Mortality and prognostic factors among inpatients with systemic lupus erythematosus in China: A 20-year retrospective study.

OBJECTIVE: To summarize the causes of death and clinical characteristics of systemic lupus erythematosus (SLE) hospitalized patients in the last 20 years to improve SLE survival rates by detecting critical SLE early. METHODS: In this case-control study, 218 SLE death cases were retrospectively analyzed from January 2002 to December 2022, with 110 SLE inpatients chosen at random as controls. The clinical symptoms, causes of death, and risk factors in patients with SLE were investigated. RESULTS: There were 218 deaths among 9538 patients with SLE, including 188 women and 30 men. The death rate fell steadily from 4.14% in 2002 to 1.96% in 2013 and remained at 1.84% from 2014 to 2022. The standardized mortality ratio (SMR) was 4.98 [95% CI (4.06-5.89)] from 2002 to 2012 and 3.39 [95% CI (2.74-4.04)] from 2013 to 2022. Infection, lupus-induced multiple organ failure syndrome (MODS), and neuropsychiatric lupus (NPLE) were the leading causes of death, accounting for 31.19%, 15.14%, and 11.47% of overall deaths. Age had a significant association with the major causes of death. Logistic regression analysis showed NPLE[OR = 10.772,95% CI (3.350,34.633), p < 0.001], lupus pulmonary involvement (LP)[OR = 3.844,95%CI (1.547,9.552), p = 0.004], pneumonia[OR = 3.439,95%CI(1.552,7.621), p = 0.002], thrombocytopenia[OR = 14.941,95%CI (4.088,54.604), p < 0.001], creatinine>177 µmol/L[OR = 8.644,95%CI (2.831,26.388), p < 0.001], glutamic transaminase(AST) > 60U/L[OR = 5.762,95%CI (2.200,15.088), p < 0.001], total bilirubin > 34 µmol/L[OR = 16.701,95%CI (3.349,83.294), p = 0.001], higher SLE Disease Activity Index (SLEDAI)[OR = 1.089,95%CI (1.032,1.149), p = 0.002] and SLE Damage Index (SDI)[OR = 3.690,95%CI (2.487,5.474), p < 0.001] correlated positively with death. CONCLUSION: From 2002 to 2013, the mortality rate among patients with SLE fell steadily but remained unchanged from 2014 to 2022. Patients with SLE had significantly higher SMR than the general population. Childhood-onset SLE had a poorer prognosis than adult-onset SLE. Infection, MODS, and NPLE were the three leading causes of death. Major organ involvement and high disease activity were risk factors for mortality.

Bisphosphonate- and disumab-related gingival disorders: case analysis from the U.S. Food and Drug Administration Adverse Event Reporting System.

Prior research has indicated that bisphosphonates (BPs) can improve periodontal disease because of their anti-osteoporosis properties. In vitro studies have shown that BPs induce cytotoxicity, inhibit wound healing, and thus affect periodontal disease. Denosumab and BPs have alternative indications. BP and denosumab are not known to correlate with gingival disorders. We assessed such a relationship by applying Bayesian and nonproportional analyses to data in the US FDA Adverse Event Reporting System (FAERS) database. The study analyzed BPs and denosumab-reported incidents with preferred terms found in the narrow Standardized MedDRA Queries for gingival disorders. A total of 5863 reported cases of gingival disorders were associated with five BPs (alendronate, pamidronate, ibandronate, risedronate, and zoledronate) and denosumab. More than 15% of patients with gingival disorders related to BPs and denosumab other than denosumab were hospitalized over short- or long-term periods. Our findings indicated BPs and denosumab had significant reporting odds ratios (ROR), proportional reporting ratios (PRR), and information components (IC) with respect to gingival disorders. Pamidronate had the highest association (ROR = 64.58, PRR = 57.99, IC = 5.71), while the weakest association was found with denosumab (ROR = 3.61, PRR = 3.60, IC = 1.77). Significant associations were found between the six drugs and gingival pain, gingival recession, gingivitis, periodontal disease, and periodontitis. In conclusion, our comprehensive overview of the correlations, clinical characteristics, and prognoses of BPs and denosumab-related gingival disorders suggests that these issues deserve continued surveillance and appropriate management.

