ObjectiveThe widespread adoption of portable fundus cameras for primary care and community screening is hindered by limitations in current autofocus(AF) technologies. Image-based methods relying on sharpness evaluation require iterative searches, resulting in slow convergence, while projection-based techniques are susceptible to optical artifacts and calibration errors. To address these challenges, this study introduces a novel AF system based on direct wavefront sensing, designed to deliver simultaneous high speed, high precision, and operational robustness within the compact form factor essential for portable ophthalmic devices. MethodsOur approach fundamentally reimagines the AF process by directly measuring the ocular wavefront aberration. We developed a custom portable fundus camera integrating a miniaturized Shack-Hartmann wavefront sensor (SHWS) into the optical path. An 850 nm laser diode projects a point source onto the retina via oblique illumination to minimize corneal reflections. Light scattered from this spot carries the eye’s refractive error through the imaging optics and is directed to the SHWS, positioned at a plane optically conjugate to the primary color CMOS imaging sensor. A microlens array within the SHWS samples the incident wavefront, generating a pattern of focal spots on a CCD. Real-time centroid analysis of these spots provides a map of local wavefront slopes. These measurements are processed through a singular value decomposition (SVD) algorithm to fit a Zernike polynomial basis set, enabling real-time reconstruction of the wavefront phase. The defocus component (S) is extracted from the second-order Zernike coefficients, providing a direct, quantitative measure of the refractive error in diopters. This value serves as a precise error signal in a closed-loop control system, which commands a voice-coil actuated focusing lens to its null position in a single, deterministic step, eliminating the need for iterative search algorithms. ResultsComprehensive evaluation demonstrated the system’s high performance. Testing on a calibrated model eye (OEMI-7) established a highly linear relationship between the computed defocus S and the focusing lens position across a ±20 Diopter (D) compensation range, achievable within a 5 mm mechanical travel. The system achieved a focusing precision of 0.08 D, corresponding to an 18-fold improvement over a conventional projection spot-size method tested under identical conditions. The total focus acquisition time, encompassing wavefront measurement, computation, and lens actuation, averaged under 0.5 s. Clinical validation with 25 human volunteers (50 eyes, refractive range -15 D to +10 D) confirmed practical efficacy. The wavefront-sensing AF succeeded in 92% of attempts with a mean time of 0.5 s, substantially outperforming a projection-based benchmark which achieved only a 32% success rate with an average time of 4.25 s. The system provided instantaneous directional guidance and maintained stability during minor ocular movements. Objective assessment of image quality, via amplitude contrast of retinal vasculature, showed consistent and significant enhancement following AF correction across the entire tested diopter range. ConclusionThis work successfully implements and validates a direct wavefront-sensing autofocus paradigm for portable fundus cameras. By directly quantifying and compensating for the optical defocus aberration, this method bypasses the fundamental limitations of image-processing and projection-based techniques, enabling rapid, precise, and deterministic diopter compensation. The developed system delivers an exceptional combination of a wide operational range (±20 D), high accuracy (0.08 D), fast convergence (0.5 s), and a compact physical footprint. This technology provides a practical and high-performance focusing solution capable of enhancing the reliability, throughput, and diagnostic utility of portable retinal imaging in large-scale screening applications. Future efforts will be directed towards system cost optimization and performance adaptation for diverse ocular conditions.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

As an innovative dosage form, traditional Chinese medicine(TCM) dry powder inhalers have emerged as a focal point in the research and development of new preparations due to its high efficiency, safety, and bioavailability. This paper systematically reviewed the relevant literature and patents associated with TCM dry powder inhalers to analyze the origins and the current research and development status. Furthermore, this paper probed into the research and development ideas of TCM dry powder inhalers regarding clinical positioning, prescription screening, and druggability. Additionally, the paper thoroughly analyzed the technical barriers in druggability studies and elaborated on corresponding research techniques and coping measures. Furthermore, it emphasized the need for improved regulations and policies governing TCM dry powder inhalers, advocated for strengthened oversight, and called for the establishment of a scientific quality evaluation system. Measures such as promoting production-education-research collaboration, enhancing personnel training, and fostering international exchanges were proposed to provide a scientific and systematic reference for the future research, development, and application of TCM dry powder inhalers, thereby facilitating the rapid modernization of TCM.
Humans ; Dry Powder Inhalers/trends* ; Drugs, Chinese Herbal/chemistry* ; Medicine, Chinese Traditional/instrumentation* ; Administration, Inhalation

