Objectives: This study aimed to determine whether errors in vascular measurements would affect device selection in endovascular aortic repair (EVAR) by comparing measurements obtained using non-contrast computed tomog...Objectives: This study aimed to determine whether errors in vascular measurements would affect device selection in endovascular aortic repair (EVAR) by comparing measurements obtained using non-contrast computed tomography (NCT) with those obtained using contrast-enhanced CT (CECT). Materials and Methods: This single-center, retrospective study included 25 patients who underwent EVAR for abdominal aortic aneurysm at our institution. A 1-mm horizontal cross-sectional slice of NCT and CECT from each patient was retrospectively reviewed. The area from the abdominal aorta to the common iliac artery was divided into four zones. A centerline was created using the NCT by manually plotting the center points. Subsequently, the centerlines were automatically extracted and manually corrected during the arterial phase of CECT. The diameter and length of each zone were measured for each modality. The mean diameters and lengths of the target vessels were compared between NCT and CECT. Results: The measurements obtained using both methods were reproducible and demonstrated good agreement. The mean differences in vessel length and diameter measurements for each segment between NCT and CECT were not statistically significant, indicating good consistency. Conclusion: NCT may be useful for preoperative EVAR evaluation in patients with renal dysfunction or allergies to contrast agents.展开更多
BACKGROUND Accurate preoperative differentiation of benign and malignant thyroid nodules is critical for optimal patient management.However,conventional imaging modalities present inherent diagnostic limitations.AIM T...BACKGROUND Accurate preoperative differentiation of benign and malignant thyroid nodules is critical for optimal patient management.However,conventional imaging modalities present inherent diagnostic limitations.AIM To develop a non-contrast computed tomography-based machine learning model integrating radiomics and clinical features for preoperative thyroid nodule classification.METHODS This multicenter retrospective study enrolled 272 patients with thyroid nodules(376 thyroid lobes)from center A(May 2021-April 2024),using histopathological findings as the reference standard.The dataset was stratified into a training cohort(264 lobes)and an internal validation cohort(112 lobes).Additional prospective temporal(97 lobes,May-August 2024,center A)and external multicenter(81 lobes,center B)test cohorts were incorporated to enhance generalizability.Thyroid lobes were segmented along the isthmus midline,with segmentation reliability confirmed by an intraclass correlation coefficient(≥0.80).Radiomics feature extraction was performed using Pearson correlation analysis followed by least absolute shrinkage and selection operator regression with 10-fold cross-validation.Seven machine learning algorithms were systematically evaluated,with model performance quantified through the area under the receiver operating characteristic curve(AUC),Brier score,decision curve analysis,and DeLong test for comparison with radiologists interpretations.Model interpretability was elucidated using SHapley Additive exPlanations(SHAP).RESULTS The extreme gradient boosting model demonstrated robust diagnostic performance across all datasets,achieving AUCs of 0.899[95%confidence interval(CI):0.845-0.932]in the training cohort,0.803(95%CI:0.715-0.890)in internal validation,0.855(95%CI:0.775-0.935)in temporal testing,and 0.802(95%CI:0.664-0.939)in external testing.These results were significantly superior to radiologists assessments(AUCs:0.596,0.529,0.558,and 0.538,respectively;P<0.001 by DeLong test).SHAP analysis identified radiomic score,age,tumor size stratification,calcification status,and cystic components as key predictive features.The model exhibited excellent calibration(Brier scores:0.125-0.144)and provided significant clinical net benefit at decision thresholds exceeding 20%,as evidenced by decision curve analysis.CONCLUSION The non-contrast computed tomography-based radiomics-clinical fusion model enables robust preoperative thyroid nodule classification,with SHAP-driven interpretability enhancing its clinical applicability for personalized decision-making.展开更多
