Objective To compare the differences between swept-source optical coherence tomography angiography(SS-OCTA)and ultra-wide-field fluorescein angiography(UWFA)in detecting non-perfusion areas(NPs)in patients with diabetic retinopathy(DR),to evaluate the accuracy of SS-OCTA in predicting NPs outside its visible range,and to explore the distribution patterns of NPs.Methods A retrospective analysis was made on 69 DR patients(88 eyes)who under-went both UWFA and SS-OCTA examinations at the Ophthalmology Department of Sichuan Provincial People's Hospital from December 2022 to September 2024.Manual NP labeling was conducted to compare the detection rate of NPs between the two imaging techniques.The distribution patterns of NPs and the accuracy of SS-OCTA for predicting NPs outside its visible range were also analyzed.Results In a scanning area of 20 mm x 24 mm,the overall NP detection rate by SS-OCTA was 47.40％,with UWFA taken as the standard.The NP detection rate by SS-OCTA was 51.56％in the superotemporal quad-rant,58.35％in the inferotemporal quadrant,45.50％in the superonasal quadrant,and 43.17％in the inferonasal quad-rant.Most NPs occurred in the inferonasal quadrant,accounting for 41.71％of the total NP.The accuracy of SS-OCTA in predicting NPs was 75.00％in the superonasal quadrant and 78.41％in the inferonasal quadrant.The ischemic indices(ISI)of the two imaging techniques were highly positively correlated(r2=0.74).Conclusion Although SS-OCTA can-not yet fully replace UWFA for NP detection in DR patients due to a small visible range,it is still an effective tool to assess retinal ischemia.SS-OCTA has the ability to predict NPs outside its visible range in its scanning range.The inferonasal quadrant is the region where NPs occur most frequently in DR patients,so it is suggested that special attention should be paid to this region in early diagnosis and follow-up periods.

Objective Deep learning provides strong technical support for early diagnosis,lesion segmentation,and treatment prediction of retinal diseases,significantly improving the efficiency and accuracy of diagnosis.But it also faces challenges in terms of different applicability and performance differences of the model,mainly due to the differences in fea-ture extraction ability,computational complexity,and clinical adaptability among different network structures,which make them have different advantages and limitations in different application scenarios.By systematically searching relevant litera-ture in PubMed and Web of Science databases over the past 5 years,this article summarizes the most commonly used deep learning network architectures in common vitreoretinal diseases,summarizes their different advantages and limitations,and analyzes the best application directions of each architecture in the field of ophthalmology,providing reference and inspira-tion for future research.

Objective To quantify the intraretinal fluid(IRF)volume in patients with diabetic macular edema(DME)using a deep learning-based three-dimensional segmentation model and to investigate the relationship of IRF volume with visual function and retinal biomarkers based on optical coherence tomography(OCT).Methods A total of 37 pa-tients(42 eyes)with DME who received treatment in the Ophthalmology Department of Sichuan Provincial People's Hospi-tal from July 2022 to September 2024 were prospectively included in this study.A three-dimensional segmentation model was used to automatically quantify the IRF volume within a 6 mm × 6 mm OCT scan at baseline and at the last follow-up.The correlation of visual acuity with IRF volume,central subfield thickness(CST),disorganization of the retinal inner lay-ers(DRIL),hyperreflective foci(HRF),external limiting membrane(ELM),ellipsoid zone(EZ),and vitreomacular in-terface abnormality(VMIA)was analyzed at baseline and the last follow-up.Additionally,the correlation of IRF volume with CST,DRIL,HRF,ELM,EZ,and VMIA was analyzed at both time points.Furthermore,the correlation of the visual acuity at the last follow-up with the IRF volume,CST,DRIL,HRF,ELM,EZ,and VMIA at baseline was analyzed.Results Compared with the baseline value,patients exhibited a significant increase in the best corrected visual acuity(BCVA)(logMAR),a significant decrease in the IRF volume and CST,a significant reduction in the HRF,a significant re-covery in the DRIL,ELM,and EZ(all P＜0.05),and no significant changes in the VMIA(P=1.000)at the last follow-up.At baseline,the BCVA was negatively correlated with the IRF volume,CST,DRIL,and HRF,but positively correlated with the ELM and EZ(all P＜0.05);there was no correlation between the BCVA and VMIA(P=0.069).At the last follow-up,the BCVA was negatively correlated with the DRIL,HRF,and VMIA,but positively correlated with the ELM and EZ(all P＜0.01);the BCVA did not correlate with the IRF volume and CST(P=0.419 and 0.994).At baseline,the IRF volume was positively correlated with the CST(P＜0.001)but negatively correlated with the ELM and EZ(P＜0.01);the IRF vol-ume did not correlate with the DRIL,HRF,and VMIA(all P＞0.05).At the last follow-up,the IRF volume was positively correlated with the CST and HRF(all P＜0.01);however,it was not correlated with the DRIL,ELM,EZ,and VMIA(all P＞0.05).The BCVA at the last follow-up was positively correlated with the BCVA,ELM,and EZ at baseline,but negative-ly correlated with the IRF volume,CST,DRIL,HRF,and VMIA at baseline(all P＜0.05).Conclusion The baseline IRF volume in DME patients is an important influencing factor of visual acuity at baseline and the last follow-up.Hence,the IRF volume may serve as a potential biomarker in the management of DME.

