Yan Gu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yan Gu has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
Yonggang Zhang
Department of Imaging, Lianyungang First People’s Hospital, Lianyungang, Jiangsu, People’s Republic of China
4 shared publicationsChongchang Miao
The Affiliated Lianyungang Hospital of xuzhou medical university
3 shared publicationsYinan Zhao
Department of Radiology, The Affiliated Jinzhou Medical University, Jinzhou, People’s Republic of China
3 shared publicationsYing Liu
Department of Radiology, The Affiliated Lianyungang Hospital of Xuzhou Medical University, Lianyungang, People’s Republic of China
3 shared publicationsZhongping Guo
Department of Radiology, The Affiliated Lianyungang Clinical College of Nanjing Medical University, Lianyungang, People’s Republic of China
2 shared publicationsBin Hu
Department of Medical Imaging, Jinling Hospital, Medical School of Nanjing University
1 shared publication
Latest Publications
- Open Access
- 1 January 2026
- Published by AME Publishing Company in Gland Surgery
- Vol. 15 (1) , 7
Optimized dual-source dual-energy computed tomography nomogram model integrating background normalization improves solid thyroid nodules diagnosis
Accurate preoperative diagnosis of benign and malignant thyroid nodules is crucial for personalized patient treatment and management. This study aimed to create a dual-source dual-energy computed tomography (DS-DECT) based nomogram model to predict the risk of malignant thyroid nodules. A total of 263 patients (288 nodules) with thyroid nodules who underwent preoperative neck DS-DECT scans and were pathologically confirmed were included in this study. The computed tomography (CT) radiological features and DS-DECT-derived quantitative parameters of the nodules were collected. Subsequently, the thyroid nodules were randomly partitioned into a training cohort (n=201) and a validation cohort (n=87) at a ratio of 7:3. Univariate logistic regression analysis identified predictors (P<0.05), followed by least absolute shrinkage and selection operator (LASSO) logistic regression to screen features in the training cohort. Multivariate logistic regression analysis was then conducted to determine independent predictors of malignancy (P<0.05) and to construct a nomogram model for predicting malignancy risk. The performance of the model was assessed using the area under the curve (AUC), calibration curve, and decision curve analysis (DCA). The applicability of the nomogram was evaluated through internal validation. In the final stepwise multivariable logistic regression model, independent predictors of malignancy included iodine concentration in the arterial phase [IC_IAP; odds ratio (OR) =0.286; 95% confidence interval (CI): 0.140–0.517; P<0.001], normalized iodine concentration relative to the thyroid parenchyma in the arterial phase (NIC_P_IAP; OR =0.133; 95% CI: 0.033–0.417; P=0.003), effective atomic number in the venous phase (Zeff_IVP; OR =0.137; 95% CI: 0.050–0.333; P0.05), and the calibration curve was close to the standard curve. DCA showed significant net benefits from using the model. The nomogram model, based on normalized multiphase quantitative DECT parameters and qualitative imaging features, serves as an effective diagnostic tool for imaging physicians to distinguish between benign and malignant nodules.
- 1 January 2026
- Published by AME Publishing Company in Gland Surgery
- Vol. 15 (1) , 7
Open AccessOptimized dual-source dual-energy computed tomography nomogram model integrating background normalization improves solid thyroid nodules diagnosis
Accurate preoperative diagnosis of benign and malignant thyroid nodules is crucial for personalized patient treatment and management. This study aimed to create a dual-source dual-energy computed tomography (DS-DECT) based nomogram model to predict the risk of malignant thyroid nodules. A total of 263 patients (288 nodules) with thyroid nodules who underwent preoperative neck DS-DECT scans and were pathologically confirmed were included in this study. The computed tomography (CT) radiological features and DS-DECT-derived quantitative parameters of the nodules were collected. Subsequently, the thyroid nodules were randomly partitioned into a training cohort (n=201) and a validation cohort (n=87) at a ratio of 7:3. Univariate logistic regression analysis identified predictors (P<0.05), followed by least absolute shrinkage and selection operator (LASSO) logistic regression to screen features in the training cohort. Multivariate logistic regression analysis was then conducted to determine independent predictors of malignancy (P<0.05) and to construct a nomogram model for predicting malignancy risk. The performance of the model was assessed using the area under the curve (AUC), calibration curve, and decision curve analysis (DCA). The applicability of the nomogram was evaluated through internal validation. In the final stepwise multivariable logistic regression model, independent predictors of malignancy included iodine concentration in the arterial phase [IC_IAP; odds ratio (OR) =0.286; 95% confidence interval (CI): 0.140–0.517; P<0.001], normalized iodine concentration relative to the thyroid parenchyma in the arterial phase (NIC_P_IAP; OR =0.133; 95% CI: 0.033–0.417; P=0.003), effective atomic number in the venous phase (Zeff_IVP; OR =0.137; 95% CI: 0.050–0.333; P0.05), and the calibration curve was close to the standard curve. DCA showed significant net benefits from using the model. The nomogram model, based on normalized multiphase quantitative DECT parameters and qualitative imaging features, serves as an effective diagnostic tool for imaging physicians to distinguish between benign and malignant nodules.
