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.

BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

OBJECTIVE To evaluate the effectiveness and safety of warfarin anticoagulation therapy guided by gene test in patients undergoing left ventricular assist device （LVAD） implantation， and to analyze the influencing factors of warfarin anticoagulation efficacy. METHODS Patients who underwent LVAD implantation at the Heart and Vascular Center of Northern Jiangsu People’s Hospital from January 2023 to October 2024 and required warfarin anticoagulant therapy were selected as the study subjects. They were divided into genetic testing group （n＝51） and empirical treatment group （n＝17） based on whether they underwent CYP2C9 and VKORC1 gene test. The gene test group was given warfarin based on the predicted dose calculated by gene test， while the empirical treatment group was given warfarin by clinical doctors based on international normalized ratio （INR） experience， all patients were given warfarin once a day. Follow-up observation was conducted for 6 months to compare the effectiveness [time in therapeutic range（TTR）， the time required to reach INR for the first time， the incidence of embolic events， the incidence of INR＜1.5 events] and safety （the incidence of major and minor bleeding events，the incidence of INR＞3.5 events） of warfarin treatment between two groups of patients. According to whether the patient’s TTR was ≥60%， they were divided into TTR≥60% group （n＝20） and TTR＜60% group （n＝48）. Univariate and multivariate binary Logistic regression analysis were used to determine the factors affecting the anticoagulant effect of warfarin in patients. RESULTS The TTR of patients in the gene test group was significantly higher than that in the empirical treatment group （P＜0.05）. The incidence of INR＜1.5 events in the gene test group was significantly lower than in the empirical treatment group （P＜0.05）. The incidence of minor bleeding events and INR＞3.5 events in the gene test group were lower than in the empirical treatment group， but the difference was not statistically significant （P＞0.05）. The results of multivariate binary Logistic regression analysis showed that gene test was an independent protective factor for warfarin anticoagulant therapy [odds ratio （OR）＝10.842， 95% confidence interval （CI）： 1.211-27.037， P＝0.033]， and the combination of statins was an independent risk factor for warfarin anticoagulant therapy [OR＝0.196， 95%CI： 0.045-0.861， P＝0.031]. CONCLUSIONS Under the guidance of gene test， warfarin anticoagulation therapy for LVAD patients after implantation can improve TTR， shorten the anticoagulation target time， and has good safety； meanwhile， it should be noted that the combination of statins may enhance the anticoagulant effect of warfarin， thereby increasing the risk of bleeding in patients.

Dormant tumor cells constitute a population of cancer cells that reside in a non-proliferative or low-proliferative state, typically arrested in the G0/G1 phase and exhibiting minimal mitotic activity. These cells are commonly observed across multiple cancer types, including breast, lung, and ovarian cancers, and represent a central cellular component of minimal residual disease (MRD) following surgical resection of the primary tumor. Dormant cells are closely associated with long-term clinical latency and late-stage relapse. Due to their quiescent nature, dormant cells are intrinsically resistant to conventional therapies—such as chemotherapy and radiotherapy—that preferentially target rapidly dividing cells. In addition, they display enhanced anti-apoptotic capacity and immune evasion, rendering them particularly difficult to eradicate. More critically, in response to microenvironmental changes or activation of specific signaling pathways, dormant cells can re-enter the cell cycle and initiate metastatic outgrowth or tumor recurrence. This ability to escape dormancy underscores their clinical threat and positions their effective detection and elimination as a major challenge in contemporary cancer treatment. Hypoxia, a hallmark of the solid tumor microenvironment, has been widely recognized as a potent inducer of tumor cell dormancy. However, the molecular mechanisms by which tumor cells sense and respond to hypoxic stress—initiating the transition into dormancy—remain poorly defined. In particular, the lack of a systems-level understanding of the dynamic and multifactorial regulatory landscape has impeded the identification of actionable targets and constrained the development of effective therapeutic strategies. Accumulating evidence indicates that hypoxia-induced