Objective To quantify the intraretinal fluid(IRF)volume in patients with diabetic macular edema(DME)using a deep learning-based three-dimensional segmentation model and to investigate the relationship of IRF volume with visual function and retinal biomarkers based on optical coherence tomography(OCT).Methods A total of 37 pa-tients(42 eyes)with DME who received treatment in the Ophthalmology Department of Sichuan Provincial People's Hospi-tal from July 2022 to September 2024 were prospectively included in this study.A three-dimensional segmentation model was used to automatically quantify the IRF volume within a 6 mm × 6 mm OCT scan at baseline and at the last follow-up.The correlation of visual acuity with IRF volume,central subfield thickness(CST),disorganization of the retinal inner lay-ers(DRIL),hyperreflective foci(HRF),external limiting membrane(ELM),ellipsoid zone(EZ),and vitreomacular in-terface abnormality(VMIA)was analyzed at baseline and the last follow-up.Additionally,the correlation of IRF volume with CST,DRIL,HRF,ELM,EZ,and VMIA was analyzed at both time points.Furthermore,the correlation of the visual acuity at the last follow-up with the IRF volume,CST,DRIL,HRF,ELM,EZ,and VMIA at baseline was analyzed.Results Compared with the baseline value,patients exhibited a significant increase in the best corrected visual acuity(BCVA)(logMAR),a significant decrease in the IRF volume and CST,a significant reduction in the HRF,a significant re-covery in the DRIL,ELM,and EZ(all P＜0.05),and no significant changes in the VMIA(P=1.000)at the last follow-up.At baseline,the BCVA was negatively correlated with the IRF volume,CST,DRIL,and HRF,but positively correlated with the ELM and EZ(all P＜0.05);there was no correlation between the BCVA and VMIA(P=0.069).At the last follow-up,the BCVA was negatively correlated with the DRIL,HRF,and VMIA,but positively correlated with the ELM and EZ(all P＜0.01);the BCVA did not correlate with the IRF volume and CST(P=0.419 and 0.994).At baseline,the IRF volume was positively correlated with the CST(P＜0.001)but negatively correlated with the ELM and EZ(P＜0.01);the IRF vol-ume did not correlate with the DRIL,HRF,and VMIA(all P＞0.05).At the last follow-up,the IRF volume was positively correlated with the CST and HRF(all P＜0.01);however,it was not correlated with the DRIL,ELM,EZ,and VMIA(all P＞0.05).The BCVA at the last follow-up was positively correlated with the BCVA,ELM,and EZ at baseline,but negative-ly correlated with the IRF volume,CST,DRIL,HRF,and VMIA at baseline(all P＜0.05).Conclusion The baseline IRF volume in DME patients is an important influencing factor of visual acuity at baseline and the last follow-up.Hence,the IRF volume may serve as a potential biomarker in the management of DME.

Objective To compare the differences between swept-source optical coherence tomography angiography(SS-OCTA)and ultra-wide-field fluorescein angiography(UWFA)in detecting non-perfusion areas(NPs)in patients with diabetic retinopathy(DR),to evaluate the accuracy of SS-OCTA in predicting NPs outside its visible range,and to explore the distribution patterns of NPs.Methods A retrospective analysis was made on 69 DR patients(88 eyes)who under-went both UWFA and SS-OCTA examinations at the Ophthalmology Department of Sichuan Provincial People's Hospital from December 2022 to September 2024.Manual NP labeling was conducted to compare the detection rate of NPs between the two imaging techniques.The distribution patterns of NPs and the accuracy of SS-OCTA for predicting NPs outside its visible range were also analyzed.Results In a scanning area of 20 mm x 24 mm,the overall NP detection rate by SS-OCTA was 47.40％,with UWFA taken as the standard.The NP detection rate by SS-OCTA was 51.56％in the superotemporal quad-rant,58.35％in the inferotemporal quadrant,45.50％in the superonasal quadrant,and 43.17％in the inferonasal quad-rant.Most NPs occurred in the inferonasal quadrant,accounting for 41.71％of the total NP.The accuracy of SS-OCTA in predicting NPs was 75.00％in the superonasal quadrant and 78.41％in the inferonasal quadrant.The ischemic indices(ISI)of the two imaging techniques were highly positively correlated(r2=0.74).Conclusion Although SS-OCTA can-not yet fully replace UWFA for NP detection in DR patients due to a small visible range,it is still an effective tool to assess retinal ischemia.SS-OCTA has the ability to predict NPs outside its visible range in its scanning range.The inferonasal quadrant is the region where NPs occur most frequently in DR patients,so it is suggested that special attention should be paid to this region in early diagnosis and follow-up periods.