- Open Access
- 1 March 2025
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 18, 1231-1246
Comparing Different Contrast Injection Methods for Multislice Spiral CT Imaging in Triple-Rule-Out Examinations: A Study on Acute Chest Pain Patients
Introduction: The traditional injection method of contrast media (CM) has limitations in terms of volume and imaging quality. Choosing the optimal CM injection method is crucial for ensuring high-quality images and accurate diagnosis in triple-rule-out computed tomography angiography (TRO-CTA). This study aims to investigate the application value of combining region of interest (ROI) with CM injection method alterations in TRO-CTA. Methods: A total of 596 patients with chest pain, chest tightness, and dyspnea as the primary clinical symptoms, underwent TRO-CTA examinations based on their clinical symptoms and test indicators. These patients were randomly assigned to one of three groups: the traditional group, the modified group, and the high-pitch helical scan (FLASH) group. Key parameters, including computed tomography (CT) values, vascular filling contrast numbers, image noise, signal-to-noise ratio, contrast noise ratio, image scores, CM volume, CM injection rate, radiation dose of the aorta, pulmonary artery, and coronary artery, were systematically compared across the three groups. Results: The CM volume and radiation doses were significantly higher in the traditional group compared to the other groups (P < 0.05). CT values for the aorta and coronary artery in the traditional group were higher than those in the modified and FLASH group, while CT values for the pulmonary artery were lower in the traditional group compared to the modified and FLASH groups (ALL P-value for traditional group versus FLASH group, P-value for modified group versus traditional group < 0.05, P-value for modified group versus FLASH group > 0.05). Conclusion: The combination of the modified CM injection method and FLASH scanning mode reduces the required CM volume by nearly 30 mL and significantly decreases radiation dose exposure during the examination. This approach shows valuable potential for improving both the safety and effectiveness of TRO-CTA, with significant implications for clinical practice.
- 1 March 2025
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 18, 1231-1246
Open AccessComparing Different Contrast Injection Methods for Multislice Spiral CT Imaging in Triple-Rule-Out Examinations: A Study on Acute Chest Pain Patients
Introduction: The traditional injection method of contrast media (CM) has limitations in terms of volume and imaging quality. Choosing the optimal CM injection method is crucial for ensuring high-quality images and accurate diagnosis in triple-rule-out computed tomography angiography (TRO-CTA). This study aims to investigate the application value of combining region of interest (ROI) with CM injection method alterations in TRO-CTA. Methods: A total of 596 patients with chest pain, chest tightness, and dyspnea as the primary clinical symptoms, underwent TRO-CTA examinations based on their clinical symptoms and test indicators. These patients were randomly assigned to one of three groups: the traditional group, the modified group, and the high-pitch helical scan (FLASH) group. Key parameters, including computed tomography (CT) values, vascular filling contrast numbers, image noise, signal-to-noise ratio, contrast noise ratio, image scores, CM volume, CM injection rate, radiation dose of the aorta, pulmonary artery, and coronary artery, were systematically compared across the three groups. Results: The CM volume and radiation doses were significantly higher in the traditional group compared to the other groups (P < 0.05). CT values for the aorta and coronary artery in the traditional group were higher than those in the modified and FLASH group, while CT values for the pulmonary artery were lower in the traditional group compared to the modified and FLASH groups (ALL P-value for traditional group versus FLASH group, P-value for modified group versus traditional group < 0.05, P-value for modified group versus FLASH group > 0.05). Conclusion: The combination of the modified CM injection method and FLASH scanning mode reduces the required CM volume by nearly 30 mL and significantly decreases radiation dose exposure during the examination. This approach shows valuable potential for improving both the safety and effectiveness of TRO-CTA, with significant implications for clinical practice.