dormancy tumor cells are accompanied by a suite of adaptive phenotypes, including cell cycle arrest, global suppression of protein synthesis, metabolic reprogramming, autophagy activation, resistance to apoptosis, immune evasion, and therapy tolerance. These changes are orchestrated by multiple converging signaling pathways—such as PI3K-AKT-mTOR, Ras-Raf-MEK-ERK, and AMPK—that together constitute a highly dynamic and interconnected regulatory network. While individual pathways have been studied in depth, most investigations remain reductionist and fail to capture the temporal progression and network-level coordination underlying dormancy transitions. Systems biology offers a powerful framework to address this complexity. By integrating high-throughput multi-omics data—such as transcriptomics and proteomics—researchers can reconstruct global regulatory networks encompassing the key signaling axes involved in dormancy regulation. These networks facilitate the identification of core regulatory modules and elucidate functional interactions among key effectors. When combined with dynamic modeling approaches—such as ordinary differential equations—these frameworks enable the simulation of temporal behaviors of critical signaling nodes, including phosphorylated AMPK (p-AMPK), phosphorylated S6 (p-S6), and the p38/ERK activity ratio, providing insights into how their dynamic changes govern transitions between proliferation and dormancy. Beyond mapping trajectories from proliferation to dormancy and from shallow to deep dormancy, such dynamic regulatory models support topological analyses to identify central hubs and molecular switches. Key factors—such as NR2F1, mTORC1, ULK1, HIF-1α, and DYRK1A—have emerged as pivotal nodes within these networks and represent promising therapeutic targets. Constructing an integrative, systems-level regulatory framework—anchored in multi-pathway coordination, omics-layer integration, and dynamic modeling—is thus essential for decoding the architecture and progression of tumor dormancy. Such a framework not only advances mechanistic understanding but also lays the foundation for precision therapies targeting dormant tumor cells during the MRD phase, addressing a critical unmet need in cancer management.

As a hot spot in clinical research today, immune checkpoint inhibitor has been recommended by guidelines in the first- and second-line treatments of advanced cervical cancer as immune monotherapy or combination therapy. It has also achieved good efficacy in clinical practice. In locally advanced cervical cancer, immune checkpoint inhibitors have been included in the guidelines for adjuvant therapy, and good tumor regression effects have been achieved in clinical practice. Based on the results of existing trials, immune checkpoint inhibitors have also shown good clinical potential as neoadjuvant therapy. Furthermore, the issue of immunotherapy rechallenge has increasingly captured clinicians’ attention, offering a potential new therapeutic strategy for cervical cancer patients with prior immunotherapy exposure. In this article, the clinical application and research progress of immune checkpoint inhibitors in the treatment of cervical cancer in recent years are summarized to provide valuable ideas and directions for clinical treatment.

OBJECTIVE: To observe the effect of "Zhibian" (BL54)-toward-"Shuidao" (ST28) needling technique on the relative protein expression of the signaling pathway of nucleotide-binding oligomerization domain-containing protein 1 (NOD1)/ receptor-interacting protein 2 (RIP2)/nuclear factor kappa-B (NF-κB) and the expression of proinflammatory cytokines in the rats with primary dysmenorrhea (PD), so as to explore the underlying mechanism of this acupuncture technique for pain alleviation in PD. METHODS: Thirty female SD rats of SPF grade with normal estrous cycle were randomized into a blank group, a model group and an acupuncture group, 10 rats in each one. Using the intraperitoneal injection with estradiol benzoate combined with oxytocin, PD model was prepared in the model group and the acupuncture group. In the acupuncture group, during model preparation, the intervention with "Zhibian" (BL54)-toward-"Shuidao" (ST28) needling technique was delivered simultaneously, 20 min each time, once daily for consecutive 10 days. On day 11, within 30 min after the intraperitoneal injection with oxytocin, the writhing reaction (latency, frequency and score) was recorded; the morphology of uterine tissue was observed with HE staining, the contents of prostaglandin E₂ (PGE₂), prostaglandin F_2α (PGF_2α), interleukin (IL)-1β, IL-18, cyclooxygenase-2 (COX-2), and tumor necrosis factor-α(TNF-α) in the serum were detected using ELISA method; the relative protein expression of NOD1, RIP2, NF-κB p65, phosphorylation-NF-κB p65 (p-NF-κB p65) was detected in the uterine tissue using Western blot method; and the mRNA expression of NOD1, RIP2 and NF-κB p65 was detected with the quantitative real-time PCR employed. RESULTS: Compared with the blank group, in the model group, the writhing latency was prolonged (P<0.01), the writhing frequency and score increased (P<0.01) in the rats; the endometrial epithelial cells showed massive degeneration and necrosis, with severe endometrial edema and widespread shedding, combined with neutrophil infiltration; the serum PGE₂ content was dropped (P<0.01), while those of PGF_2α, IL-1β, IL-18, COX-2, and TNF-α elevated (P<0.01); the protein expression of NOD1, RIP2, NF-κB p65 and p-NF-κB p65, and the mRNA expression of NOD1, RIP2 and NF-κB p65 in uterine tissue increased (P<0.01). In comparison with the model group, in the acupuncture group, the writhing latency was prolonged (P<0.01), the writhing frequency and score were reduced (P<0.01) in the rats; there was less degeneration and necrosis of endometrial epithelial cells, with mild endometrial edema and very little neutrophil infiltration; the serum PGE₂ content increased (P<0.01), while those of PGF_2α, IL-1β, IL-18, COX-2, and TNF-α decreased (P<0.01); the protein expression of NOD1, RIP2, NF-κB p65 and p-NF-κB p65 and the mRNA expression of NOD1, RIP2 and NF-κB p65 in uterine tissue were dropped (P<0.05, P<0.01). CONCLUSION "Zhibian" (BL54)-toward-"Shuidao" (ST28) needling technique can alleviate the pain symptom of PD rats, and its action mechanism may be related to inhibiting the active expression of NOD1/RIP2/NF-κB signaling pathway in the uterine tissue, thereby reducing the inflammatory response.
Animals ; Female ; Rats, Sprague-Dawley ; Rats ; Signal Transduction ; Dysmenorrhea/metabolism* ; NF-kappa B/metabolism* ; Acupuncture Points ; Humans ; Acupuncture Analgesia ; Nod1 Signaling Adaptor Protein/metabolism* ; Receptor-Interacting Protein Serine-Threonine Kinase 2/metabolism* ; Acupuncture Therapy

OBJECTIVE: To investigate the underlying mechanism of "Zhibian" (BL54)-toward-"Shuidao" (ST28) acupuncture technique for improving reproductive function in mice with asthenospermia by regulating ferroptosis pathway. METHODS: Sixty male C57BL/6 mice were randomly divided into a blank group, a model group, an acupuncture group and a Fer-1 group, 15 mice in each one. Except the blank group, the intraperitoneal injection with cyclophosphamide (50·kg^-1·d^-1) was administered to establish the asthenospermia model in the mice of the rest 3 groups for 5 consecutive days. In the acupuncture group, "Zhibian" (BL54)-toward-"Shuidao" (ST28) acupuncture technique was operated in the mice, for 20 min each time; and in the Fer-1 group, Fer-1 solution (1 mg/kg) was injected intraperitoneally. The interventions of these two groups were delivered once daily and for 2 consecutive weeks. The testicular wet weight was measured and the testicular coefficient was calculated. Using sperm quality detection system, the sperm quality was detected. With ELISA used, the contents of testosterone (T), follicle-stimulating hormone (FSH) and luteinizing hormone (LH) in the serum were detected. With HE staining, testicular and epididymal morphology was observed. Immunofluorescence was used to detect the expression of reactive oxygen species (ROS) in the testes. Biochemical assay was conducted to determine the contents of malondialdehyde (MDA), reduced glutathione (GSH), and total iron ion (TFe) in the testicular tissue. Transmission electron microscopy was used to examine mitochondrial structure of the testis, while JC-1 staining was used to assess mitochondrial membrane potential in the testicular tissue. Fluorescence quantitative PCR and Western blot analyses were employed to measure the mRNA and protein expression of solute carrier family 7 member 11 (SLC7A11), glutathione peroxidase 4 (GPX4), ferritin heavy chain 1 (FTH1), and acyl-CoA synthetase long-chain family member 4 (ACSL4) in the testicular tissue. RESULTS: Compared with the blank group, in the model group, the testicular wet weight and testicular coefficient decreased (P<0.01); the sperm concentration and sperm motility reduced (P<0.01), and the contents of T, FSH, and LH decreased in the serum (P<0.01); and the seminiferous tubules in the testis showed loose structure and deformed lumen, sperm cells were disorganized and the sperm numbers reduced; the tubular walls became thinner, and