This paper proposes a two-stage method integrating visual foundation models （VFM） and diffusion models. The segment anything model （SAM） as VFM is combined with the SegRefiner diffusion model to construct the SAM-SegRefiner framework for automated segmentation of edema， inflammation， and thrombus regions in histopathological images of hemorrhoidal tissue， providing a reproducible technical tool for the objective quantification of pathological morphology and its application in traditional Chinese medicine （TCM） syndrome research. Trained and validated on multi-center retrospective data， the SAM-SegRefiner model achieved an average pixel accuracy of 95.32% and a mean intersection over union （mIoU） of 66.81% on an independent test set， significantly outperfor-ming comparative models such as U-Net， MixU-Net， and SAM-Med2D， and also demonstrating robust cross-center generalization capability. Furthermore， by correlating the quantitatively segmented results from the model with the patients' TCM syndrome types， the potential associations between pathomorphological features and TCM syndrome differentiation have been explored. The analysis revealed no statistically significant differences in the degree of inflammatory infiltration and thrombus formation among different syndrome types， suggesting a complex relationship between local pathological changes and systemic syndrome manifestations.

Objective:To screen the optimal machine learning model for predicting the recurrence condition of hepatocellular carcinoma (HCC) at different time points post-surgery, based on the cutoff value of the Ki67 positive proliferation index condition calculated from recurrence-free survival and combined with various clinical features.Methods:retrospective study included initially treated patients with solitary HCC who underwent radical surgery at the Fifth Medical Center of the PLA General Hospital from January 2013 to March 2023. Data included general clinical data, preoperative laboratory parameters, and surgical pathology information about the subjects. The postoperative recurrence status was assessed by querying the medical record system or by telephone follow-up. The Ki67 positive index cutoff value was determined by the X-tile software based on the patient's recurrence-free survival status and time analysis. Survival rates were calculated using the Kaplan-Meier method, and survival curves were plotted. The study population was randomly divided into training and testing groups in a 7:3 ratio using a computer-generated random number method. The minimum redundancy maximum relevance (mRMR) method was used for feature variable selection. Predictive models for postoperative HCC recurrence conditions in patients with HCC were constructed using random forest, support vector machine, logistic regression, and gradient boosting decision tree machine learning algorithms. Inter-group comparisons for continuous data were performed using the t-test or Mann-Whitney U test. Inter-group comparisons of enumeration data were performed using the Pearson χ2 test, continuity-corrected χ2 test, or Fisher's exact test. Results:The cutoff values for the Ki67 positivity index were 0.3 and 0.5 in 510 cases, with a follow-up time ranging from 1.2 to 11.4 years (median: 6.2 years). The recurrence-free survival time was between 1 and 135 months (median: 32 months), with recurrence-free survival rates post-surgery at 1, 2, 3, and 5 years were 87.5%, 77.1%, 61.2%, and 54.5%, respectively. The top five variables predicted HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years, in accordance with information obtained by the mRMR screen out. The Ki67 positivity index screened a successfully constructed machine learning model to predict HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years. The machine learning model based on the gradient boosting decision tree algorithm had the best prediction performance among them (areas under the receiver operating characteristic curves for predicting HCC recurrence within six months in the training and validation sets were 0.996 and 0.946, and accuracies were 0.972 and 0.935, respectively).Conclusion:A machine learning model was successfully constructed using the Ki67 positivity index combined with four readily available clinical features to predict HCC recurrence. The machine learning model based on the gradient boosting decision tree algorithm demonstrated the best performance in terms of predicting HCC recurrence within six months after surgery.