IL-1ß promotes adipogenesis by directly targeting adipocyte precursors.

Postprandial IL-1ß surges are predominant in the white adipose tissue (WAT), but its consequences are unknown. Here, we investigate the role of IL-1ß in WAT energy storage and show that adipocyte-specific deletion of IL-1 receptor 1 (IL1R1) has no metabolic consequences, whereas ubiquitous lack of IL1R1 reduces body weight, WAT mass, and adipocyte formation in mice. Among all major WAT-resident cell types, progenitors express the highest IL1R1 levels. In vitro, IL-1ß potently promotes adipogenesis in murine and human adipose-derived stem cells. This effect is exclusive to early-differentiation-stage cells, in which the adipogenic transcription factors C/EBPδ and C/EBPß are rapidly upregulated by IL-1ß and enriched near important adipogenic genes. The pro-adipogenic, but not pro-inflammatory effect of IL-1ß is potentiated by acute treatment and blocked by chronic exposure. Thus, we propose that transient postprandial IL-1ß surges regulate WAT remodeling by promoting adipogenesis, whereas chronically elevated IL-1ß levels in obesity blunts this physiological function.

Amide naphthotube as a novel supramolecular sequestration agent for tetracaine and decamethonium.

RATIONALE: Anesthetics are widely used for optimizing surgical conditions, postoperative pain management, and treating various chronic pain conditions. Tetracaine and decamethonium are representative drugs of local anesthetics and neuromuscular blocking agents, respectively. However, overdose and toxicity of the drugs always lead to serious adverse events. Thus, there is a strong demand for effective antidotes. METHODS: The binding interactions of amide naphthotubes with tetracaine and decamethonium were systematically studied using 1H NMR, ITC, and DFT calculations. The antidotal effects of amide naphthotube to tetracaine toxicity were assessed in vitro and in vivo, and the mechanism of detoxification was explored at a cellular level. Additionally, mouse models were established to evaluate the reversal activities of amide naphthotube on decamethonium-induced mortality and muscle relaxation, and the reversal mechanism was investigated through pharmacokinetic experiments. RESULTS: We have demonstrated that the anti-isomer of amide naphthotube exhibits significant binding affinities towards tetracaine (K a = 1.89×107 M-1) and decamethonium (K a = 1.01×107 M-1) in water. The host displayed good biocompatibility both in vitro and in vivo. The administration of amide naphthotube following tetracaine overdose in mouse models notably increased the overall survival rate, indicating its effective antidotal properties. The host could reverse the tetracaine-induced Na+ channels blockage at the cellular level. Moreover, the injection of amide naphthotube also reversed the mortality and paralysis induced by decamethonium in mouse models following a pharmacokinetic mechanism. CONCLUSION: An emerging artificial receptor, amide naphthotube, has strong binding affinities towards tetracaine and decamethonium. It functions as a supramolecular antidote for tetracaine poisoning and a reversal agent for decamethonium by selectively sequestering these compounds in vivo.

Knowledge-aware Multisite Adaptive Graph Transformer for Brain Disorder Diagnosis.