This study aims to explore the effects and mechanisms of 4'-O-methylbavachalcone(MeBavaC), an active compound from Psoraleae Fructus, in regulating white matter neuroinflammation to improve vascular cognitive impairment. Male Sprague-Dawley(SD) rats were randomly divided into four groups: sham group, model group, high-dose MeBavaC group(14 mg·kg~(-1)), and low-dose MeBavaC group(7 mg·kg~(-1)). The rat model of chronic cerebral hypoperfusion(CCH) was established using bilateral common carotid artery occlusion. The Morris water maze test was performed to evaluate the learning and memory abilities of the rats. Luxol fast blue staining, Nissl staining, immunofluorescence, immunohistochemistry, and transmission electron microscopy were utilized to observe the morphology and ultrastructure of the white matter myelin sheaths, axon integrity, the morphology and number of hippocampal neurons, and the loss and activation of glial cells in the white matter. Transcriptome analysis was performed to explore the potential mechanisms of white matter injury induced by CCH. Western blot and quantitative real-time polymerase chain reaction(qRT-PCR) assays were conducted to measure the expression levels of NOD-like receptor protein 3(NLRP3), absent in melanoma 2(AIM2), gasdermin D(GSDMD), cysteinyl aspartate-specific proteinase-1(caspase-1), interleukin-18(IL-18), interleukin-1β(IL-1β), erythropoietin(EPO), nuclear factor erythroid 2-related factor 2(Nrf2), and heme oxygenase-1(HO-1) in the white matter of rats. The results showed that compared with the model group, MeBavaC significantly improved the learning and memory abilities of rats with CCH, improved the damage of white matter myelin sheath, maintained axonal integrity, reduced the loss of hippocampal neurons and oligodendrocytes in the white matter, inhibited the activation of microglia and the proliferation of astrocytes in the white matter, and suppressed the NLRP3/AIM2/caspase-1/GSDMD pathway. The expression levels of inflammatory cytokines IL-1β and IL-18 were significantly reduced, while EPO expression and the expression of Nrf2/HO-1 antioxidant pathway were notably elevated. In conclusion, MeBavaC can alleviate cognitive impairment in rats with CCH and suppress neuroinflammation in cerebral white matter. The mechanism of action may involve activation of EPO activity, promotion of endogenous antioxidant pathways, and inhibition of neuroinflammation in the white matter. This study suggests that MeBavaC exhibits antioxidant and anti-neuroinflammatory effects, showing potential application in improving cognitive dysfunction.
Animals ; Male ; Rats, Sprague-Dawley ; NF-E2-Related Factor 2/immunology* ; Rats ; Chalcones/administration & dosage* ; Cognitive Dysfunction/metabolism* ; Signal Transduction/drug effects* ; Neuroinflammatory Diseases/drug therapy* ; Heme Oxygenase-1/metabolism* ; Humans ; Heme Oxygenase (Decyclizing)/genetics*

OBJECTIVE: To observe the clinical efficacy and safety of automatic pressure-controlled pressure cupping in patients with lumbar disc herniation, and compare it with traditional cupping. METHODS: A total of 100 patients diagnosed with lumbar disc herniation from January 2022 to August 2024 were selected and divided into two groups:the automatic pressure-controlled pressure cupping group (controlled pressure cupping group) and the traditional cupping group (control group), 50 cases in each group. In the controlled pressure cupping group, there were 18 males and 32 females, with an age of (51.98±12.69) years;in the control group, there were 16 males and 34 females, with an age of (51.32±12.05) years. The visual analogue scale(VAS), comfort score, and lumbar range of motion were observed before treatment and after the 1st, 3rd, and 7th treatments to evaluate the efficacy and safety. RESULTS: All patients completed the treatment intervention, with complete follow-up data collected. No adverse reactions or complications occurred during treatment and follow-up. After the 3rd treatment, the VAS score of the controlled pressure cupping group was (2.38±0.49), which was lower than that of the control group (2.94±0.68), with a statistically significant difference (P<0.001). In the controlled pressure cupping group, the VAS scores after the 1st, 3rd, and 7th treatments were significantly better than those before treatment (P=0.026);in the control group, the VAS scores after the 3rd and 7th treatments were better than those before treatment, but the difference was not statistically significant(P=0.182). Repeated-measures analysis of variance (ANOVA) on VAS scores at different time points in both groups showed that there were statistically significant differences in inter-group, time, and interaction effects (P<0.05). After the 1st treatment, in the controlled pressure cupping group, 0 patients felt comfortable, 42 patients (84%) felt mild discomfort, and 8 patients (16%) felt moderate discomfort;in the control group, 0 patients felt comfortable, 28 patients (56%) felt mild discomfort, and 22 patients(44%) felt moderate discomfort;the difference between the two groups was statistically significant(P=0.005). After the 3rd treatment, in the controlled pressure cupping group, 30 patients(60%) felt comfortable, 20 patients (40%) felt mild discomfort, and 0 patients felt moderate discomfort; in the control group, 9 patients (18%) felt comfortable, 41 patients (82%) felt mild discomfort, and 0 patients felt moderate discomfort;the difference between the two groups was statistically significant(P<0.001). There was no statistically significant difference in comfort between the two groups after the 7th treatment(P>0.001). There was no statistically significant difference in lumbar range of motion between the two groups before and after treatment(P>0.05);compared with before treatment, the lumbar range of motion of both groups after treatment was significantly improved, with statistically significant differences (P<0.001). CONCLUSION Automatic pressure-controlled pressure cupping can effectively relieve symptoms in patients with lumbar disc herniation, with excellent safety.
Humans ; Female ; Male ; Intervertebral Disc Displacement/physiopathology* ; Middle Aged ; Adult ; Lumbar Vertebrae/physiopathology* ; Cupping Therapy/methods* ; Pressure ; Aged ; Treatment Outcome