- Open Access
- 1 March 2024
- Published by Taylor & Francis in Vascular Health and Risk Management
- Vol. ume 20, 77-87
Evaluation of the Correlation Between Distribution Location and Vulnerability of Carotid Plaque in Patients with Transient Ischemic Attack
Purpose: To analyze the relationship among distribution location, characteristics, and vulnerability of carotid plaque using CTA and provide more information on the risk factors of carotid atherosclerotic plaque. Patients and Methods: We retrospectively analyzed the CTA images of the head and neck of 93 patients with carotid atherosclerosis. Atherosclerosis was developed in 148 carotid arteries. The plaques were divided into a high-risk plaque group and a low-risk plaque group according to whether the plaques had high-risk characteristics. The maximum cross-sectional area of carotid artery bifurcation plaque on the axial image was selected, and the cross-sectional lumen was equally divided into four 90-degree sectors, ventral side wall, dorsal side wall, inner side wall, and outer side wall. The differences in the characteristics and distribution locations of the plaques in the two groups were analyzed. The characteristic parameters of the cross-sectional plaques at the bifurcation of the carotid artery. The logistic regression analysis was used to further analyze the risk factors associated with plaque vulnerability. Results: Among 148 carotid arteries,80 were classified as high-risk and 68 as low-risk groups. There were significant differences between the two groups concerning the thickness, length, maximum cross-sectional area, burden, and cross-sectional distribution of the plaques (P < 0.05). The plaque distribution on the dorsal side wall of the carotid bifurcation was higher in the high-risk group than that in the low-risk group (P < 0.05), dorsal side wall plaque-independent risk factors for the development of vulnerability of plaques in transient ischemic attack (TIA) patients (95% CI:1.522~6.991, P< 0.05). Conclusion: High-risk plaques tend to occur on the dorsal side wall of the carotid bifurcation, whereas low-risk plaques tend to occur on the outer side wall of the carotid bifurcation.
- 1 March 2024
- Published by Taylor & Francis in Vascular Health and Risk Management
- Vol. ume 20, 77-87
Open AccessEvaluation of the Correlation Between Distribution Location and Vulnerability of Carotid Plaque in Patients with Transient Ischemic Attack
Purpose: To analyze the relationship among distribution location, characteristics, and vulnerability of carotid plaque using CTA and provide more information on the risk factors of carotid atherosclerotic plaque. Patients and Methods: We retrospectively analyzed the CTA images of the head and neck of 93 patients with carotid atherosclerosis. Atherosclerosis was developed in 148 carotid arteries. The plaques were divided into a high-risk plaque group and a low-risk plaque group according to whether the plaques had high-risk characteristics. The maximum cross-sectional area of carotid artery bifurcation plaque on the axial image was selected, and the cross-sectional lumen was equally divided into four 90-degree sectors, ventral side wall, dorsal side wall, inner side wall, and outer side wall. The differences in the characteristics and distribution locations of the plaques in the two groups were analyzed. The characteristic parameters of the cross-sectional plaques at the bifurcation of the carotid artery. The logistic regression analysis was used to further analyze the risk factors associated with plaque vulnerability. Results: Among 148 carotid arteries,80 were classified as high-risk and 68 as low-risk groups. There were significant differences between the two groups concerning the thickness, length, maximum cross-sectional area, burden, and cross-sectional distribution of the plaques (P < 0.05). The plaque distribution on the dorsal side wall of the carotid bifurcation was higher in the high-risk group than that in the low-risk group (P < 0.05), dorsal side wall plaque-independent risk factors for the development of vulnerability of plaques in transient ischemic attack (TIA) patients (95% CI:1.522~6.991, P< 0.05). Conclusion: High-risk plaques tend to occur on the dorsal side wall of the carotid bifurcation, whereas low-risk plaques tend to occur on the outer side wall of the carotid bifurcation.
