Objective:To develop a convolutional neural network model of whole slide imaging of cerebrospinal fluid cells for rapid and accurate identification and classification of tumor cells in cerebrospinal fluid.Methods:A total of 8 692 cerebrospinal fluid cytology smears from Huashan Hospital Affiliated to Fudan University from January 2nd, 2019, to December 27th, 2024. As randomly assigned, the training set included 4 941 benign and 1 745 malignant samples, while the validation set comprised of 1 368 benign and 638 malignant samples. Whole-slide digital images were acquired using a cytopathology scanner, cells (clusters) were annotated for classification, and a deep learning model was constructed via tiled image patches for cell detection and classification. Model performance was evaluated using accuracy, sensitivity, specificity, and other indicators. The classification efficiency of manual microscopy was compared.Results:The model achieved a mean precision of 96.75% for cerebrospinal fluid cell classification. For malignant tumor cells, the classification accuracy was 96.61% (mAP=98.36%, AUC=0.97). Subtype classification accuracies for epithelial/epithelioid tumors and small round cell tumors were 97.13% (AUC=0.98) and 95.58% (AUC=0.93), respectively. Compared with manual microscopy, which took (9.70±0.82) minutes for classifying 200 cells, (18.27±1.21) minutes for 500 cells, and often exceeded 60 minutes or infeasible for full slides, the AI model took (3.46±0.49) seconds for 200 cells, (6.76±0.82) seconds for 500 cells, and a median of 48.57 seconds for full slides ( P<0.001), representing an efficiency improvement of approximately 161-170 times, significantly enhancing diagnostic efficiency. Conclusion:This fully automated hierarchical deep learning model enables efficient and accurate tumor cell identification and classification in CSF, providing an effective auxiliary tool for the rapid diagnosis of meningeal carcinomatosis.

At present,the development of clinical specialties in public hospitals is faced with internal and external superposition problems such as high internal and external operating pressure,lagging connotation construction,insufficient driving force for sustainable development,and inadequate implementation of the autonomy of specialty development.In this regard,from the perspective of building an internal and external linkage mechanism,external governance policies and internal management measures of public hospitals should be effectively linked and coordinated to promote the transformation of clinical specialties from high-speed development to high-quality development.Among them,improving the technical ability of diagnosis and treatment of special diseases and ensuring the health of regional people is the fundamental goal,the dual-wheel drive of talent and technology is the key,the personnel and salary management oriented by business development and value output is the driving force,the reform of external governance mechanism and the effective allocation of resources is the guarantee,the performance evaluation and supervision and management is the means,and the continuous medical service system is the support.To provide a fundamental guarantee for the high-quality development of clinical specialties in public hospitals.

At present,the development of clinical specialties in public hospitals is faced with internal and external superposition problems such as high internal and external operating pressure,lagging connotation construction,insufficient driving force for sustainable development,and inadequate implementation of the autonomy of specialty development.In this regard,from the perspective of building an internal and external linkage mechanism,external governance policies and internal management measures of public hospitals should be effectively linked and coordinated to promote the transformation of clinical specialties from high-speed development to high-quality development.Among them,improving the technical ability of diagnosis and treatment of special diseases and ensuring the health of regional people is the fundamental goal,the dual-wheel drive of talent and technology is the key,the personnel and salary management oriented by business development and value output is the driving force,the reform of external governance mechanism and the effective allocation of resources is the guarantee,the performance evaluation and supervision and management is the means,and the continuous medical service system is the support.To provide a fundamental guarantee for the high-quality development of clinical specialties in public hospitals.

Objective:To develop a convolutional neural network model of whole slide imaging of cerebrospinal fluid cells for rapid and accurate identification and classification of tumor cells in cerebrospinal fluid.Methods:A total of 8 692 cerebrospinal fluid cytology smears from Huashan Hospital Affiliated to Fudan University from January 2nd, 2019, to December 27th, 2024. As randomly assigned, the training set included 4 941 benign and 1 745 malignant samples, while the validation set comprised of 1 368 benign and 638 malignant samples. Whole-slide digital images were acquired using a cytopathology scanner, cells (clusters) were annotated for classification, and a deep learning model was constructed via tiled image patches for cell detection and classification. Model performance was evaluated using accuracy, sensitivity, specificity, and other indicators. The classification efficiency of manual microscopy was compared.Results:The model achieved a mean precision of 96.75% for cerebrospinal fluid cell classification. For malignant tumor cells, the classification accuracy was 96.61% (mAP=98.36%, AUC=0.97). Subtype classification accuracies for epithelial/epithelioid tumors and small round cell tumors were 97.13% (AUC=0.98) and 95.58% (AUC=0.93), respectively. Compared with manual microscopy, which took (9.70±0.82) minutes for classifying 200 cells, (18.27±1.21) minutes for 500 cells, and often exceeded 60 minutes or infeasible for full slides, the AI model took (3.46±0.49) seconds for 200 cells, (6.76±0.82) seconds for 500 cells, and a median of 48.57 seconds for full slides ( P<0.001), representing an efficiency improvement of approximately 161-170 times, significantly enhancing diagnostic efficiency. Conclusion:This fully automated hierarchical deep learning model enables efficient and accurate tumor cell identification and classification in CSF, providing an effective auxiliary tool for the rapid diagnosis of meningeal carcinomatosis.