BACKGROUND Intracerebral hemorrhage(ICH)comprises 9%-27%of stroke patients.Hematoma expansion(HE)occurs in approximately 20%of patients following ICH,typically within the first 24 hours.HE increases mortality and long...BACKGROUND Intracerebral hemorrhage(ICH)comprises 9%-27%of stroke patients.Hematoma expansion(HE)occurs in approximately 20%of patients following ICH,typically within the first 24 hours.HE increases mortality and long-term disability in these patients and is being investigated as a therapeutic target to improve the outcome in these patients by limiting HE.Non-contrast computed tomography(NCCT)has potential in predicting HE,which can identify the individuals at risk.AIM To evaluate NCCT markers for predicting HE in patients with ICH and to develop a simple,practical grading system for risk stratification.METHODS This prospective observational study evaluated 192 patients with spontaneous ICH who underwent a baseline NCCT within four hours of admission,followed by a follow-up scan after six hours or earlier if there was clinical deterioration.Hematoma volumes and imaging characteristics that predicted HE were evaluated.A simple five-point grading system score was created to predict HE.In RESULTS Of the 192 patients studied,HE was seen in 106(55.2%).The mean baseline hematoma volume was significantly greater among patients in the HE group(44.1 mL)compared to those in the non-HE group(12.2 mL),with a P-value<0.05.Additionally,imaging biomarkers such as the island sign,swirl sign,and black hole sign were observed with significantly higher frequency in the HE group relative to the non-HE cohort(all P-values<0.05).The island sign was strongly associated with HE[odds ratio(OR)13.7;95%confidence interval(CI):10.15-16.37;P<0.001].Similarly,the black hole sign(OR 9.4;95%CI:7.4-11.62;P<0.001)and the swirl sign(OR 5.2;95%CI:3.72-6.53;P<0.001)emerged as significant predictors of HE.Initial hematoma volume≥30 mL also showed a sig-nificant association(OR 1.9;95%CI:1.41-2.74;P=0.039).A five-point predictive scoring model demonstrated a strong positive association between increasing scores and the probability of HE.Specifically,the likelihood of HE corresponding to scores of 0,1,2,3,4,and 5 was 7.4%,37.5%,75%,85%,93.3%,and 100%,respectively.CONCLUSION The five variables demonstrated statistically significant associations with HE.This simple and practical 5-point prediction score can enable identification of patients at elevated risk of HE based on baseline NCCT findings.This can facilitate timely recognition of high-risk individuals who may benefit from targeted anti-expansion therapy.展开更多
目的研究冠状动脉CT血管成像(CCTA)评价冠状动脉慢性完全闭塞病变(chronic total occlusion,CTO)形态学参数在介入治疗指导中的应用价值。方法选取2021年1月至2023年12月金华市人民医院收治的经冠状动脉造影(ICA)证实的CTO患者300例,患...目的研究冠状动脉CT血管成像(CCTA)评价冠状动脉慢性完全闭塞病变(chronic total occlusion,CTO)形态学参数在介入治疗指导中的应用价值。方法选取2021年1月至2023年12月金华市人民医院收治的经冠状动脉造影(ICA)证实的CTO患者300例,患者术前均接受CCTA检查。记录CCTA形态学参数闭塞段近端形态、闭塞血管长度、闭塞段内线样强化长度、闭塞段内线样强化长度/闭塞血管长度、闭塞段血管线样强化、闭塞段内血管钙化情况、闭塞段内血管钙化面积≥横截面50%、病变走行迂曲(>45°)、侧支血管情况、血管开口病变,并分析以上参数与PCI治疗结果的关系。结果300例CTO患者病变共325处,PCI治疗成功227处(69.85%),PCI治疗失败98处(30.15%);失败组闭塞段近端钝形、闭塞血管长度、闭塞段内血管钙化面积≥横截面50%、病变走行迂曲(>45°)明显高于成功组(P<0.05),闭塞段内线样强化长度、闭塞段内线样强化长度/闭塞血管长度、闭塞段内线样强化明显低于成功组(P<0.05),两组其余参数差异均无统计学意义(P>0.05);多因素logistic回归分析结果显示,闭塞段内线样强化长度(OR=1.975,95%CI:1.306~2.988)、闭塞段内线样强化长度/闭塞血管长度(OR=3.831,95%CI:1.332~11.017)、闭塞段内线样强化(OR=1.702,95%CI:1.007~2.879)是预测PCI治疗成功的相关因素(P<0.05)。结论CCTA评价冠状动脉CTO形态学参数在介入治疗中具有一定的指导作用,其中闭塞段内线样强化长度、闭塞段内线样强化长度/闭塞血管长度、闭塞段内线样强化是预测PCI治疗成功的相关因素。展开更多