OBJECTIVES: To study the effect of prophylactic phenytoin (PHT) or levetiracetam (LEV) on busulfan (BU) blood concentration in children undergoing hematopoietic stem cell transplantation. METHODS: Pediatric patients conditioned with BU plus cyclophosphamide and fludarabine at the First People's Hospital of Chenzhou from September 2023 to February 2025 were retrospectively included. Patients were grouped by prophylactic antiepileptic regimen into PHT (n=24) and LEV (n=26). BU blood concentrations at the end of infusion (0 hour) and at 1, 2, and 4 hours post-infusion were compared between groups. RESULTS: At 0 hour post-infusion, BU blood concentrations did not differ significantly between groups (P>0.05). At 1, 2, and 4 hours post-infusion, BU blood concentrations were higher in the LEV group than in the PHT group (P<0.05). The area under the concentration-time curve from 0 to ∞ (AUC_0-∞) was greater in the LEV group (P<0.001), and the attainment rate of AUC_0-∞ was higher in the LEV group than in the PHT group (73% vs 21%, P<0.001). No significant differences were observed between groups in time to hematopoietic engraftment or in the incidence of BU-related adverse drug reactions (P>0.05). CONCLUSIONS Compared with PHT, LEV prophylaxis is associated with higher BU blood concentration and a higher AUC_0-∞ attainment rate. There is no observed difference in BU efficacy or safety between PHT and LEV.
Humans ; Levetiracetam/therapeutic use* ; Busulfan/pharmacokinetics* ; Hematopoietic Stem Cell Transplantation ; Male ; Female ; Child ; Child, Preschool ; Phenytoin/pharmacology* ; Infant ; Retrospective Studies ; Anticonvulsants/pharmacology* ; Adolescent

OBJECTIVE: To analyze the distribution characteristics of glucose-6-phosphate-dehydrogenase (G6PD) mutations and their enzyme activity in newborns patients with G6PD deficiency in Guilin region. METHODS: From July 2022 to July 2024, umbilical cord blood samples from 4 554 newborns in Guilin were analyzed for G6PD mutations using fluorescence PCR melting curve analysis. Enzyme activity was detected in 4 467 cases using the rate assay. RESULTS: Among 4 467 newborns who underwent G6PD activity testing, 162 newborns (3.63%) were identified as G6PD-deficient, including 142 males (6.04%) and 20 females (0.94%), the prevalence of G6PD deficiency was significantly higher in males than in females (P < 0.001). Genetic analysis of 4 554 newborns detected G6PD mutations in 410 cases (9%), including 171 males (7.13%) and 239 females (11.09%), with a significantly higher mutation detection rate in females than in males (P < 0.001). A total of nine single mutations and four compound heterozygous mutations were identified. The most common mutations were c.1388G>A (33.66%), c.1376G>T (23.66%) and c.95A>G (16.34%). Among newborns who underwent both enzyme activity and genetic mutation testing, males with G6PD mutations had significantly lower enzyme activity than that of females with G6PD mutations(P < 0.001). Specifically, among newborns carrying the mutations c.1388G>A, c.1376G>T, c.95A>G, c.1024C>T or c.871G>A, males consistently exhibited lower enzymatic activity than females with the same mutations (P < 0.001). Furthermore, in male G6PD-deficient newborns, the enzyme activity levels in those carrying c.1388G>A, c.1376G>T, c.95A>G, c.1024C>T, or c.871G>A were lower than those in both the control group and the c.519C>T group (P < 0.05). CONCLUSION This study provides a comprehensive profile of G6PD deficiency incidence and mutation spectrum in the Guilin region. By analyzing enzyme activity and genetic mutation results, this study provides insights into potential intervention strategies and personalized management approaches for the prevention and treatment of neonatal G6PD deficiency in the region.
Humans ; Infant, Newborn ; Glucosephosphate Dehydrogenase Deficiency/epidemiology* ; Glucosephosphate Dehydrogenase/genetics* ; Female ; Male ; Mutation ; China/epidemiology*