To report a case of onychocytic carcinoma in China. A 76-year-old female patient presented with a subnail mass on her left thumb for 3 years. Skin examination showed thickening of the left thumb plate whose surface was partially yellow and black in color, slight curving of the distal plate, obvious telangiectasia under the nail plate, and verrucous hyperplasia under the nail plate and at the lateral edge of the nail plate. Histopathological examination showed that the tumor grew and infiltrated downward in clumps, with obvious keratosis and necrosis in the center of the mass, and many "squamous vortex" structures, atypical cells, and pathological mitotic figures could be seen. Immunohistochemical study revealed that HK31 and HK85 were positively expressed in the squamous vortex region, K6/K16 was positively expressed in the basal part of the tumor, and HK75 was strongly expressed in the whole tumor region. The diagnosis of onychocytic carcinoma was confirmed. The patient received surgical excision combined with 3 sessions of photodynamic therapy, and no recurrence was observed during 18 months of follow-up.

Objective To explore the mechanism underlying the protective effect of Lonicerae japonicae flos(LJF)extract against doxorubicin(DOX)-induced liver injury in mice.Methods Network pharmacology methods were used to obtain the intersection genes between LJF targets and disease targets,based on which the protein-protein interaction(PPI)network was constructed using STRING database for screening the core targets using Cytoscape software.DAVID database was used for bioinformatics analysis,and the core components and core targets were verified using molecular docking study.In a mouse model of DOX-induced liver injury,the effect of LJF extract on liver pathologies,serum levels of ALT and AST,and hepatic expressions of HYP,ROS,TNF-α,IL-6,COL-IV and P53 proteins were evaluated using HE and Masson staining,ELISA,and Western blotting.Results We identified 12 core targets from 43 intersection genes involving cancer pathway,IL-17 signaling pathway,and TNF signaling pathways.Molecular docking study suggested that 10 core components of LJF could bind to different core targets.The mice with DOX-induced liver injury showed elevated serum AST and ALT levels with obvious liver injury and fibrosis,increased ROS content,and enhanced expressions of TNF-α,IL-6,HYP,COL-IV and P53 proteins in the liver tissue.All these changes in the mouse models were significantly alleviated by treatment with LJF extract,suggesting obviously lowered levels of oxidative stress,inflammation and fibrosis in the liver tissues.Conclusion LJF extract is capable of alleviating DOX-induced liver injury in mice by downregulating Trp53,TNF and IL-6 to reduce liver oxidative stress,inflammation and fibrosis.

ObjectiveTo investigate the effect of Yunkang oral liquid on postpartum kidney deficiency in mice. MethodPostpartum mice were randomized into model and low-dose （6 mL·kg^-1）， medium-dose （9 mL·kg^-1）， and high-dose （12 mL·kg^-1） Yunkang oral liquid groups. The mouse model of postpartum kidney deficiency was established by sleep deprivation combined with forced swimming. Another 9 female ICR mice were selected as the normal control group. The mice were administrated with Yunkang oral liquid during the period of modeling. The levels of estradiol （E₂）， progesterone （P）， follicle-stimulating hormone （FSH）， and luteinizing hormone （LH） in the serum were measured by the enzyme-linked immunosorbent assay. The morphological changes of ovaries and uterus were observed by hematoxylin-eosin （HE） staining， and the expression of transforming growth factor （TGF）-β and Smad2/3 was determined by immunohistochemistry and Western blotting. ResultThe mice in the model group showed prolonged estrous cycle， reduced voluntary activity， dorsal temperature， grip strength， and bone strength， and whitening tongue coating. Compared with the model group， Yunkang oral liquid shortened the estrous cycle， increased the voluntary activity， dorsal temperature， grip strength， and bone strength， and alleviated the whitening of tongue coating. Moreover， it elevated the E₂ and P levels and lowered the FSH and LH levels in the serum， decreased ovarian follicular atresia rate， promoted uterine repair， and down-regulated the expression of TGF-β and Smad2/3 in the ovarian and uterine tissues. ConclusionYunkang oral liquid can ameliorate postpartum kidney deficiency in mice by regulating the TGF-β/Smads signaling pathway.