- Open Access
- 25 October 2023
- Published by Wolters Kluwer Health in Journal of Computer Assisted Tomography
- Vol. 48 (2) , 311-316
Pericarotid Fat Stranding at Computed Tomography Angiography: A Marker of the Short-Term Prognosis of Acute Ischemic Stroke
Purpose: Perivascular epicardial fat stranding detected in the coronary computed tomography (CT) angiography is associated with culprit lesions and provides helpful information on the risk of acute coronary syndrome. This study aimed to evaluate the potential clinical significance of pericarotid fat stranding (PCFS) and investigate the association between PCFS and short-term prognosis in acute stroke using head and neck CT angiography (CTA). Methods: This study included 80 patients (mean age, 69.69 ± 11.03; 58 men) who underwent both head and neck CTA and magnetic resonance imaging within a 1-week period. Baseline characteristics, pericarotid adipose tissue attenuation, plaque characteristics, ischemic penumbra, infarct core volume, infarct core growth rate (CGR), and the grade of collateral status were recorded and compared between a PCFS group and a non-PCFS group. Data were compared using the 2-sample t test, Mann-Whitney U test, Fisher exact test, and Spearman rank correlation analysis. Results: We found that patients with PCFS had a significantly higher pericarotid adipose tissue density than patients without PCFS (−55.75 ± 5.53 vs −65.82 ± 9.65, P < 0.001). Patients with PCFS showed a larger infarct core volume (166.43 ± 73.07 vs 91.43 ± 55.03, P = 0.001) and faster CGR (39.57 ± 12.01 vs 19.83 ± 32.77; P < 0.001), and the frequency of adverse prognosis was more significant than in control participants (83.33% vs 19.11%). Conclusions: Individuals with PCFS showed higher CGR, which was substantially related to worse outcomes in patients with acute stroke with ipsilateral carotid atherosclerosis. Recognition of PCFS may help predict stroke prognosis and allow doctors to take early action to improve patient prognosis.
- 25 October 2023
- Published by Wolters Kluwer Health in Journal of Computer Assisted Tomography
- Vol. 48 (2) , 311-316
Open AccessPericarotid Fat Stranding at Computed Tomography Angiography: A Marker of the Short-Term Prognosis of Acute Ischemic Stroke
Purpose: Perivascular epicardial fat stranding detected in the coronary computed tomography (CT) angiography is associated with culprit lesions and provides helpful information on the risk of acute coronary syndrome. This study aimed to evaluate the potential clinical significance of pericarotid fat stranding (PCFS) and investigate the association between PCFS and short-term prognosis in acute stroke using head and neck CT angiography (CTA). Methods: This study included 80 patients (mean age, 69.69 ± 11.03; 58 men) who underwent both head and neck CTA and magnetic resonance imaging within a 1-week period. Baseline characteristics, pericarotid adipose tissue attenuation, plaque characteristics, ischemic penumbra, infarct core volume, infarct core growth rate (CGR), and the grade of collateral status were recorded and compared between a PCFS group and a non-PCFS group. Data were compared using the 2-sample t test, Mann-Whitney U test, Fisher exact test, and Spearman rank correlation analysis. Results: We found that patients with PCFS had a significantly higher pericarotid adipose tissue density than patients without PCFS (−55.75 ± 5.53 vs −65.82 ± 9.65, P < 0.001). Patients with PCFS showed a larger infarct core volume (166.43 ± 73.07 vs 91.43 ± 55.03, P = 0.001) and faster CGR (39.57 ± 12.01 vs 19.83 ± 32.77; P < 0.001), and the frequency of adverse prognosis was more significant than in control participants (83.33% vs 19.11%). Conclusions: Individuals with PCFS showed higher CGR, which was substantially related to worse outcomes in patients with acute stroke with ipsilateral carotid atherosclerosis. Recognition of PCFS may help predict stroke prognosis and allow doctors to take early action to improve patient prognosis.
- Open Access
- 24 May 2023
- Published by Springer Nature in Research Square
A Case of Cerebral Cavernous Malformations Misdiagnosed as Multiple Myeloma
Cerebral cavernous malformations (CCMs) are angiographically occult cerebrovascular malformations (AOVM) of the central nervous system. They are composed of a sponge-like abnormal vascular mass with numerous thin-walled vessels. CCM incidence rate is approximately 7% of cerebrovascular malformations. The clinical manifestations mainly include seizures, bleeding, headache, etc., with seizures being the most common first symptom, and some patients may be without clinical manifestations. MRI has been the specific diagnostic method for CCMs. This article reports the case of a patient initially diagnosed with multiple myeloma in the left temporal lobe at our hospital in September 2020. Later, the patient was confirmed to have CCMs through surgery and pathology. The main symptom was a sudden onset of cognitive dysfunction, and the patient was discharged after receiving symptomatic and surgical interventions.
