Intravitreal anti-vascular endothelial growth factor(anti-VEGF)therapy has been widely used, but the variability in its therapeutic efficacy limits individualized treatment. In recent years, the application of artificial intelligence(AI)has opened up new avenues for personalized treatment response prediction, and its core branches include machine learning(ML)and deep learning(DL). This review systematically retrieved and analyzed 41 relevant studies published up to April 2025. Comprehensive analysis reveals that AI predictive models are evolving from forecasting single endpoints(such as visual acuity or central retinal thickness)to integrating multi-dimensional endpoints(encompassing anatomical, functional, and treatment demand parameters)and generating predictive imaging outputs. In terms of technical approaches, DL models(28 studies, accounting for 68.3%)dominate this field due to their robust image interpretation capabilities, while ML models(10 studies, 24.4%)retain significant value in the analysis of structured clinical data. Cross-disease comparisons indicate that research efforts are most concentrated on age-related macular degeneration(ARMD)and diabetic macular edema(DME), with shared conceptual frameworks for model construction, yet distinct anatomical and functional indicators are prioritized for each disease. Currently, the field confronts several key challenges, including insufficient prospective clinical validation, limited model interpretability(the “black box problem”), and a scarcity of high-quality multi-center datasets. Moving forward, it is imperative to advance real-world validation and develop explainable AI techniques to expedite the clinical translation of these predictive models.

Objective:To explore the diagnostic value of preoperative diffusion weighted imaging (DWI) histogram parameters in the depth of invasion of early rectal cancer.Methods:A total of 180 patients with early rectal cancer admitted to 904th Hospital of Joint Logistics Support Force of Chinese People's Liberation Army from August 2020 to August 2024 were selected as the study objects. Patients were divided into intramucosal cancer group ( n=102) and submucosal cancer group ( n=78) according to the depth of tumor invasion. The general data of the two groups were compared. The intraclass correlation coefficient (ICC) was used to analyze the consistency of DWI histogram parameters extracted by the two radiologists, and the differences between the two groups were compared. Receiver operator characteristic (ROC) curve was used to analyze the predictive value of each parameter to the depth of tumor invasion. Multivariate logistic regression was used to analyze the independent influencing factors of invasion depth, and a predictive model was constructed. The ROC curve was drawn to analyze the predictive value of the model for tumor invasion depth, and the Hosmer-Lemeshow test was used to analyze the goodness of fit of the model. Results:There were statistically significant differences in age ( t=8.15, P<0.001), maximum tumor diameter ( χ2=29.29, P<0.001), endoscopic type ( χ2=20.96, P<0.001), histological type ( χ2=24.93, P<0.001) and differentiation degree ( χ2=73.35, P<0.001) between intramucosal cancer group and submucosal cancer group. The mean, variance, skewness, kurtosis, the 1 st, 10 th, 50 th, 90 th, and 99 th percentiles of the histogram parameters of DWI had good consistency (all ICC>0.75). There were statistically significant differences in the mean ( t=5.69, P<0.001), variance ( t=9.75, P<0.001), skewness ( t=10.88, P<0.001), kurtosis ( t=10.06, P<0.001), the 1 st percentile ( t=3.43, P<0.001), 10 th percentile ( t=3.59, P<0.001), 50 th percentile ( t=9.97, P<0.001), 