Objective:To investigate the clinical characteristics of patients with high-myopia macular hole retinal detachment (MHRD) combined with choroidal detachment and to preliminarily analyze factors associated with postoperative hole closure.Methods:A retrospective clinical case series study. A total of 68 patients with high myopia (68 eyes) with MHRD diagnosed by Department of Ophthalmology, Peking University People’s Hospital from January 2019 to April 2024 were included in this study. Among them, there were 14 males (14 eyes) and 54 females (54 eyes). The mean age was (61.10±9.66) years. All eyes were treated with pars plana vitrectomy (PPV) combined with silicone oil or gas filling. Best corrected visual acuity (BCVA), intraocular pressure, and B-mode ultrasonography were performed. The BCVA test was performed using the Snellen visual acuity chart, which was statistically converted to logarithm of the minimum angle of resolution (logMAR) visual acuity. The range of choroidal detachment was defined according to the number of involved quadrants observed in B-mode ultrasound or surgery, which was divided into 1 to 4 quadrants. Axial length (AL) was measured under retinal reattachment. In 68 eyes, there were 17 eyes with choroidal detachment and 51 eyes without choroidal detachment, respectively. There were 17 eyes with choroidal detachment, and the detachment range involved 1, 2, 2 and 12 eyes in 1, 2, 3 and 4 quadrants, respectively. During operation, 13% C 3F 8 was filled in 2 eyes, all of which were not complicated with choroidal detachment. 66 eyes were filled with silicone oil. According to whether the patients were complicated with choroidal detachment, the patients were divided into the group without choroidal detachment and the group with choroidal detachment. Independent sample t test, Welch two-sample t test or Mann-Whitney U test were used for comparison between groups. Generalized linear regression and logistic regression were used to analyze the relationship between the aperture size of postoperative unclosed holes and the closed hole after surgery and clinical factors. Results:At 3 months after surgery, the logMAR BCVA of the affected eye was 1.29±0.43, with a preoperative to postoperative difference ranging from -1.60 to 0.70 (-0.51±0.51) logMAR units. The AL ranged from 26.6 to 34.3 (29.60±2.12) mm. Among 68 eyes, macular hole of 37 (54.4%, 37/68) eyes were open and 31 (45.6%, 31/68) eyes were closed, respectively. The hole diameter of the open eye was (753±424) μm. There was no significant difference in age, course of disease and AL between the two groups ( W=412.0, 477.5, 427.0; P>0.05). Before operation, BCVA in patients with choroidal detachment was worse ( W=257.5) and intraocular pressure was lower ( t=4.051) in patients with choroidal detachment compared with those without choroidal detachment, with statistical significance ( P<0.05). At 3 months after surgery, BCVA in patients with choroidal detachment was significantly worse than that in patients without choroidal detachment, with statistical significance ( W=284.0, P<0.05). There were no significant differences in logMAR BCVA difference ( t=0.616) and macular hole closure rate ( χ 2=0.000) before and after surgery ( P>0.05). The reoperation rate of retinal detachment due to persistent or recurrent retinal detachment was significantly higher in the group with choroid detachment than in the group without choroid detachment, and the difference was statistically significant (odds ratio=6.424, P<0.05). Logistic regression analysis showed that young age was significantly correlated with macular hole closure failure after surgery ( β=0.077, P=0.015). There was no correlation between AL, duration of disease, BCVA before surgery, intraocular pressure, wether combined with choroid detachment range and postoperative hole closure ( β=-0.072, 0.000, 0.672, -0.085, -0.391; P>0.05). Conclusions:Concomitant choroidal detachment adversely affected on both pre-operative and post-operative visual acuity in high myopia MHRD. It is closely associated with the risk of recurrent retinal detachment and the needs of multiple operations, but has no significant effect on hole closure rate. Lower age of onset may be a risk factor for macular hole closure.