ObjectivePrimary liver cancer, predominantly hepatocellular carcinoma (HCC), is a significant global health issue, ranking as the sixth most diagnosed cancer and the third leading cause of cancer-related mortality. Accurate and early diagnosis of HCC is crucial for effective treatment, as HCC and non-HCC malignancies like intrahepatic cholangiocarcinoma (ICC) exhibit different prognoses and treatment responses. Traditional diagnostic methods, including liver biopsy and contrast-enhanced ultrasound (CEUS), face limitations in applicability and objectivity. The primary objective of this study was to develop an advanced, light-weighted classification network capable of distinguishing HCC from other non-HCC malignancies by leveraging the automatic analysis of brightness changes in CEUS images. The ultimate goal was to create a user-friendly and cost-efficient computer-aided diagnostic tool that could assist radiologists in making more accurate and efficient clinical decisions. MethodsThis retrospective study encompassed a total of 161 patients, comprising 131 diagnosed with HCC and 30 with non-HCC malignancies. To achieve accurate tumor detection, the YOLOX network was employed to identify the region of interest (ROI) on both B-mode ultrasound and CEUS images. A custom-developed algorithm was then utilized to extract brightness change curves from the tumor and adjacent liver parenchyma regions within the CEUS images. These curves provided critical data for the subsequent analysis and classification process. To analyze the extracted brightness change curves and classify the malignancies, we developed and compared several models. These included one-dimensional convolutional neural networks (1D-ResNet, 1D-ConvNeXt, and 1D-CNN), as well as traditional machine-learning methods such as support vector machine (SVM), ensemble learning (EL), k-nearest neighbor (KNN), and decision tree (DT). The diagnostic performance of each method in distinguishing HCC from non-HCC malignancies was rigorously evaluated using four key metrics: area under the receiver operating characteristic (AUC), accuracy (ACC), sensitivity (SE), and specificity (SP). ResultsThe evaluation of the machine-learning methods revealed AUC values of 0.70 for SVM, 0.56 for ensemble learning, 0.63 for KNN, and 0.72 for the decision tree. These results indicated moderate to fair performance in classifying the malignancies based on the brightness change curves. In contrast, the deep learning models demonstrated significantly higher AUCs, with 1D-ResNet achieving an AUC of 0.72, 1D-ConvNeXt reaching 0.82, and 1D-CNN obtaining the highest AUC of 0.84. Moreover, under the five-fold cross-validation scheme, the 1D-CNN model outperformed other models in both accuracy and specificity. Specifically, it achieved accuracy improvements of 3.8% to 10.0% and specificity enhancements of 6.6% to 43.3% over competing approaches. The superior performance of the 1D-CNN model highlighted its potential as a powerful tool for accurate classification. ConclusionThe 1D-CNN model proved to be the most effective in differentiating HCC from non-HCC malignancies, surpassing both traditional machine-learning methods and other deep learning models. This study successfully developed a user-friendly and cost-efficient computer-aided diagnostic solution that would significantly enhances radiologists’ diagnostic capabilities. By improving the accuracy and efficiency of clinical decision-making, this tool has the potential to positively impact patient care and outcomes. Future work may focus on further refining the model and exploring its integration with multimodal ultrasound data to maximize its accuracy and applicability.

With the rapid development of generative artificial intelligence(GAI)technologies,their widespread application in academic research and writing is continuously expanding the boundaries of sci-entific inquiry.However,this trend has also raised a series of ethical and regulatory challenges,inclu-ding issues related to authorship,content authenticity,citation accuracy,and accountability.In light of the growing involvement of AI in generating academic content,establishing an open,controllable,and trustworthy ethical governance framework has become a key task for safeguarding research integrity and maintaining trust within the academic community.This expert consensus outlines ethical requirements across key stages of AI-assisted academic writing-including topic selection,data management,citation practices,and authorship attribution.It aims to clarify the boundaries and ethical obligations surrounding AI use in academic writing,ensuring that technological tools enhance efficiency without compromising in-tegrity.The goal is to provide guidance and institutional support for building a responsible and sustainable research ecosystem.