sperm numbers in the lumen less; the expression of ROS in testicular tissue, as well as the contents of MDA and TFe increased (P<0.01), and the content of GSH decreased (P<0.01); and the numbers of mitochondria reduced, the structure of cristae was serious damaged, and mitochondrial membrane potential level declined (P<0.01); the mRNA and protein expression of SLC7A11, GPX4, and FTH1 decreased (P<0.01), while the mRNA and protein expression of ACSL4 increased (P<0.01). In comparison with the model group, the acupuncture and Fer-1 groups showed the increase of testicular wet weight and coefficient (P<0.01), sperm concentration and motility (P<0.01), and the serum contents of T, FSH, and LH (P<0.01); and the improvements in testicular and epididymal histopathology; ROS expression and the contents of MDA and TFe decreased (P<0.01), and the content of GSH elevated (P<0.05); the mitochondrial structure and numbers were ameliorated and mitochondrial membrane potential rose (P<0.01). Besides, in comparison with the model group, the mRNA expression of SLC7A11 was higher (P<0.05, P<0.01), the mRNA and protein expression of GPX4 and FTH1 increased (P<0.01, P<0.05), and the mRNA and protein expression of ACSL4 decreased (P<0.01) in the acupuncture and the Fer-1 groups; and the protein expression of SLC7A11 was higher in the Fer-1 group (P<0.05). CONCLUSION "Zhibian" (BL54)-toward-"Shuidao" (ST28) acupuncture technique may improve the reproductive capacity in the mice with asthenospermia by alleviating ferroptosis-induced cellular damage and ameliorating testicular function.
Animals ; Male ; Ferroptosis ; Mice ; Acupuncture Therapy ; Mice, Inbred C57BL ; Asthenozoospermia/metabolism* ; Humans ; Acupuncture Points ; Testis/metabolism* ; Luteinizing Hormone/metabolism* ; Malondialdehyde/metabolism* ; Reproduction ; Testosterone/metabolism*

OBJECTIVE: To explore the possible mechanism by which the "Zhibian" (BL54) toward "Shuidao" (ST28) acupuncture improves ovarian function in mice with poor ovarian response (POR) by observing its effect on gut microbiota. METHODS: A total of 35 SPF-grade C57BL/6 female mice were screened for normal estrous cycles using vaginal smears, and 30 mice were selected. Ten mice were assigned to the blank group, while the remaining mice were used to establish the POR model by intragastric administration of tripterygium wilfordii suspension. The successfully modeled mice were randomly divided into a model group and an acupuncture group, with 10 mice in each group. After modeling, the acupuncture group received the "Zhibian" (BL54) toward "Shuidao" (ST28) acupuncture method once daily for 20 minutes per session. Ovulation induction began the day after the intervention, and samples were collected after ovulation induction. Vaginal cytology was used to observe estrous cycle changes, and the number of oocytes obtained, ovarian wet weight, and ovarian index were recorded. Serum levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), and anti-Müllerian hormone (AMH) were detected using ELISA. HE staining was used to observe ovarian histology. Gut microbiota was analyzed using 16S rRNA gene sequencing technology. Western blot was used to detect the relative protein expression levels of Occludin and zonula occludens-1 (ZO-1) in colonic tissue. Correlation analysis was conducted among serum hormone indexes, the number of oocytes obtained, ovarian index and gut microbiota. RESULTS: Compared with the blank group, the model group showed a higher estrous cycle disorder rate (P<0.01), increased serum FSH and LH levels, and a higher LH/FSH ratio (P<0.01), while the number of oocytes obtained, ovarian wet weight, ovarian index, and serum E2 and AMH levels were significantly reduced (P<0.01). Compared with the model group, the acupuncture group showed a lower estrous cycle disorder rate (P<0.01), decreased serum FSH and LH levels, and a lower LH/FSH ratio (P<0.01), along with an increased number of oocytes obtained, higher ovarian wet weight, ovarian index, and elevated serum AMH and E2 levels (P<0.01, P<0.05). The blank group had a large number of well-developed primordial follicles, with abundant and closely arranged follicles at various stages. In the model group, there was a significant increase in the number of atretic follicles, a reduction in the number of follicles at various stages, and loosely arranged ovarian tissue. Compared with the blank group, the model group showed a significant decrease in the number of normal follicles (P<0.01) and an increase in the number of atretic follicles (P<0.01). The acupuncture group showed a reduction in atretic follicles