Objective:To observe and analyze the correlation between ectopic foveal inner layer (EIFL) and the EIFL-based idiopathic epiretinal membrane (ERM) staging system and the anatomic and functional prognosis of ERM eyes post pars plana vitrectomy (PPV).Methods:A retrospective study. From January 1, 2020 to October 30, 2023, 345 eyes of 330 patients diagnosed with idiopathic ERM in Department of Ophthalmology of Peking University People's Hospital and treated with standard transciliary flat three-channel 25G PPV combined with ERM and internal limiting membrane exfoliation were included in the study. Among them, 96 were males (111 eyes) and 234 were females (234 eyes). The mean age was (66.8±7.7) years. All study eyes received standard three-port 25G PPV combined with ERM and internal limiting membrane peeling. All study eyes underwent best corrected visual acuity (BCVA) and optical coherence tomography (OCT) examinations. BCVA was performed using a standard logarithmic visual acuity chart and converted to logarithm of the minimum angle of resolution visual acuity for statistical analysis. EIFL thickness and central foveal thickness (CFT) on OCT were measured. ERM eyes were grouped into stage Ⅰ, Ⅱ, Ⅲ and Ⅳ according to ERM staging scheme based on EIFL; disorganization of the retinal inner layers (DRIL) of study eyes were assessed and grouped into no, mild and severe groups. The correlation between ERM staging as well as EIFL thickness and the anatomical and functional prognosis 6 months post-PPV were analyzed.Results:Among 345 study eyes, 12, 87, 174 and 72 eyes were stage Ⅰ-Ⅳ ERM respectively, 63 with no DRIL, 216 with mild DRIL and 66 with severe DRIL. Among the 153 eyes with macular edema, the edema subsided in 66 eyes (43.1%, 66/153) 6 months after the operation. Eighty-seven eyes (56.9%, 87/153) did not regress. The edema subsided 6 months after the operation was not significantly correlated with the ERM stage before the operation ( χ2=3.331, R=?0.145, P=0.304) or the degree of DRIL ( χ2=0.655, R=?0.108, P=0.445). The results of the correlation analysis showed that logMAR BCVA 6 months after the surgery was positively correlated with the degree of DRIL before the surgery ( Tau-b=0.236), ERM stage ( Tau-b=0.194), CFT ( r=0.383), and EIFL thickness ( r=0.317) ( P<0.05). There was no significant correlation with the thickness of the outer nuclear layer before the operation ( r=0.004, P>0.05). Preoperative ERM stage ( Tau-b=0.303, P<0.001) and DRIL severity ( Tau-b= 0.238, P=0.001) were positively correlated with CFT at 6 months after surgery. Conclusion:The ERM stage and EIFL thickness before the operation are positively correlated with logMAR BCVA and CFT 6 months after the operation.

Objective:To screen the optimal machine learning model for predicting the recurrence condition of hepatocellular carcinoma (HCC) at different time points post-surgery, based on the cutoff value of the Ki67 positive proliferation index condition calculated from recurrence-free survival and combined with various clinical features.Methods:retrospective study included initially treated patients with solitary HCC who underwent radical surgery at the Fifth Medical Center of the PLA General Hospital from January 2013 to March 2023. Data included general clinical data, preoperative laboratory parameters, and surgical pathology information about the subjects. The postoperative recurrence status was assessed by querying the medical record system or by telephone follow-up. The Ki67 positive index cutoff value was determined by the X-tile software based on the patient's recurrence-free survival status and time analysis. Survival rates were calculated using the Kaplan-Meier method, and survival curves were plotted. The study population was randomly divided into training and testing groups in a 7:3 ratio using a computer-generated random number method. The minimum redundancy maximum relevance (mRMR) method was used for feature variable selection. Predictive models for postoperative HCC recurrence conditions in patients with HCC were constructed using random forest, support vector machine, logistic regression, and gradient boosting decision tree machine learning algorithms. Inter-group comparisons for continuous data were performed using the t-test or Mann-Whitney U test. Inter-group comparisons of enumeration data were performed using the Pearson χ2 test, continuity-corrected χ2 test, or Fisher's exact test. Results:The cutoff values for the Ki67 positivity index were 0.3 and 0.5 in 510 cases, with a follow-up time ranging from 1.2 to 11.4 years (median: 6.2 years). The recurrence-free survival time was between 1 and 135 months (median: 32 months), with recurrence-free survival rates post-surgery at 1, 2, 3, and 5 years were 87.5%, 77.1%, 61.2%, and 54.5%, respectively. The top five variables predicted HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years, in accordance with information obtained by the mRMR screen out. The Ki67 positivity index screened a successfully constructed machine learning model to predict HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years. The machine learning model based on the gradient boosting decision tree algorithm had the best prediction performance among them (areas under the receiver operating characteristic curves for predicting HCC recurrence within six months in the training and validation sets were 0.996 and 0.946, and accuracies were 0.972 and 0.935, respectively).Conclusion:A machine learning model was successfully constructed using the Ki67 positivity index combined with four readily available clinical features to predict HCC recurrence. The machine learning model based on the gradient boosting decision tree algorithm demonstrated the best performance in terms of predicting HCC recurrence within six months after surgery.