- 24 May 2023
- Published by Springer Nature in Research Square
Open AccessA Case of Cerebral Cavernous Malformations Misdiagnosed as Multiple Myeloma
Cerebral cavernous malformations (CCMs) are angiographically occult cerebrovascular malformations (AOVM) of the central nervous system. They are composed of a sponge-like abnormal vascular mass with numerous thin-walled vessels. CCM incidence rate is approximately 7% of cerebrovascular malformations. The clinical manifestations mainly include seizures, bleeding, headache, etc., with seizures being the most common first symptom, and some patients may be without clinical manifestations. MRI has been the specific diagnostic method for CCMs. This article reports the case of a patient initially diagnosed with multiple myeloma in the left temporal lobe at our hospital in September 2020. Later, the patient was confirmed to have CCMs through surgery and pathology. The main symptom was a sudden onset of cognitive dysfunction, and the patient was discharged after receiving symptomatic and surgical interventions.
- Open Access
- 4 January 2023
- Published by Springer Nature in Research Square
Pericarotid fat stranding at CT angiography:a marker of the short-term prognosis of acute ischemic stroke
Purpose Perivascular epicardial fat stranding detected in the coronary CT angiography was associated with culprit lesion and provided helpful information of risk for ACS. This study aimed to evaluate the potential clinical significance of pericarotid fat stranding (PCFS) and investigate the association between PCFS and short-term prognosis of acute stroke with head and neck CT angiography (CTA). Methods This study included 80 patients (mean age 69.69士11.03; 58 men) who underwent both head and neck CTA and magnetic resonance imaging (MRI) within 2 weeks. Baseline characteristics, PCAT attenuation, plaque characteristics, ischemic penumbra, infarct core volume, infarct core growth rate (CGR) and grade of collateral status between PCFS group and non-PCFS group were recorded, these data were compared by using two-sample t,Mann-Whitney U,Fisher tests and Spearman’s correlation test. Results We found that patients with PCFS had a significantly higher PCAT density than non-PCFS patients (-55.75 ± 5.53 vs -65.82 ± 9.65;P༜0.001 ). Those with PCFS showed a larger infarct core volume (166.43 ± 73.07 vs 91.43 ± 55.03;P = 0.001 ) and faster CGR (39.57 ± 12.01 vs 19.83 ± 32.77;P༜0.001), and that the frequency of adverse prognosis was more significant than in control subjects (83.33% vs 19.11%). The difference was statistically significant. Conclusions We discovered that individuals with PCFS had greater CGR, which was substantially related with a worse outcome in acute stroke patients with ipsilateral carotid atherosclerosis. Recognition of PCFS may help to predict stroke prognosis and allow doctors to take early action to improve patient prognosis.
- 4 January 2023
- Published by Springer Nature in Research Square
Open AccessPericarotid fat stranding at CT angiography:a marker of the short-term prognosis of acute ischemic stroke
Purpose Perivascular epicardial fat stranding detected in the coronary CT angiography was associated with culprit lesion and provided helpful information of risk for ACS. This study aimed to evaluate the potential clinical significance of pericarotid fat stranding (PCFS) and investigate the association between PCFS and short-term prognosis of acute stroke with head and neck CT angiography (CTA). Methods This study included 80 patients (mean age 69.69士11.03; 58 men) who underwent both head and neck CTA and magnetic resonance imaging (MRI) within 2 weeks. Baseline characteristics, PCAT attenuation, plaque characteristics, ischemic penumbra, infarct core volume, infarct core growth rate (CGR) and grade of collateral status between PCFS group and non-PCFS group were recorded, these data were compared by using two-sample t,Mann-Whitney U,Fisher tests and Spearman’s correlation test. Results We found that patients with PCFS had a significantly higher PCAT density than non-PCFS patients (-55.75 ± 5.53 vs -65.82 ± 9.65;P༜0.001 ). Those with PCFS showed a larger infarct core volume (166.43 ± 73.07 vs 91.43 ± 55.03;P = 0.001 ) and faster CGR (39.57 ± 12.01 vs 19.83 ± 32.77;P༜0.001), and that the frequency of adverse prognosis was more significant than in control subjects (83.33% vs 19.11%). The difference was statistically significant. Conclusions We discovered that individuals with PCFS had greater CGR, which was substantially related with a worse outcome in acute stroke patients with ipsilateral carotid atherosclerosis. Recognition of PCFS may help to predict stroke prognosis and allow doctors to take early action to improve patient prognosis.