and an increase in the number of follicles at various stages compared with the model group, with a significant increase in normal follicles (P<0.01) and a decrease in atretic follicles (P<0.01). Compared with the blank group, the model group exhibited reduced gut microbiota diversity and richness, with significantly lower Chao1 and Shannon indices (P<0.01), and a greater clustering distance from the blank group. The model group also showed an increase in the relative abundance of Firmicutes_D, Verrucomicrobiota, Paramuribaculum, Dubosiella, and Muribaculum (P<0.01, P<0.05), while the relative abundance of Firmicutes_A and the relative protein expression of Occludin and ZO-1 in colonic tissue were decreased (P<0.01). Compared with the model group, the acupuncture group showed improved gut microbiota diversity and richness, with increased Chao1 and Shannon indices (P<0.05), and a clustering distance closer to the blank group. The acupuncture group exhibited reduced relative abundance of Firmicutes_D, Verrucomicrobiota, and Muribaculum (P<0.05, P<0.01), while the relative abundance of Firmicutes_A and the relative protein expression of Occludin and ZO-1 were significantly increased (P<0.01, P<0.05). Correlation analysis indicated a relationship between gut microbiota and serum hormone indicators, as well as the ovarian index. Kyoto encyclopedia of genes and genomes (KEGG) pathway analysis showed that the metabolic pathways of the intersecting species were related to amino acid biosynthesis and nucleotide metabolism. CONCLUSION The "Zhibian" (BL54) toward "Shuidao" (ST28) acupuncture method improves ovarian function in POR mice, and its mechanism may be related to regulating gut microbiota structure and maintaining intestinal barrier homeostasis.
Animals ; Female ; Gastrointestinal Microbiome ; Mice ; Acupuncture Therapy ; Mice, Inbred C57BL ; Humans ; Ovary/physiopathology* ; Acupuncture Points ; Follicle Stimulating Hormone/metabolism* ; Luteinizing Hormone/metabolism* ; Estrous Cycle ; Anti-Mullerian Hormone/blood*

In this paper, near-infrared spectroscopy(NIRS) was employed to analyze 129 batches of commercial products of Reduning Injection. The batch reporting rate was estimated according to the report of Reduning Injection in the direct adverse drug reaction(ADR) reporting system of the drug marketing authorization holder of the Center for Drug Reevaluation of the National Medical Products Administration(National Center for ADR Monitoring) from August 2021 to August 2022. According to the batch reporting rate, the samples of Reduning Injection were classified into those with potential risks and those being safe. No processing, random oversampling(ROS), random undersampling(RUS), and synthetic minority over-sampling technique(SMOTE) were then employed to balance the unbalanced data. After the samples were classified according to appropriate sampling methods, competitive adaptive reweighted sampling(CARS), successive projections algorithm(SPA), uninformative variables elimination(UVE), and genetic algorithm(GA) were respectively adopted to screen the features of spectral data. Then, support vector machine(SVM), logistic regression(LR), k-nearest neighbors(KNN), naive bayes(NB), random forest(RF), and artificial neural network(ANN) were adopted to establish the risk prediction models. The effects of the four feature extraction methods on the accuracy of the models were compared. The optimal method was selected, and bayesian optimization was performned to optimize the model parameters to improve the accuracy and robustness of model prediction. To explore the correlations between potential risks of clinical use and quality test data, TreeNet was employed to identify potential quality parameters affecting the clinical safety of Reduning Injection. The results showed that the models established with the SVM, LR, KNN, NB, RF, and ANN algorithms had the F1 scores of 0.85, 0.85, 0.86, 0.80, 0.88, and 0.85 and the accuracy of 88%, 88%, 88%, 85%, 91%, and 88%, respectively, and the prediction time was less than 5 s. The results indicated that the established models were accurate and efficient. Therefore, near infrared spectroscopy combined with machine learning algorithms can quickly predict the potential risks of clinical use of Reduning Injection in batches. Three key quality parameters that may affect clinical safety were identified by TreeNet, which provided a scientific basis for improving the safety standards of Reduning Injection.
Spectroscopy, Near-Infrared/methods* ; Drugs, Chinese Herbal/administration & dosage* ; Machine Learning ; Algorithms ; Humans ; Quality Control