Objective:To investigate the clinical characteristics of patients with high-myopia macular hole retinal detachment (MHRD) combined with choroidal detachment and to preliminarily analyze factors associated with postoperative hole closure.Methods:A retrospective clinical case series study. A total of 68 patients with high myopia (68 eyes) with MHRD diagnosed by Department of Ophthalmology, Peking University People’s Hospital from January 2019 to April 2024 were included in this study. Among them, there were 14 males (14 eyes) and 54 females (54 eyes). The mean age was (61.10±9.66) years. All eyes were treated with pars plana vitrectomy (PPV) combined with silicone oil or gas filling. Best corrected visual acuity (BCVA), intraocular pressure, and B-mode ultrasonography were performed. The BCVA test was performed using the Snellen visual acuity chart, which was statistically converted to logarithm of the minimum angle of resolution (logMAR) visual acuity. The range of choroidal detachment was defined according to the number of involved quadrants observed in B-mode ultrasound or surgery, which was divided into 1 to 4 quadrants. Axial length (AL) was measured under retinal reattachment. In 68 eyes, there were 17 eyes with choroidal detachment and 51 eyes without choroidal detachment, respectively. There were 17 eyes with choroidal detachment, and the detachment range involved 1, 2, 2 and 12 eyes in 1, 2, 3 and 4 quadrants, respectively. During operation, 13% C 3F 8 was filled in 2 eyes, all of which were not complicated with choroidal detachment. 66 eyes were filled with silicone oil. According to whether the patients were complicated with choroidal detachment, the patients were divided into the group without choroidal detachment and the group with choroidal detachment. Independent sample t test, Welch two-sample t test or Mann-Whitney U test were used for comparison between groups. Generalized linear regression and logistic regression were used to analyze the relationship between the aperture size of postoperative unclosed holes and the closed hole after surgery and clinical factors. Results:At 3 months after surgery, the logMAR BCVA of the affected eye was 1.29±0.43, with a preoperative to postoperative difference ranging from -1.60 to 0.70 (-0.51±0.51) logMAR units. The AL ranged from 26.6 to 34.3 (29.60±2.12) mm. Among 68 eyes, macular hole of 37 (54.4%, 37/68) eyes were open and 31 (45.6%, 31/68) eyes were closed, respectively. The hole diameter of the open eye was (753±424) μm. There was no significant difference in age, course of disease and AL between the two groups ( W=412.0, 477.5, 427.0; P>0.05). Before operation, BCVA in patients with choroidal detachment was worse ( W=257.5) and intraocular pressure was lower ( t=4.051) in patients with choroidal detachment compared with those without choroidal detachment, with statistical significance ( P<0.05). At 3 months after surgery, BCVA in patients with choroidal detachment was significantly worse than that in patients without choroidal detachment, with statistical significance ( W=284.0, P<0.05). There were no significant differences in logMAR BCVA difference ( t=0.616) and macular hole closure rate ( χ 2=0.000) before and after surgery ( P>0.05). The reoperation rate of retinal detachment due to persistent or recurrent retinal detachment was significantly higher in the group with choroid detachment than in the group without choroid detachment, and the difference was statistically significant (odds ratio=6.424, P<0.05). Logistic regression analysis showed that young age was significantly correlated with macular hole closure failure after surgery ( β=0.077, P=0.015). There was no correlation between AL, duration of disease, BCVA before surgery, intraocular pressure, wether combined with choroid detachment range and postoperative hole closure ( β=-0.072, 0.000, 0.672, -0.085, -0.391; P>0.05). Conclusions:Concomitant choroidal detachment adversely affected on both pre-operative and post-operative visual acuity in high myopia MHRD. It is closely associated with the risk of recurrent retinal detachment and the needs of multiple operations, but has no significant effect on hole closure rate. Lower age of onset may be a risk factor for macular hole closure.