Brain disorder diagnosis via resting-state functional magnetic resonance imaging (rs-fMRI) is usually limited due to the complex imaging features and sample size. For brain disorder diagnosis, the graph convolutional network (GCN) has achieved remarkable success by capturing interactions between individuals and the population. However, there are mainly three limitations: 1) The previous GCN approaches consider the non-imaging information in edge construction but ignore the sensitivity differences of features to non-imaging information. 2) The previous GCN approaches solely focus on establishing interactions between subjects (i.e., individuals and the population), disregarding the essential relationship between features. 3) Multisite data increase the sample size to help classifier training, but the inter-site heterogeneity limits the performance to some extent. This paper proposes a knowledge-aware multisite adaptive graph Transformer to address the above problems. First, we evaluate the sensitivity of features to each piece of non-imaging information, and then construct feature-sensitive and feature-insensitive subgraphs. Second, after fusing the above subgraphs, we integrate a Transformer module to capture the intrinsic relationship between features. Third, we design a domain adaptive GCN using multiple loss function terms to relieve data heterogeneity and to produce the final classification results. Last, the proposed framework is validated on two brain disorder diagnostic tasks. Experimental results show that the proposed framework can achieve state-of-the-art performance.

Histone modifications and their roles in macrophage-mediated inflammation: a new target for diabetic wound healing.

Impaired wound healing is one of the main clinical complications of type 2 diabetes (T2D) and a major cause of lower limb amputation. Diabetic wounds exhibit a sustained inflammatory state, and reducing inflammation is crucial to diabetic wounds management. Macrophages are key regulators in wound healing, and their dysfunction would cause exacerbated inflammation and poor healing in diabetic wounds. Gene regulation caused by histone modifications can affect macrophage phenotype and function during diabetic wound healing. Recent studies have revealed that targeting histone-modifying enzymes in a local, macrophage-specific manner can reduce inflammatory responses and improve diabetic wound healing. This article will review the significance of macrophage phenotype and function in wound healing, as well as illustrate how histone modifications affect macrophage polarization in diabetic wounds. Targeting macrophage phenotype with histone-modifying enzymes may provide novel therapeutic strategies for the treatment of diabetic wound healing.

Tofacitinib versus thalidomide for mucocutaneous lesions of systemic lupus erythematosus: A real-world CSTAR cohort study XXVII.

OBJECTIVE: Thalidomide is an effective medication for refractory mucocutaneous lesions of systemic lupus erythematosus (SLE) and can treat arthritis in some autoimmune diseases, but it has some adverse reactions. Recently, the effectiveness of tofacitinib in treating mucocutaneous lesions of SLE has been reported. We aimed to compare the efficacy and safety of tofacitinib with thalidomide in treating mucocutaneous and musculoskeletal lesions in patients with SLE. METHODS: This study was a real-world cohort study based on the Chinese SLE Treatment and Research group (CSTAR) registry. SLE patients who manifested mucocutaneous and/or musculoskeletal symptoms and were prescribed tofacitinib or thalidomide were included. We retrospectively conducted comparisons between the tofacitinib and thalidomide groups regarding clinical improvements, SLE disease activity, serological indicators, glucocorticoid doses, and adverse events at the 1, 3, and 6-months time points. RESULTS: At 3 and 6 months, the tofacitinib group exhibited a higher proportion of patients with improvement in mucocutaneous and musculoskeletal issues. Additionally, a greater percentage of patients in the tofacitinib group achieved remission or a low disease activity state (LLDAS) at these time points. No significant serological improvements were observed in either the tofacitinib or thalidomide groups. Fewer adverse events were observed in the tofacitinib group than in the thalidomide group. CONCLUSIONS: Tofacitinib might be superior to thalidomide in the improvement of mucocutaneous and musculoskeletal lesions in SLE, and had a good safety profile.

[Application scenario design and prospect of generative artificial intelligence (AI) in intelligent manufacturing and supply chain of traditional Chinese medicine].