文摘Objectives: This study aimed to determine whether errors in vascular measurements would affect device selection in endovascular aortic repair (EVAR) by comparing measurements obtained using non-contrast computed tomography (NCT) with those obtained using contrast-enhanced CT (CECT). Materials and Methods: This single-center, retrospective study included 25 patients who underwent EVAR for abdominal aortic aneurysm at our institution. A 1-mm horizontal cross-sectional slice of NCT and CECT from each patient was retrospectively reviewed. The area from the abdominal aorta to the common iliac artery was divided into four zones. A centerline was created using the NCT by manually plotting the center points. Subsequently, the centerlines were automatically extracted and manually corrected during the arterial phase of CECT. The diameter and length of each zone were measured for each modality. The mean diameters and lengths of the target vessels were compared between NCT and CECT. Results: The measurements obtained using both methods were reproducible and demonstrated good agreement. The mean differences in vessel length and diameter measurements for each segment between NCT and CECT were not statistically significant, indicating good consistency. Conclusion: NCT may be useful for preoperative EVAR evaluation in patients with renal dysfunction or allergies to contrast agents.
基金Supported by the Science and Technology Development Fund of Nanjing Medical University,No.NMUB20230037the Youth Scientific Research Nurturing Fund of Jiangbei Campus of Zhongda Hospital Affiliated with Southeast University,No.JB2024Q01.
文摘BACKGROUND Accurate preoperative differentiation of benign and malignant thyroid nodules is critical for optimal patient management.However,conventional imaging modalities present inherent diagnostic limitations.AIM To develop a non-contrast computed tomography-based machine learning model integrating radiomics and clinical features for preoperative thyroid nodule classification.METHODS This multicenter retrospective study enrolled 272 patients with thyroid nodules(376 thyroid lobes)from center A(May 2021-April 2024),using histopathological findings as the reference standard.The dataset was stratified into a training cohort(264 lobes)and an internal validation cohort(112 lobes).Additional prospective temporal(97 lobes,May-August 2024,center A)and external multicenter(81 lobes,center B)test cohorts were incorporated to enhance generalizability.Thyroid lobes were segmented along the isthmus midline,with segmentation reliability confirmed by an intraclass correlation coefficient(≥0.80).Radiomics feature extraction was performed using Pearson correlation analysis followed by least absolute shrinkage and selection operator regression with 10-fold cross-validation.Seven machine learning algorithms were systematically evaluated,with model performance quantified through the area under the receiver operating characteristic curve(AUC),Brier score,decision curve analysis,and DeLong test for comparison with radiologists interpretations.Model interpretability was elucidated using SHapley Additive exPlanations(SHAP).RESULTS The extreme gradient boosting model demonstrated robust diagnostic performance across all datasets,achieving AUCs of 0.899[95%confidence interval(CI):0.845-0.932]in the training cohort,0.803(95%CI:0.715-0.890)in internal validation,0.855(95%CI:0.775-0.935)in temporal testing,and 0.802(95%CI:0.664-0.939)in external testing.These results were significantly superior to radiologists assessments(AUCs:0.596,0.529,0.558,and 0.538,respectively;P<0.001 by DeLong test).SHAP analysis identified radiomic score,age,tumor size stratification,calcification status,and cystic components as key predictive features.The model exhibited excellent calibration(Brier scores:0.125-0.144)and provided significant clinical net benefit at decision thresholds exceeding 20%,as evidenced by decision curve analysis.CONCLUSION The non-contrast computed tomography-based radiomics-clinical fusion model enables robust preoperative thyroid nodule classification,with SHAP-driven interpretability enhancing its clinical applicability for personalized decision-making.