Objective:To observe and analyze the correlation between ectopic foveal inner layer (EIFL) and the EIFL-based idiopathic epiretinal membrane (ERM) staging system and the anatomic and functional prognosis of ERM eyes post pars plana vitrectomy (PPV).Methods:A retrospective study. From January 1, 2020 to October 30, 2023, 345 eyes of 330 patients diagnosed with idiopathic ERM in Department of Ophthalmology of Peking University People's Hospital and treated with standard transciliary flat three-channel 25G PPV combined with ERM and internal limiting membrane exfoliation were included in the study. Among them, 96 were males (111 eyes) and 234 were females (234 eyes). The mean age was (66.8±7.7) years. All study eyes received standard three-port 25G PPV combined with ERM and internal limiting membrane peeling. All study eyes underwent best corrected visual acuity (BCVA) and optical coherence tomography (OCT) examinations. BCVA was performed using a standard logarithmic visual acuity chart and converted to logarithm of the minimum angle of resolution visual acuity for statistical analysis. EIFL thickness and central foveal thickness (CFT) on OCT were measured. ERM eyes were grouped into stage Ⅰ, Ⅱ, Ⅲ and Ⅳ according to ERM staging scheme based on EIFL; disorganization of the retinal inner layers (DRIL) of study eyes were assessed and grouped into no, mild and severe groups. The correlation between ERM staging as well as EIFL thickness and the anatomical and functional prognosis 6 months post-PPV were analyzed.Results:Among 345 study eyes, 12, 87, 174 and 72 eyes were stage Ⅰ-Ⅳ ERM respectively, 63 with no DRIL, 216 with mild DRIL and 66 with severe DRIL. Among the 153 eyes with macular edema, the edema subsided in 66 eyes (43.1%, 66/153) 6 months after the operation. Eighty-seven eyes (56.9%, 87/153) did not regress. The edema subsided 6 months after the operation was not significantly correlated with the ERM stage before the operation ( χ2=3.331, R=?0.145, P=0.304) or the degree of DRIL ( χ2=0.655, R=?0.108, P=0.445). The results of the correlation analysis showed that logMAR BCVA 6 months after the surgery was positively correlated with the degree of DRIL before the surgery ( Tau-b=0.236), ERM stage ( Tau-b=0.194), CFT ( r=0.383), and EIFL thickness ( r=0.317) ( P<0.05). There was no significant correlation with the thickness of the outer nuclear layer before the operation ( r=0.004, P>0.05). Preoperative ERM stage ( Tau-b=0.303, P<0.001) and DRIL severity ( Tau-b= 0.238, P=0.001) were positively correlated with CFT at 6 months after surgery. Conclusion:The ERM stage and EIFL thickness before the operation are positively correlated with logMAR BCVA and CFT 6 months after the operation.

Objective:To evaluate the therapeutic effect of ultrasound-guided microwave ablation (MWA) in the treatment of breast cancer liver metastases (BCLM), and to select the beneficiaries.Methods:A retrospective study was performed in 63 patients with BCLM who underwent ultrasound-guided microwave ablation in Department of Interventional Ultrasound, the Fifth Medical Center of Chinese PLA General Hospital, and had detailed follow-up data from February 2010 to October 2022. General patient data and laboratory results were collected. Univariate and multivariate Cox regression analyses were performed to determine the independent influencing factors for prognosis in patients with BCLM. A nomogram model was established to predict the survival rates of patients after operation.Calibration curve and decision curve were plotted to evaluate the calibration degree and the clinical benefit of the model, respectively. Time-dependent ROC curve was plotted to evaluate the predicting ability of the model. All patients were divided into high-risk group and low-risk group according to prognostic index. Survival curves were plotted to compare differences in survival between the two groups.Results:Multivariate Cox regression analysis showed that tumor burden score(TBS) and the presence of extrahepatic metastasis before MWA were independent influencing factors for prognosis of BCLM(all P<0.001). The combined prediction model was established based on TBS, extrahepatic metastasis and axillary nodal metastases. The areas under the 1-year, 2-year and 3-year ROC curves were 0.849, 0.855 and 0.878, respectively, suggesting that the model had good predicting ability. The calibration curve showed that the model had good calibration degree, while the decision curve showed that the model had good clinical practicability. The median survival time between the two groups was statistically significant[13.5 (9, 20) months vs 63.0 (39, 140) months, P<0.001]. Conclusions:Low-risk patients with BCLM gain significant clinical benefits after microwave ablation treatment, while high-risk patients do not exhibit a noticeable survival advantage.