90 th percentile ( t=4.63, P<0.001), and 99 th percentile ( t=2.44, P=0.016) of the DWI histogram parameters between the intramucosal cancer group and the submucosal cancer group. ROC curve analysis results showed that mean [area under the curve (AUC) =0.77], variance (AUC=0.88), skewness (AUC=0.88), kurtosis (AUC=0.78), 50 th percentile (AUC=0.86) and 90 th percentile (AUC=0.82) had certain diagnostic value for submucous cancer. Multivariate analysis showed that age ( OR=9.98, 95% CI: 1.10-90.70, P=0.041), maximum tumor diameter ( OR=7.36, 95% CI: 1.08-50.23, P=0.042), and differentiation degree ( OR=19.88, 95% CI: 1.21-327.92, P=0.037), variance ( OR=16.24, 95% CI: 2.26-116.68, P=0.006), skewness ( OR=21.13, 95% CI: 2.80-59.61, P=0.003), 1 st percentile ( OR=9.78, 95% CI: 1.17-81.76, P=0.035) were independent factors in predicting tumor invasion depth in patients with early rectal cancer. The predictive model based on the above indicators was logit ( P) =1.51+2.30×age+2.00×maximum tumor diameter+2.99×differentiation degree+2.79×variance+3.05×skewness+ 2.28×the 1 st percentile. ROC curve analysis showed that the predictive model had an AUC of 0.97 (95% CI: 0.95-0.99) for judging the occurrence of submucosal cancer in patients with early rectal cancer, the sensitivity was 0.95, and the specificity was 0.88. The Hosmer-Lemeshow test results showed that the goodness of fit of the model was ideal ( P=0.823) . Conclusions:Age, maximum tumor diameter, differentiation degree, variance, skewness, and the 1 st percentile are independent factors in predicting tumor invasion depth in patients with early rectal cancer. The predictive model constructed based on these factors can effectively predict the risk of submucosal cancer in patients with early rectal cancer.

OBJECTIVES: To explore the association between GPSM2 expression level and gastric cancer progression and analyze the functional pathways and action mechanism of GPSM2. METHODS: We analyzed GPSM2 expression levels in gastric cancer tumors based on data from the GEPIA database and the clinical data of 109 patients. Public databases enrichment analysis were used to assess the impact of GPSM2 expression level on survival outcomes and the functional pathways and action mechanism of GPSM2. We further observed the effects of GPSM2 knockdown and overexpression on proliferation, migration and apoptosis of MGC803 cells using CCK-8 assay, colony formation assay, flow cytometry and immunoblotting and on the growth of MGC803 cell xenografts in nude mice. RESULTS: Bioinformatic analysis and immunohistochemical staining of the clinical specimens both revealed high GPSM2 expressions in gastric cancer (P<0.01). A high GPSM2 expression was significantly correlated with T3-4 stages, N2-3 stages, a carcinoembryonic antigen (CEA) level ≥5 μg/L, and a carbohydrate antigen (CA) 19-9 level ≥37 kU/L (P<0.05). Cox regression analysis identified high GPSM2 expression as an independent risk factor affecting 5-year survival of the patients (P<0.05). Gene ontology (GO) analysis suggested that GPSM2 was involved in cell cycle regulation. In MGC803 cells, GPSM2 overexpression significantly promoted cell proliferation and G1/S transition and xenograft growth in nude mice. KEGG pathway enrichment analysis indicated that GPSM2 executed its biological functions by regulating the p53 signaling pathway, which was confirmed by the results of immunoblotting experiments showing suppression of p53 signaling pathway activity in GPSM2-over expressing MGC803 cells. CONCLUSIONS GPSM2 is highly expressed in gastric cancer to affect patient prognosis by promoting tumor cell proliferation and G1/S transition possibly via inhibiting the p53 pathway.
