bias due to dataset quality. Primary limitations identified across studies include small sample size, a lack of external validation, and the need for prospective clinical deployment studies. No study provided documented considerations regarding model safety or data security. Conclusions: This systematic review suggests that multimodal VLMs consistently outperform unimodal models with access to only image or text. While model performance as standalone agents compared to humans remains inconclusive, a copilot model has demonstrated high diagnostic accuracy. Given substantial methodological concerns across studies, cautious interpretation is required, No firm clinical recommendation can be made regarding the use of standalone VLMs. Further research employing high-quality datasets is needed to ensure the reliability and clinical applicability of future VLMs. Full article

There is an enormous rise in the amount of user-generated content on social media. That makes it easier for hateful and fake messages to spread, and threatens both societal stability and public trust in institutions. Most of current solutions have fundamental limitations due to modal limitations (i.e., each solution only uses one type of data at a time), lack of user context integration, poor synchronization across different types of data, and poor resilience to manipulation by adversaries. As a result, most solutions are subject to compound loss in terms of their ability to generalize well, classify correctly, or remain reliable when deployed in real-world environments. To address all of the above challenges, we propose a comprehensive and modular analytical framework consisting of five interconnected components that integrate contextual representation learning, multimodal semantic alignment, graph-based propagation modeling, adaptive inference, and consistency validation for hate speech and fake post detection. First is our Context-Driven Social Vector Extraction methodology, which provides enriched contextual embeddings by extracting and combining text-based metadata, image-based metadata, temporal metadata, and behavioral metadata. We use those embeddings in our second module, Multimodal Label Fusion via Mutual Co-Attention (CMF-MCA). Our CMF-MCA module incorporates two transformers with co-attention mechanisms that can mutually annotate text and images. In our third methodology, Semantic Propagation Graph for Hate and Fake Correlation (SPG-HFC), we implement a relational graph attention mechanism that captures both the influence of semantics and how communities propagate information about hate and fake posts. The fourth module, Adaptive Modality Routing via Reinforcement (AMR-R), routes based on the modality of the input and whether the input is simple enough to be classified using machine learning or complex enough to require deep learning. Finally, our Counterfactual Consistency Validation Engine (CCVE) is used after prediction to validate that the model’s predictions are consistent with the output data by creating counterfactuals and validating them. Therefore, in addition to improving the overall accuracy of hate speech and fake post detections, our proposed framework also improves its scalability and inference reliability. Additionally, because our framework allows multimodal classifications that include both context and behavior, it enables the scalable and trustworthy development of content moderation systems. Full article

Solar insecticidal lamps are important physical control devices for green pest management, but faults in their photovoltaic power supply units can reduce trapping efficiency and shorten service life. To improve fault identification under complex agricultural environments, this study proposes a signal-image-level multimodal fusion network (SIL-MMFN) for detecting and classifying photovoltaic panel operating states in solar insecticidal lamps. The method combines time-series measurements with short-time Fourier transform (STFT)-based time–frequency images. A convolutional image branch extracts spatial features from time–frequency representations, whereas a bidirectional GRU branch with attention models temporal dependencies in the original signals. In addition, physics-informed features based on the illumination–current residual and output power are introduced to enhance discriminative fault information. Field data collected from four agricultural deployment nodes were used to classify normal, open-circuit, and mismatch states. Experimental results show that the proposed method achieved an accuracy of 97.5%, precision of 96.7%, recall of 97.8%, and macro-F1 score of 97.3%, outperforming single-modality and representative comparison models. The results indicate that multimodal fusion helps reduce confusion between open-circuit and mismatch faults and provides a potential approach for operating-state monitoring and maintenance of agricultural photovoltaic equipment. In this study, fault diagnosis refers to the detection and classification of photovoltaic panel operating states, including normal, open-circuit, and mismatch conditions. Full article