文摘BACKGROUND Intracerebral hemorrhage(ICH)comprises 9%-27%of stroke patients.Hematoma expansion(HE)occurs in approximately 20%of patients following ICH,typically within the first 24 hours.HE increases mortality and long-term disability in these patients and is being investigated as a therapeutic target to improve the outcome in these patients by limiting HE.Non-contrast computed tomography(NCCT)has potential in predicting HE,which can identify the individuals at risk.AIM To evaluate NCCT markers for predicting HE in patients with ICH and to develop a simple,practical grading system for risk stratification.METHODS This prospective observational study evaluated 192 patients with spontaneous ICH who underwent a baseline NCCT within four hours of admission,followed by a follow-up scan after six hours or earlier if there was clinical deterioration.Hematoma volumes and imaging characteristics that predicted HE were evaluated.A simple five-point grading system score was created to predict HE.In RESULTS Of the 192 patients studied,HE was seen in 106(55.2%).The mean baseline hematoma volume was significantly greater among patients in the HE group(44.1 mL)compared to those in the non-HE group(12.2 mL),with a P-value<0.05.Additionally,imaging biomarkers such as the island sign,swirl sign,and black hole sign were observed with significantly higher frequency in the HE group relative to the non-HE cohort(all P-values<0.05).The island sign was strongly associated with HE[odds ratio(OR)13.7;95%confidence interval(CI):10.15-16.37;P<0.001].Similarly,the black hole sign(OR 9.4;95%CI:7.4-11.62;P<0.001)and the swirl sign(OR 5.2;95%CI:3.72-6.53;P<0.001)emerged as significant predictors of HE.Initial hematoma volume≥30 mL also showed a sig-nificant association(OR 1.9;95%CI:1.41-2.74;P=0.039).A five-point predictive scoring model demonstrated a strong positive association between increasing scores and the probability of HE.Specifically,the likelihood of HE corresponding to scores of 0,1,2,3,4,and 5 was 7.4%,37.5%,75%,85%,93.3%,and 100%,respectively.CONCLUSION The five variables demonstrated statistically significant associations with HE.This simple and practical 5-point prediction score can enable identification of patients at elevated risk of HE based on baseline NCCT findings.This can facilitate timely recognition of high-risk individuals who may benefit from targeted anti-expansion therapy.
文摘目的研究冠状动脉CT血管成像(CCTA)评价冠状动脉慢性完全闭塞病变(chronic total occlusion,CTO)形态学参数在介入治疗指导中的应用价值。方法选取2021年1月至2023年12月金华市人民医院收治的经冠状动脉造影(ICA)证实的CTO患者300例,患者术前均接受CCTA检查。记录CCTA形态学参数闭塞段近端形态、闭塞血管长度、闭塞段内线样强化长度、闭塞段内线样强化长度/闭塞血管长度、闭塞段血管线样强化、闭塞段内血管钙化情况、闭塞段内血管钙化面积≥横截面50%、病变走行迂曲(>45°)、侧支血管情况、血管开口病变,并分析以上参数与PCI治疗结果的关系。结果300例CTO患者病变共325处,PCI治疗成功227处(69.85%),PCI治疗失败98处(30.15%);失败组闭塞段近端钝形、闭塞血管长度、闭塞段内血管钙化面积≥横截面50%、病变走行迂曲(>45°)明显高于成功组(P<0.05),闭塞段内线样强化长度、闭塞段内线样强化长度/闭塞血管长度、闭塞段内线样强化明显低于成功组(P<0.05),两组其余参数差异均无统计学意义(P>0.05);多因素logistic回归分析结果显示,闭塞段内线样强化长度(OR=1.975,95%CI:1.306~2.988)、闭塞段内线样强化长度/闭塞血管长度(OR=3.831,95%CI:1.332~11.017)、闭塞段内线样强化(OR=1.702,95%CI:1.007~2.879)是预测PCI治疗成功的相关因素(P<0.05)。结论CCTA评价冠状动脉CTO形态学参数在介入治疗中具有一定的指导作用,其中闭塞段内线样强化长度、闭塞段内线样强化长度/闭塞血管长度、闭塞段内线样强化是预测PCI治疗成功的相关因素。