Stomach Neoplasms/metabolism* ; Humans ; Cell Proliferation ; Prognosis ; Animals ; Mice, Nude ; Cell Line, Tumor ; Mice ; Apoptosis ; Tumor Suppressor Protein p53/metabolism* ; Cell Movement

OBJECTIVES: To analyze MYO1B expression in gastric cancer, its association with long-term prognosis and its role in regulating biological behaviors of gastric cancer cells. METHODS: We analyzed MYO1B expression in gastric cancer and its correlation with tumor grade, tumor stage, and patient survival using the Cancer Public Database. We also examined MYO1B expression with immunohistochemistry in gastric cancer and paired adjacent tissues from 105 patients receiving radical surgery and analyzed its correlation with cancer progression and postoperative 5-year survival of the patients. GO and KEGG enrichment analyses were used to explore the biological functions of MYO1B and the key pathways. In cultured gastric cancer cells, we examined the changes in cell proliferation, migration and invasion following MYO1B overexpression and knockdown. RESULTS: Data from the Cancer Public Database showed that MYO1B expression was significantly higher in gastric cancer tissues than in normal tissues with strong correlations with tumor grade, stage and patient prognosis (P<0.05). In the clinical tissue samples, MYO1B was significantly overexpressed in gastric cancer tissues in positive correlation with Ki67 expression (r=0.689, P<0.05) and the parameters indicative of gastric cancer progression (CEA ≥5 μg/L, CA19-9 ≥37 kU/L, G3-4, T3-4, and N2-3) (P<0.05). Kaplan-Meier analysis and multivariate Cox regression analysis suggested that high MYO1B expression was associated with decreased postoperative 5-year survival and was an independent risk factor (HR: 3.522, 95%CI: 1.783-6.985, P<0.05). MYO1B expression level was a strong predictor of postoperative survival (cut-off value: 3.11, AUC: 0.753, P<0.05). GO and KEGG analyses suggested that MYO1B may regulate cell migration and the mTOR signaling pathway. In cultured gastric cancer cells, MYO1B overexpression significantly enhanced cell proliferation, migration, and invasion and promoted the phosphorylation of Akt and mTOR. CONCLUSIONS High MYO1B expression promotes proliferation, migration and invasion of gastric cancer cells and is correlated with poor patient prognosis.
Humans ; Stomach Neoplasms/metabolism* ; Cell Proliferation ; Prognosis ; Cell Movement ; Myosin Type I/genetics* ; Neoplasm Invasiveness ; Cell Line, Tumor ; Female ; Male

Objective:To explore the value of Alberta stroke program early CT Score (ASPECTS) regional net water intake (NWU) based on plain CT in evaluating the neurological outcome of patients with acute large vessel occlusive stroke (ALVOS) after mechanical thrombectomy.Methods:The study was a prospective cross-sectional study. The clinical and imaging data of patients with ALVOS who underwent mechanical thrombectomy in the 904 Hospital of the PLA Joint Service Support Force from June 2022 to June 2024 were prospectively collected. Clinical data analysis included age, gender distribution, National Institutes of Health Stroke Scale (NIHSS) score and Glasgow Coma Scale (GCS) score at admission. All patients underwent plain CT and CT angiography (CTA). Automated processing software quantified ischemic brain edema by calculating ASPECTS and measuring the ASPECTS regional NWU rate (CT-ASPECTS-NWU) based on plain CT. Cerebral collateral circulation was assessed using CTA images. According to the modified Rankin Scale score of patients with mechanical thrombectomy 90 days after the telephone follow-up, 0-2 points were defined as good neurological outcome, and 3-6 points were defined as poor neurological outcome. Independent samples t test, Mann-Whitney U test and χ2 test were used to analyze the differences of clinical and imaging indicators between patients with good and poor neurological outcome, indicators with statistically significant differences were included in multivariate logistic regression analysis to screen out the independent influencing factors that predicted the neurological outcome of patients with ALVOS after mechanical thrombectomys. The efficacy of related indicators to predict neurological outcomes after mechanical thrombectomy in patients with ALVOS was evaluated using receiver operating characteristic (ROC) curve analysis, with the area under the curve (AUC) calculated. Results:A total of 122 patients with ALVOS were included, including 101 patients with good neurological outcome and 21 patients with poor neurological outcome after mechanical thrombectomy. There were statistically significant differences in preoperative GCS score, collateral circulation status, NIHSS score at admission, preoperative ASPECTS and CT-ASPECTS-NWU between patients with good neurological outcome and patients with poor neurological outcome ( P<0.05). Multivariate logistic regression analysis showed that the status of collateral circulation ( OR=3.450, 95% CI 1.158-10.277, P=0.026), preoperative ASPECTS ( OR=0.510, 95% CI 0.274-0.952, P=0.034) and CT-ASPECTS-NWU ( OR=2.131, 95% CI 1.301-3.493, P=0.003) were independent predictors of neurological outcome in patients with ALVOS after mechanical thrombectomy. ROC curve analysis showed that CT-ASPECTS-NWU had the highest predictive value, with an AUC of 0.881, a sensitivity of 85.7%, a specificity of 85.1%, and an optimal cutoff value of 7.55. Conclusion:CT-ASPECTS-NWU demonstrates high diagnostic value in evaluating the neurological outcome of patients with ALVOS after mechanical thrombectomy, and can provide an objective imaging biomarker to guide clinical decision-making.