Intelligent manufacturing technologies, including databases, mathematical modeling, and information systems have played a significant role in process control, production management, and supply chain management in traditional Chinese medicine(TCM) industry. However, their ability to process and utilize unstructured data, such as research and development reports, batch production records, quality inspection records, and supplier documents, is relatively weak. For text, images, language, and other unstructured data, generative artificial intelligence(AI) technology has shown strong potential for development in extracting information, extracting knowledge, semantic retrieval, and content generation. Generative AI is expected to provide a feasible set of tools for the utilization of unstructured data resources in the TCM industry. Based on years of research and industrial application experience in TCM intelligent manufacturing technology, this study reviewed the current situation of intelligent manufacturing in TCM and the utilization of unstructured data, analyzed the application value of generative AI in the TCM manufacturing process and supply chain, summarized four typical application scenarios, including intelligent pharmaceutical knowledge base/knowledge graph, intelligent on-the-job trai-ning, intelligent production quality control, and intelligent supply chain. Furthermore, this study also explained the data collection and processing, business process design, application potential, and value of each scenario based on industry demands. Finally, based on the integration of generative AI and TCM industrial models, the study proposed a preliminary concept of a smart industrial brain for TCM, aiming to provide a reference for the application of AI technology in the field of TCM manufacturing.

A systematic review and Bayesian analysis of the adverse effects of dienogest.

BACKGROUND AND OBJECTIVE: Endometriosis and adenomyosis are two common diseases that impair women's health, and dienogest is one of the pharmacologic treatments which is the first-line therapeutic option for patients with pelvic pain and individuals who have no desire for immediate pregnancy. The goal of this study was to summarize the current evidence of adverse events associated with dienogest as well as the prevalence of these adverse events during treatment with dienogest. METHODS: Several databases (PubMed, Embase, Cochrane Central and Clinicaltrials.gov, etc.) and the US FDA Adverse Event Reporting System (FAERS) Public Dashboard were searched on May 31, 2023, using the topic words alongside free words of dienogest and "adverse reaction". Studies were incorporated into this research if they reported or assessed safety issues or adverse reactions of dienogest during the period of endometriosis treatment or adenomyosis therapy. The extracted information comprised trial design, dienogest and control group demographics, as well as reported side effects. RESULTS: This systematic review comprehended 39 publications in total. The mean age of patients in the included studies was 34.43 years. The follow-up duration varied from 3 to 60 months. Most adverse reactions were common and not serious, and the most common adverse reactions during dienogest medication were abnormal uterine bleeding (55%, 95% CI 37-73%), amenorrhea (17%, 95% CI 2-42%) and swelling (13%, 95% CI 3-28%). Uncommon adverse reactions included dysmenorrhea (0.2%, n = 1), dyspepsia (0.4%, n = 1), and (lower) abdominal pain (1%, 95% CI 0-3%), urticaria (1%, 95% CI 0-3%) and peritonitis (1%, n = 1). Serious adverse reactions including decreased lumbar spine Bone Mineral Density (BMD), depression, peritonitis and so on have been reported. Heterogeneity assessment revealed that patient number and study design are influencing factors to adverse reaction prevalence. Moreover, abdominal pain, diarrhea, nausea and vomiting, back pain and anemia are side effects reported both in the FAERS database and in the systematic review. CONCLUSIONS: Dienogest's most frequent side effects were not severe. Dienogest is generally safe for treating endometriosis and adenomyosis. Nevertheless, people should be aware of serious adverse reactions, such as decreased lumbar spine BMD and hemorrhagic shock.

Mechanism and application of fibrous proteins in diabetic wound healing: a literature review.

Diabetic wounds are more complex than normal chronic wounds because of factors such as hypoxia, reduced local angiogenesis, and prolonged inflammation phase. Fibrous proteins, including collagen, fibrin, laminin, fibronectin, elastin etc., possess excellent inherent properties that make them highly advantageous in the area of wound healing. Accumulating evidence suggests that they contribute to the healing process of diabetic wounds by facilitating the repair and remodel of extracellular matrix, stimulating the development of vascular and granulation tissue, and so on. However, there is currently a lack of a comprehensive review of the application of these proteins in diabetes wounds. An overview of fibrous protein characteristics and the alterations linked to diabetic wounds is given in this article's initial section. Next is a summary of the advanced applications of fibrous proteins in the last five years, including acellular dermal matrix, hydrogel, foam, scaffold, and electrospun nanofibrous membrane. These dressings have the ability to actively promote healing in addition to just covering wounds compared to traditional wound dressings like gauze or bandage. Research on fibrous proteins and their role in diabetic wound healing may result in novel therapeutic modalities that lower the incidence of diabetic wounds and thereby enhance the health of diabetic patients.