Based on the theory of "mutual interference of clear and turbid Qi" in Huangdi Neijing（《黄帝内经》）， this study explored the pathogenesis of spleen and stomach diseases and the therapeutic effects of Banxia Xiexintang on them. It suggested that "mutual interference of clear and turbid Qi" represents a pathological state of Yin and Yang disturbance and imbalance in Qi circulation due to the mixture of clear and turbid Qi， which can elucidate the pathogenesis of spleen and stomach diseases. According to this theory， the pathogenesis of spleen and stomach diseases was summarized as Qi disorder in spleen and stomach， disharmony between Ying Qi and Wei Qi， and conflict between cold and heat. Banxia Xiexintang， as a crucial prescription for treating spleen and stomach diseases， achieves its therapeutic effects by dispersing stagnation with pungent flavor， descending adverse Qi with bitterness， regulating Ying Qi and Wei Qi， and harmonizing cold and heat. By regulating Qi circulation， balancing internal and external factors， and addressing deficiency and excess， it can rectify the pathological state of "mutual interference of clear and turbid Qi" of spleen and stomach diseases. Modern research reveals that Banxia Xiexintang can modulate gastrointestinal motility， restore mucosal immune barrier function of the digestive system， and exhibit optimal therapeutic effects when combined with both cold-cool and warm-hot medicines， aligning with its therapeutic role under the theory of "mutual interference of clear and turbid Qi". By delving into the essence of the "mutual interference of clear and turbid Qi" theory and exploring the pathogenesis of spleen and stomach diseases and the therapeutic effects of Banxia Xiexintang based on this theory， this study further elucidated the inherent connection between spleen and stomach diseases and the "mutual interference of clear and turbid Qi" theory， offering insights and theoretical references for the clinical diagnosis and treatment of spleen and stomach diseases.

Objective To explore the mechanism underlying the protective effect of Lonicerae japonicae flos(LJF)extract against doxorubicin(DOX)-induced liver injury in mice.Methods Network pharmacology methods were used to obtain the intersection genes between LJF targets and disease targets,based on which the protein-protein interaction(PPI)network was constructed using STRING database for screening the core targets using Cytoscape software.DAVID database was used for bioinformatics analysis,and the core components and core targets were verified using molecular docking study.In a mouse model of DOX-induced liver injury,the effect of LJF extract on liver pathologies,serum levels of ALT and AST,and hepatic expressions of HYP,ROS,TNF-α,IL-6,COL-IV and P53 proteins were evaluated using HE and Masson staining,ELISA,and Western blotting.Results We identified 12 core targets from 43 intersection genes involving cancer pathway,IL-17 signaling pathway,and TNF signaling pathways.Molecular docking study suggested that 10 core components of LJF could bind to different core targets.The mice with DOX-induced liver injury showed elevated serum AST and ALT levels with obvious liver injury and fibrosis,increased ROS content,and enhanced expressions of TNF-α,IL-6,HYP,COL-IV and P53 proteins in the liver tissue.All these changes in the mouse models were significantly alleviated by treatment with LJF extract,suggesting obviously lowered levels of oxidative stress,inflammation and fibrosis in the liver tissues.Conclusion LJF extract is capable of alleviating DOX-induced liver injury in mice by downregulating Trp53,TNF and IL-6 to reduce liver oxidative stress,inflammation and fibrosis.

Objective:To explore the clinical application value of third-generation sequencing technology in preimplantation genetic testing (PGT) of an ATR-X syndrome pedigree with chromosomal microduplication.Methods:The study selected a pedigree with a suspected ATR-X syndrome child at Assisted Reproductive Center of Jiangxi Maternal and Child Health Hospital in October 2022. After chromosome copy number variation sequencing (CNV Seq) detection, it was found that the female carried a 550 kb heterozygous microrepeat with unclear clinical significance in the Xq21.1 region, which involved partial sequences of the ATRX gene. The third generation long read sequencing technology was used to detect the female genome sequence, determine the physical location of the insertion of the above repeat into the genome, clarify the pathogenicity of the repeat, and obtain a single nucleotide polymorphism (SNP) haplotype linked to the above micro repeat inheritance. After the couple's full informed consent, PGT was performed to assist pregnancy. One haploid embryo without pathogenic microduplication was selected for transfer. To verify the consistency with PGT test results, amniocentesis prenatal diagnosis was performed in the second trimester after successful pregnancy, and the fetus was followed up after birth. Results:The results of the third generation long read sequencing and Sanger sequencing verification showed that the Xq21.1 microrepeat carried by the female was inserted into the genome chrX: 76804463-76804464 (GRCh37/hg19), which is an intra tandem repeat of the ATRX gene, and it is predicted that it may cause damage to the normal function of the ATRX protein. After one cycle of PGT treatment, 27 oocytes were obtained and 13 blastocysts were successfully developed after intracytoplasmic sperm injection (ICSI). Through genetic testing, it was found that two blastocysts were haploid embryos without carrying the aforementioned pathogenic microduplication. After thawing and transferring one of the blastocysts, the pregnancy was achieved, and the prenatal diagnosis results of amniocentesis in the second trimester were consistent with the PGT results. In November 2023, at 39 +5 weeks of pregnancy, a female live baby was delivered by natural delivery, and she is in good health. Conclusion:The third-generation sequencing technology has significant advantages in PGT detection of clinically ambiguous microreplicates with functional deficiency due to its long read length characteristics. It can not only determine the location of microreplicates inserted into the genome and determine their pathogenicity, but also obtain SNP haplotypes that are linked to the target mutation, thus preparing for subsequent embryo detection.