Objective:To explore the value of Alberta stroke program early CT Score (ASPECTS) regional net water intake (NWU) based on plain CT in evaluating the neurological outcome of patients with acute large vessel occlusive stroke (ALVOS) after mechanical thrombectomy.Methods:The study was a prospective cross-sectional study. The clinical and imaging data of patients with ALVOS who underwent mechanical thrombectomy in the 904 Hospital of the PLA Joint Service Support Force from June 2022 to June 2024 were prospectively collected. Clinical data analysis included age, gender distribution, National Institutes of Health Stroke Scale (NIHSS) score and Glasgow Coma Scale (GCS) score at admission. All patients underwent plain CT and CT angiography (CTA). Automated processing software quantified ischemic brain edema by calculating ASPECTS and measuring the ASPECTS regional NWU rate (CT-ASPECTS-NWU) based on plain CT. Cerebral collateral circulation was assessed using CTA images. According to the modified Rankin Scale score of patients with mechanical thrombectomy 90 days after the telephone follow-up, 0-2 points were defined as good neurological outcome, and 3-6 points were defined as poor neurological outcome. Independent samples t test, Mann-Whitney U test and χ2 test were used to analyze the differences of clinical and imaging indicators between patients with good and poor neurological outcome, indicators with statistically significant differences were included in multivariate logistic regression analysis to screen out the independent influencing factors that predicted the neurological outcome of patients with ALVOS after mechanical thrombectomys. The efficacy of related indicators to predict neurological outcomes after mechanical thrombectomy in patients with ALVOS was evaluated using receiver operating characteristic (ROC) curve analysis, with the area under the curve (AUC) calculated. Results:A total of 122 patients with ALVOS were included, including 101 patients with good neurological outcome and 21 patients with poor neurological outcome after mechanical thrombectomy. There were statistically significant differences in preoperative GCS score, collateral circulation status, NIHSS score at admission, preoperative ASPECTS and CT-ASPECTS-NWU between patients with good neurological outcome and patients with poor neurological outcome ( P<0.05). Multivariate logistic regression analysis showed that the status of collateral circulation ( OR=3.450, 95% CI 1.158-10.277, P=0.026), preoperative ASPECTS ( OR=0.510, 95% CI 0.274-0.952, P=0.034) and CT-ASPECTS-NWU ( OR=2.131, 95% CI 1.301-3.493, P=0.003) were independent predictors of neurological outcome in patients with ALVOS after mechanical thrombectomy. ROC curve analysis showed that CT-ASPECTS-NWU had the highest predictive value, with an AUC of 0.881, a sensitivity of 85.7%, a specificity of 85.1%, and an optimal cutoff value of 7.55. Conclusion:CT-ASPECTS-NWU demonstrates high diagnostic value in evaluating the neurological outcome of patients with ALVOS after mechanical thrombectomy, and can provide an objective imaging biomarker to guide clinical decision-making.