Objective Deep learning provides strong technical support for early diagnosis,lesion segmentation,and treatment prediction of retinal diseases,significantly improving the efficiency and accuracy of diagnosis.But it also faces challenges in terms of different applicability and performance differences of the model,mainly due to the differences in fea-ture extraction ability,computational complexity,and clinical adaptability among different network structures,which make them have different advantages and limitations in different application scenarios.By systematically searching relevant litera-ture in PubMed and Web of Science databases over the past 5 years,this article summarizes the most commonly used deep learning network architectures in common vitreoretinal diseases,summarizes their different advantages and limitations,and analyzes the best application directions of each architecture in the field of ophthalmology,providing reference and inspira-tion for future research.

Objective:To investigate whether the existence of thoracic artery blood supply in peripheral pulmonary lesions is the key factor affecting the accuracy of contrast-enhanced ultrasound in differentiating benign and malignant lesions.Methods:From June 2020 to December 2021, a total of 170 patients with peripheral pulmonary lesions were consecutively enrolled in Xi′an Chest Hospital, and all patients underwent conventional ultrasound and contrast-enhanced ultrasound(CEUS). Taking ΔAT(lesion-lung arrival time difference ) of 2.5 seconds as the cut-off point for differentiating benign and malignant lesions(ΔAT<2.5 s for benign, ΔAT≥2.5 s for malignant), and the final pathological results as the gold standard, these patiens were divided into correct classification group and wrong classification group, and the main influencing factors of wrong classification were analyzed.Results:Compared with the correct classification group, the proportion of thoracic artery blood supply in the wrong classification group was significantly higher ( P<0.001). After adjusting the dendritic venous reflux, thoracic artery blood supply was an independent influencing factor for CEUS misclassification ( OR=3.531, 95% CI=1.805-6.908, P<0.001). In the patients with thoracic artery blood supply, the sensitivity of the absence of dendritic venous reflux in judging malignant peripulmonary lesions was 75.0%, the specificity was 91.3%, and the area under the ROC curve (AUC) was 0.832 (95% CI=0.715-0.915, P<0.001), while the sensitivity, specificity and AUC of CEUS in judging malignant peripulmonary lesions were 68.7%, 67.4% and 0.659 (95% CI=0.528-0.775, P=0.006), and there was significant statistical difference in the AUC between them( P<0.001). In the group of patients without thoracic arterial blood supply, the sensitivity of the absence of dendritic venous reflux in judging malignant peripulmonary lesions was 82.8%, the specificity was 63.3%, and the AUC was 0.730 (95% CI=0.636-0.811, P<0.001), while the sensitivity, specificity and AUC of CEUS in judging malignant peripulmonary lesions were 62.1%, 81.0% and 0.684 (95% CI=0.587-0.770, P=0.003), and there was no significant difference in the AUC between them ( P=0.425). Conclusions:The presence of thoracic artery blood supply significantly decrease the diagnostic efficiency of CEUS in differentiating benign and malignant of peripheral pulmonary lesions. For peripheral pulmonary lesions with thoracic arterial blood supply, the diagnostic efficiency of dendritic venous reflux is better than CEUS, while for lesions without thoracic artery supply, the diagnostic efficiency of dendritic venous reflux is equivalent to CEUS.

Objective To understand the effects of chemoradiotherapy combined with radiotherapy on peripheral differential blood cell counts on tumor patients.Methods 40 patients receiving concurrent chemoradiotherapy were randomly selected in the radiotherapy department in our hospital.Peripheral blood WBC,PLT radiotherapy for 1-4 weeks were retrospective analysis.Results The first 1-4 week of radiotherapy,peripheral blood cells of WBC,PLT decreased.The difference was statistically significant (WBC:x2 =16.902,P =0.001,PLT:x2 =8.492,P =0.037),and obviously in the second week.Conclusion Tumor radiotherapy decreases blood cell counting and classification in the WBC,PLT,and obviously in the second week.