Objective:To investigate the spatial distribution pattern of local tumor progression (LTP) for hepatocellular carcinoma (HCC) ≤5 cm after microwave ablation.Methods:A retrospective analysis was performed on 169 HCCs with matched MRI before and after ablation from December 2009 to December 2019. A tumor MRI was reconstructed using three-dimensional visualization technology. LTP was classified as contact or non-contact, early or late stage, according to whether LTP was in contact with the edge of the ablation zone and the occurrence time (24 months). The tumor-surrounded area was divided into eight quadrants by using the eight-quadrant map method. An analysis was conducted on the spatial correlation between the quadrant where the ablative margin (AM) safety boundary was located and the quadrant where different types of LTP occurred. The t-test, or rank-sum test, was used for the measurement data. 2-test for count data was used to compare the difference between the two groups.Results:The AM quadrant had a distribution of 54.4% LTP, 64.2% early LTP stage, and 69.1% contact LTP, suggesting this quadrant was much more concentrated than the other quadrants ( P ?

Objectives:To investigate the clinical characteristics and prognosis of bone metastasis of gastric cancer, analyze the influencing factors of bone metastasis and the effects of different treatment methods, and provide a basis for early detection and treatment optimization of bone metastasis of gastric cancer.Methods:A total of 142 gastric cancer patients with bone metastasis admitted to the First Hospital of Lanzhou University from January 2011 to December 2021 were enrolled, including 60 cases of simple bone metastasis and 82 cases of bone metastasis combined with extraosseous metastasis. 142 patients with stage Ⅲgastric cancer without distant metastasis and 142 gastric cancer patients with visceral metastasis admitted to this hospital during the same period were also enrolled for comparison. Logistic regression analysis was used to determine the influencing factors of bone metastasis, and the Cox proportional hazards regression model was used to evaluate the influencing factors of overall survival (OS) of patients with bone metastasis.Results:Among the 142 patients with bone metastasis, poorly differentiated adenocarcinoma was the main type (123 cases), and 45 patients had simultaneous bone metastasis. Rib metastasis (100 cases), spine metastasis (88 cases), and pelvis metastasis (84 cases) were more common. A total of 110 patients had multiple bone metastasis, and 82 patients had extraosseous metastasis. Results of the stage Ⅲ gastric cancer group, the visceral metastasis group, the bone metastasis group, and the bone metastasis with extraosseous metastasis group were compared. There were significant differences in age, degree of differentiation, Borrmann type, alkaline phosphatase, lactate dehydrogenase, serum calcium, alanine aminotransferase, aspartate aminotransferase, creatine kinase isoenzyme, lymphocyte, hemoglobin, platelet, CEA, CA19-9, and CA724 (all P<0.05). Multivariate logistic regression analysis showed that Borrmann type was an independent protective factor of bone metastasis of gastric cancer (type 3: OR=0.07, 95% CI: 0.01-0.64, P=0.018). Alkaline phosphatase ( OR=2.54, 95% CI: 1.07-6.01, P=0.034), serum calcium ( OR=2.71, 95% CI: 1.15-6.41, P=0.023), creatine kinase isoenzyme ( OR=16.33, 95% CI: 1.83-145.58, P=0.012), platelet ( OR=10.08, 95% CI:1.89-53.85, P=0.007), and CA19-9 ( OR=2.40, 95% CI: 1.14-5.05, P=0.021) were independent risk factors of bone metastasis of gastric cancer. The median OS of the stage Ⅲ gastric cancer group, the visceral metastasis group, the bone metastasis group, and the bone metastasis with extrabony group were 47, 13, 18, and 6 months, respectively, and the difference was statistically significant ( P<0.001). The median OS of patients with bone metastasis only who underwent primary tumor surgery was 33 months, better than 6 months of patients without surgery ( P=0.048). Multivariate Cox regression analysis showed that extraosseous metastasis ( HR=2.45, 95% CI: 1.56-3.85, P<0.001) and decreased hemoglobin ( HR=1.54, 95% CI: 1.02-2.34, P=0.042) were independent risk factors of OS of gastric cancer patients with bone metastasis. Conclusions:The prognosis of gastric cancer patients with bone metastasis alone is significantly better than that of other stage Ⅳ patients. For such patients, surgery on the primary site combined with chemotherapy after full evaluation may prolong the survival time.