Objective To explore the correlation of CPNE3 expression with long-term prognosis of patients with gastric cancer(GC)and the possible mechanism.Methods We retrospectively collected the data of 104 GC patients undergoing radical surgery in our hospital from February,2013 to October,2017.TCGA database and immunohistochemistry were used to analyze CPNE3 expression level in GC tissues and its effects on tumor progression and long-term prognosis of the patients.GO analysis was performed to predict the biological role of CPNE3 in GC.We also conducted cell experiments with MGC803 cells and observed the effects of CPNE3 knockdown,CPNE3 overexpression and LY294002(a PI3K/AKT inhibitor)treatment on cell apoptosis and cellular expressions of apoptotic proteins using flow cytometry and Western blotting.Results TCGA analysis and immunohistochemistry both showed high expressions of CPNE3 in GC(P<0.05).The patients with high CPNE3 expressions had a reduced 5-year survival(P<0.01),and a high CPNE3 expression,CEA level≥5 μg/L,CA19-9 level≥37 kU/L,T3-T4 stage,and N2-N3 stage were all independent risk factors for a lowered 5-year survival rate after surgery.The sensitivity and specificity of CPNE3 for predicting 5-year mortality was 79.59%and 74.55%,respectively(P<0.05).GO analysis predicted that CPNE3 negatively regulated GC cell apoptosis.In MGC803 cells,CPNE3 knockdown significantly increased cell apoptosis,enhanced Bax and Cleaved Caspase-3 expressions and decreased Bcl-2 expression,while CPNE3 overexpression produced the opposite results(P<0.05).The cellular expressions of p-PI3K and p-AKT were significantly decreased following CPNE3 knockdown and increased following CPNE3 overexpression(P<0.05).Treatment with LY294002 obviously attenuated the inhibitory effect of CPNE3 overexpression on apoptosis of MGC803 cells(P<0.05).Conclusion CPNE3 is highly expressed in GC tissues and affects the long-term prognosis of the patients possibly by inhibiting GC cell apoptosis through activation of PI3K/AKT signaling.

Objective To investigate the expression level of basic leucine zipper and W2 domain-containing protein 1(BZW1)in gastric cancer,its impact on patient prognosis and the underlying mechanisms.Methods TIMER,UALCAN and Kaplan-Meier Plotter databases were used for analyzing BZW1 expression level gastric cancer tissues and its correlation with tumor grade and stage and the patients'prognosis.We further analyzed BZW1 expressions,disease progression,and postoperative 5-year survival in 102 patients undergoing radical surgery for gastric cancer at our hospital between January,2014 and December,2016.Gastric cancer MGC803 cells were examined for changes in migration,invasion,and epithelial-mesenchymal transition(EMT)following lentivirus-mediated BZW1 overexpression or knockdown.Results The protein and mRNA expressions of BZW1 in gastric cancer tissues were 3.30 and 6.54 times of those in adjacent tissues,respectively(P<0.01).BZW1 expression in gastric cancer tissues were positively correlated with peripheral blood CEA and CA199 levels(P<0.01).A high BZW1 expression was an independent risk factor for 5-year survival of gastric cancer patients after radical surgery(P<0.05,HR=2.070,95%CI:1.021-4.196).At the cut-off value of 3.61,BZW1 expression had a sensitivity of 75.56%and a specificity of 71.93%for predicting postoperative 5-year mortality(P<0.01).In MGC803 cells,BZW1 overexpression obviously promoted cell migration and invasion(P<0.05),enhanced cellular expressions of N-cadherin and vimentin(P<0.05)and inhibited the expression of E-cadherin(P<0.05).Enrichment analysis suggested the involvement of BZW1 in the Wnt/β-catenin signaling pathway.Western blotting confirmed that BZW1 overexpression promoted while BZW1 knockdown inhibited the expressions of Wnt3a,β-catenin and C-myc in MGC803 cells(P<0.05).Conclusion BZW1 is highly expressed in gastric cancer tissues to affect the patient prognosis possibly by activation the Wnt/β-catenin signaling pathway to promote EMT of gastric cancer cells.