PTBP3 Mediates IL-18 Exon Skipping to Promote Immune Escape in Gallbladder Cancer.

Gallbladder cancer (GBC) is the most common malignant tumor of the biliary system, with poor response to current treatments. Abnormal alternative splicing has been associated with the development of a variety of tumors. Combining the GEO database and GBC mRNA-seq analysis, it is found high expression of the splicing factor polypyrimidine region- binding protein 3 (PTBP3) in GBC. Multi-omics analysis revealed that PTBP3 promoted exon skipping of interleukin-18 (IL-18), resulting in the expression of ΔIL-18, an isoform specifically expressed in tumors. That ΔIL-18 promotes GBC immune escape by down-regulating FBXO38 transcription levels in CD8+T cells to reduce PD-1 ubiquitin-mediated degradation is revealed. Using a HuPBMC mouse model, the role of PTBP3 and ΔIL-18 in promoting GBC growth is confirmed, and showed that an antisense oligonucleotide that blocked ΔIL-18 production displayed anti-tumor activity. Furthermore, that the H3K36me3 promotes exon skipping of IL-18 by recruiting PTBP3 via MRG15 is demonstrated, thereby coupling the processes of IL-18 transcription and alternative splicing. Interestingly, it is also found that the H3K36 methyltransferase SETD2 binds to hnRNPL, thereby interfering with PTBP3 binding to IL-18 pre-mRNA. Overall, this study provides new insights into how aberrant alternative splicing mechanisms affect immune escape, and provides potential new perspectives for improving GBC immunotherapy.

Copper-catalyzed highly switchable defluoroborylation and hydrodefluorination of 1-(trifluoromethyl)alkynes.

CF2-containing compounds hold significant potential in drug discovery, organic synthesis, and materials science. However, synthesizing various CF2-containing building blocks from a single compound remains challenging. Here, we present a Cu-catalyzed, switchable defluoroborylation and hydrodefluorination of trifluoromethylated alkynes, yielding four types of CF2-containing compounds. The chemo- and regio-selective sp2/sp3 1,2-diborylation and sp2 monoborylation of 1-(trifluoromethyl)alkynes are controlled by adjusting the solvent and ligand quantity. Additionally, altering the base allows selective generation of gem-difluoroalkenes or difluoromethylalkenes. Notably, our method prevents over-defluorination of the CF3 group on unsaturated C-C bonds during nucleophilic additions, preserving the pharmaceutically valuable CF2 group. Experimental data and density functional theory (DFT) calculations elucidate the regioselectivities of Cu-Bpin addition and the regulatory role of the ligand in selective deborylation processes.

Proper application of DNA dyes in agarose gel electrophoresis.

Various dyes are used to visualize DNA bands in agarose gel electrophoresis (AGE) by the methods of pre- or post-staining. The DNA dye user's guides generally state that the binding of the dye to DNA will affect DNA mobility in electrophoresis, thus recommending post-staining for accurate measurement of DNA size. However, many AGE performers prefer pre-staining procedures for reasons such as convenience, real-time observation of DNA bands, and/or the use of a minimal amount of dye. The detrimental effect of the dye on DNA mobility and the associated risk for inaccurate measurement of DNA size are often overlooked by AGE performers. Here we quantitatively determine the impact on DNA migration imposed by frequently used dyes, including GelRed, ethidium bromide (EB), and Gold View. It was observed that pre-staining with GelRed and EB significantly slowed down DNA migration to cause as much as 39.1% overestimation on the size of sample DNA, whereas Gold View had little effect. The slowdown of DNA migration increased with dye concentration until it plateaued when the dye concentration reached a saturated level. Thus, to take advantage of pre-staining, saturated levels of DNA dyes should always be applied for both DNA samples and DNA markers to ensure a fair comparison of DNA sizes. In addition, GelRed and EB display much higher sensitivity than Gold View in the detection of DNA bands in post-staining. The saturated concentrations, cost considerations, and other useful features of these frequently used dyes are summarized for the information of AGE performers.