- Open Access
- 27 August 2021
- Published by International Scientific Information, Inc. in Medical Science Monitor
- Vol. 27, e933351-1-e933351-10
Impact of H-Type Hypertension on Pericarotid Adipose Tissue and Plaque Characteristics Based on Computed Tomography (CT) Angiography: A Propensity Score Matching Study
Impact of H-Type Hypertension on Pericarotid Adipose Tissue and Plaque Characteristics Based on Computed Tomography (CT) Angiography: A Propensity Score Matching Study - Article abstract #933351
- 27 August 2021
- Published by International Scientific Information, Inc. in Medical Science Monitor
- Vol. 27, e933351-1-e933351-10
Open AccessImpact of H-Type Hypertension on Pericarotid Adipose Tissue and Plaque Characteristics Based on Computed Tomography (CT) Angiography: A Propensity Score Matching Study
Impact of H-Type Hypertension on Pericarotid Adipose Tissue and Plaque Characteristics Based on Computed Tomography (CT) Angiography: A Propensity Score Matching Study - Article abstract #933351
- 13 January 2021
- Published by International Scientific Information, Inc. in Medical Science Monitor
- Vol. 27, e929445-1-e929445-8
High-Resolution Magnetic Resonance Imaging (HR-MRI) Imaging Characteristics of Vertebral Artery Dissection with Negative MR Routine Scan and Hypoperfusion in Arterial Spin Labeling
High-Resolution Magnetic Resonance Imaging (HR-MRI) Imaging Characteristics of Vertebral Artery Dissection with Negative MR Routine Scan and Hypoperfusion in Arterial Spin Labeling - Article abstract #929445
- 13 January 2021
- Published by International Scientific Information, Inc. in Medical Science Monitor
- Vol. 27, e929445-1-e929445-8
High-Resolution Magnetic Resonance Imaging (HR-MRI) Imaging Characteristics of Vertebral Artery Dissection with Negative MR Routine Scan and Hypoperfusion in Arterial Spin Labeling
High-Resolution Magnetic Resonance Imaging (HR-MRI) Imaging Characteristics of Vertebral Artery Dissection with Negative MR Routine Scan and Hypoperfusion in Arterial Spin Labeling - Article abstract #929445
- Open Access
Clinically Applicable Deep Learning for Intracranial Aneurysm Detection in Computed Tomography Angiography Images: A Comprehensive Multicohort Study
Intracranial aneurysm is a common life-threatening disease. CTA is recommended as a standard diagnosis tool, while the interpretation is time-consuming and challenging. We presented a novel deep-learning-based framework trained on 1,177 DSA verified bone-removal CTA cases. The framework had excellent tolerance to the influence of occult cases of CTA-negative but DSA-positive aneurysms, image quality, and manufacturers. Simulated real-world studies were conducted in consecutive internal and external cohorts, achieving improved sensitivity and negative predictive value than radiologists. A specific cohort of suspected acute ischemic stroke was employed and found 96.8% predicted-negative cases can be trusted with high confidence, leading to reducing in human burden. A prospective study is warranted to determine whether the algorithm could improve patients’ care in comparison to radiologists’ assessment.
Open AccessClinically Applicable Deep Learning for Intracranial Aneurysm Detection in Computed Tomography Angiography Images: A Comprehensive Multicohort Study
Intracranial aneurysm is a common life-threatening disease. CTA is recommended as a standard diagnosis tool, while the interpretation is time-consuming and challenging. We presented a novel deep-learning-based framework trained on 1,177 DSA verified bone-removal CTA cases. The framework had excellent tolerance to the influence of occult cases of CTA-negative but DSA-positive aneurysms, image quality, and manufacturers. Simulated real-world studies were conducted in consecutive internal and external cohorts, achieving improved sensitivity and negative predictive value than radiologists. A specific cohort of suspected acute ischemic stroke was employed and found 96.8% predicted-negative cases can be trusted with high confidence, leading to reducing in human burden. A prospective study is warranted to determine whether the algorithm could improve patients’ care in comparison to radiologists’ assessment.