Objective To analyze the expression level of ATP5A1 in gastric carcinoma and its influence on the prognosis of the patients and glucose metabolism in the tumor cells.Methods We retrospectively analyzed the data of 115 patients undergoing radical resection of gastric carcinoma in our hospital from February,2013 to November,2016.ATP5A1 expression in the surgical specimens were detected using immunohistochemistry,and the long-term prognosis of the patients with high(n=58)and low ATP5A1 expression(n=57)were analyzed.In gastric carcinoma MGC803 cells,the effects of lentivirus-mediated ATP5A1 knockdown or overexpression on glucose metabolism were investigated.We also observed the growth and glucose metabolism of xenografts derived from MGC803 cells with ATP5A1 knockdown or overexpression in nude mice.Results ATP5A1 was significantly overexpressed in gastric carcinoma tissues in close correlation with blood CEA and CA19-9 levels,pathological grade,T stage and N stage(P<0.05).ATP5A1 overexpression was an independent risk factor for a significantly lowered 5-year survival rate of patients with gastric carcinoma(P<0.05).ROC curve analysis demonstrated the predictive value of high ATP5A1 expression for the patients'prognosis(P<0.001).In MGC803 cells,ATP5A1 overexpression significantly up-regulated cellular glucose uptake and lactate production and increased the protein levels of HK2,PFK1,and LDHA(P<0.05),while ATP5A1 knockdown produced the opposite changes(P<0.05).In the tumor-bearing mice,overexpression of ATP5A1 increased glucose metabolism of the tumor cells and promoted tumor growth(P<0.05).Overexpression of ATP5A1 promoted the expressions of p-JNK and p-JUN in MGC803 cells(P<0.05),and the JNK inhibitor SP600125 significantly inhibited the enhancement of cellular glucose metabolism induced by ATP5A1 overexpression(P<0.05).Conclusion High ATP5A1 expression in gastric cancer is associated a poor long-term prognosis of the patients,and its effect is mediated at least partly by promoting glucose metabolism of the cells through the JNK/JUN pathway.

Objective To investigate the effects of kuwanon G(KG)on proliferation,apoptosis,migration and invasion of gastric cancer cells and the molecular mechanisms.Methods The effects of KG on proliferation and growth of gastric cancer cells were assessed with CCK-8 assay and cell clone formation assay,by observing tumor formation on the back of nude mice and using immunohistochemical analysis of Ki-67.The effect of KG on cell apoptosis was analyzed using Annexin V-FITC/PI apoptosis detection kit,Western blotting and TUNEL staining.The effects of KG on cell migration and invasion were detected using Transwell migration and invasion assay and Western blotting for matrix metalloproteinase(MMP).The role of phosphatidylinositol 3-kinase(PI3K)/protein kinase B(AKT)/mammalian target of rapamycin(mTOR)pathway in KG-mediated regulation of gastric cancer cell proliferation,migration,and invasion was verified by Western blotting and rescue assay.Results KG significantly inhibited proliferation and reduced clone formation ability of gastric cancer cells in a concentration-dependent manner(P<0.05).KG treatment also increased apoptosis,enhanced the expressions of cleaved caspase-3 and Bax,down-regulated Bcl-2,lowered migration and invasion capacities and inhibited the expression of MMP2 and MMP9 in gastric cancer cells(P<0.05).Mechanistic validation showed that KG inhibited the activation of the PI3K/AKT/mTOR pathway,and IGF-1,an activator of the PI3K/AKT/mTOR pathway,reversed the effects of KG on proliferation,migration and invasion of gastric cancer cells(P<0.05).Conclusion KG inhibits proliferation,migration and invasion and promotes apoptosis of gastric cancer cells at least in part by inhibiting the activation of the PI3K/AKT/mTOR pathway.