Combining Gut Microbiota Modulation and Enzymatic-Triggered Colonic Delivery by Prebiotic Nanoparticles Improves Mouse Colitis Therapy.

The efficacy of ulcerative colitis (UC) therapy is closely connected to the composition of gut microbiota in the gastrointestinal tract. Prebiotic-based nanoparticles (NPs) provide a more precise approach to alleviate UC via modulating gut microbiota dysbiosis. The present study develops an efficient prebiotic-based colon-targeted drug delivery system (PCDDS) by using prebiotic pectin (Pcn) and chitosan (Csn) polysaccharides as a prebiotic shell, with the anti-inflammatory drug sulfasalazine (SAS) loaded into a poly(lactic-co-glycolic acid) (PLGA) core to construct SAS@PLGA-Csn-Pcn NPs. Then, we examine its characterization, cellular uptake, and in vivo therapeutic efficacy. The results of our study indicate that the Pcn/Csn shell confers efficient pH-sensitivity properties. The gut microbiota-secreted pectinase serves as the trigger agent for Pcn/Csn shell degradation, and the resulting Pcn oligosaccharides possess a substantial prebiotic property. Meanwhile, the formed PCDDSs exhibit robust biodistribution and accumulation in the colon tissue, rapid cellular uptake, efficient in vivo therapeutic efficacy, and modulation of gut microbiota dysbiosis in a mouse colitis model. Collectively, our synthetic PCDDSs demonstrate a promising and synergistic strategy for UC therapy.

An endo-functionalized molecular cage for selective potentiometric determination of creatinine.

Potentiometric ion-selective electrodes (ISEs), which rely on selective and lipophilic ionophores, are commonly employed in clinical diagnostics. However, there are very limited specific ionophores for the detection of creatinine, a critical biomarker for renal function assessment. In the present research, we designed and synthesized an endo-functionalized cage, which is able to selectively bind the creatininium cation (K a = 8.6 × 105 M-1) through the formation of multiple C-H⋯O and N-H⋯N hydrogen bonds and cation⋯π interactions. ISEs prepared with this host show a Nernstian response to creatinine and exhibit excellent selectivity and a low detection limit of 0.95 µM. In addition, the creatinine levels in urine or plasma samples determined by our sensor are consistent with those analyzed using enzymatic assay on a Cobas c702. The method is simple, fast and accurate, and amenable to clinical detection of creatinine levels.

Synthesis of Benzopyrone-Fused Hydrobenzo[c,d]indoles via Cascade Annulation of p-Quinamines and 3-Formylchromones.

A 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)-catalyzed cascade annulation reaction between p-quinamines and 3-formylchromones was developed, affording a series of benzopyrone-fused hydrobenzo[c,d]indoles in moderate to high yields with excellent diastereoselectivity. This cascade reaction is efficient since two new rings as well as one C-N, one C═C, and two C-C bonds are created in a single step. The scale-up synthesis and versatile transformations of the products further demonstrated the practicality and utility of this approach.

An insoluble cellulose nanofiber with robust expansion capacity protects against obesity.

An imbalance between energy intake and energy expenditure predisposes obesity and its related metabolic diseases. Soluble dietary fiber has been shown to improve metabolic homeostasis mainly via microbiota reshaping. However, the application and metabolic effects of insoluble fiber are less understood. Herein, we employed nanotechnology to design citric acid-crosslinked carboxymethyl cellulose nanofibers (CL-CNF) with a robust capacity of expansion upon swelling. Supplementation with CL-CNF reduced food intake and delayed digestion rate in mice by occupying stomach. Besides, CL-CNF treatment mitigated diet-induced obesity and insulin resistance in mice with enhanced energy expenditure, as well as ameliorated inflammation in adipose tissue, intestine and liver and reduced hepatic steatosis, without any discernible signs of toxicity. Additionally, CL-CNF supplementation resulted in enrichment of probiotics such as Bifidobacterium and decreased in the relative abundances of deleterious microbiota expressing bile salt hydrolase, which led to increased levels of conjugated bile acids and inhibited intestinal FXR signaling to stimulate the release of GLP-1. Taken together, our findings demonstrate that CL-CNF administration protects mice from diet-induced obesity and metabolic dysfunction by reducing food intake, enhancing energy expenditure and remodeling gut microbiota, making it a potential therapeutic strategy against metabolic diseases.

Attention-Guided Learning with Feature Reconstruction for Skin Lesion Diagnosis using Clinical and Ultrasound Images.

Skin lesion is one of the most common diseases, and most categories are highly similar in morphology and appearance. Deep learning models effectively reduce the variability between classes and within classes, and improve diagnostic accuracy. However, the existing multi-modal methods are only limited to the surface information of lesions in skin clinical and dermatoscopic modalities, which hinders the further improvement of skin lesion diagnostic accuracy. This requires us to further study the depth information of lesions in skin ultrasound. In this paper, we propose a novel skin lesion diagnosis network, which combines clinical and ultrasound modalities to fuse the surface and depth information of the lesion to improve diagnostic accuracy. Specifically, we propose an attention-guided learning (AL) module that fuses clinical and ultrasound modalities from both local and global perspectives to enhance feature representation. The AL module consists of two parts, attention-guided local learning (ALL) computes the intra-modality and inter-modality correlations to fuse multi-scale information, which makes the network focus on the local information of each modality, and attention-guided global learning (AGL) fuses global information to further enhance the feature representation. In addition, we propose a feature reconstruction learning (FRL) strategy which encourages the network to extract more discriminative features and corrects the focus of the network to enhance the model's robustness and certainty. We conduct extensive experiments and the results confirm the superiority of our proposed method. Our code is available at: https://github.com/XCL-hub/AGFnet.

Leaf rolling detection in maize under complex environments using an improved deep learning method.

Leaf rolling is a common adaptive response that plants have evolved to counteract the detrimental effects of various environmental stresses. Gaining insight into the mechanisms underlying leaf rolling alterations presents researchers with a unique opportunity to enhance stress tolerance in crops exhibiting leaf rolling, such as maize. In order to achieve a more profound understanding of leaf rolling, it is imperative to ascertain the occurrence and extent of this phenotype. While traditional manual leaf rolling detection is slow and laborious, research into high-throughput methods for detecting leaf rolling within our investigation scope remains limited. In this study, we present an approach for detecting leaf rolling in maize using the YOLOv8 model. Our method, LRD-YOLO, integrates two significant improvements: a Convolutional Block Attention Module to augment feature extraction capabilities, and a Deformable ConvNets v2 to enhance adaptability to changes in target shape and scale. Through experiments on a dataset encompassing severe occlusion, variations in leaf scale and shape, and complex background scenarios, our approach achieves an impressive mean average precision of 81.6%, surpassing current state-of-the-art methods. Furthermore, the LRD-YOLO model demands only 8.0 G floating point operations and the parameters of 3.48 M. We have proposed an innovative method for leaf rolling detection in maize, and experimental outcomes showcase the efficacy of LRD-YOLO in precisely detecting leaf rolling in complex scenarios while maintaining real-time inference speed.