The Journal of Practical Medicine >

2026 , Vol. 42 >Issue 11: 1906 - 1914

DOI: https://doi.org/10.3969/j.issn.1006-5725.2026.11.003

Cardiovascular and Cerebrovascular Diseases Column

Development and validation of a model based on AI and imaging features from non-contrast chest-abdomen CT for diagnosing aortic dissection

  • Shuxin LI ,
  • Rongli ZHANG ,
  • Xiongfeng QIU ,
  • Zhenyun HU ,
  • Weiming WU ,
  • Huilin JIANG ,
  • Min LI ,
  • Huai CHEN
Expand
  • 1.Department of Radiology，the Second Affiliated Hospital of Guangzhou Medical University，Guangzhou 510260，Guangdong，China
    3.School of Biomedical Engineering，Guangzhou Medical University，Guangzhou 511436，Guangdong，China
    2.Department of Emergency Medicine，the Second Affiliated Hospital of Guangzhou Medical University，Guangzhou 510260，Guangdong，China

Received date: 2026-01-07

  Online published: 2026-06-15

Fold

Abstract

Objective To develop and validate a combined diagnostic model for aortic dissection （AD） by integrating artificial intelligence （AI）?derived prediction probabilities with interpretable imaging features from non-contrast computed tomography （CT）， thereby enhancing its clinical utility for AD evaluation. Methods We conducted a retrospective study of 221 patients with suspected AD who underwent non-contrast chest and abdominal CT at our institution between July 2020 and February 2024. Patients were randomly allocated at a 7∶3 ratio into a training cohort （n = 153， 76 AD cases） and a validation cohort （n = 68， 33 AD cases）. Clinical variables and CT imaging features were extracted. Independent predictors of AD were identified using multivariate logistic regression. The final model incorporated younger age as the sole clinical predictor， alongside key imaging features. Three diagnostic models were constructed： a traditional feature-based model， a standalone AI model， and a combined model integrating both imaging features and AI-derived prediction probabilities. Diagnostic performance was evaluated using receiver operating characteristic （ROC） curve analysis， with area under the curve （AUC） values compared via the DeLong test. Model calibration was assessed using the Hosmer-Lemeshow test and calibration plots. Results Multivariate analysis identified younger age， ascending aortic dilation， crescentic high-attenuation intramural density， and an intimal flap as independent predictors of AD （all P < 0.05）. In the training cohort， the combined model yielded a significantly higher AUC than both the traditional model and the standalone AI model （0.921 vs. 0.897 vs. 0.779， respectively， all P < 0.05）. This advantage was confirmed in the validation cohort， where the combined model achieved an AUC of 0.841， with a sensitivity of 84.8% and specificity of 77.1%. All models demonstrated adequate calibration （Hosmer-Lemeshow P > 0.05）， and the calibration plot for the combined model showed strong agreement between predicted probabilities and actual AD diagnoses confirmed by aortic CTA. Conclusions Integrating AI-derived probabilities with interpretable non-contrast CT imaging features significantly enhances AD detection accuracy. This combined model offers a robust， non-invasive tool to aid in the rapid diagnosis and clinical triage of suspected AD.

Key words： aortic dissection; non-contrast CT; artificial intelligence; imaging features; diagnostic model

Cite this article

Shuxin LI , Rongli ZHANG , Xiongfeng QIU , Zhenyun HU , Weiming WU , Huilin JIANG , Min LI , Huai CHEN . Development and validation of a model based on AI and imaging features from non-contrast chest-abdomen CT for diagnosing aortic dissection[J]. The Journal of Practical Medicine, 2026 , 42(11) : 1906 -1914 . DOI: 10.3969/j.issn.1006-5725.2026.11.003

References

[1] MAZZOLAI L， TEIXIDO-TURA G， LANZI S， et al. 2024 ESC Guidelines for the management of peripheral arterial and aortic diseases［J］. Eur Heart J， 2024， 45（36）： 3538-3700. doi： 10.1093/eurheartj/ehae179 .
[2] 中华医学会放射学分会介入学组. 中国Stanford B型主动脉夹层影像诊断和介入治疗临床指南［J］. 中华放射学杂志， 2023， 57（5）： 457-473. doi： 10.3760/cma.j.cn112149-20230211-00091
[3] HATA A， YANAGAWA M， YAMAGATA K， et al. Deep learning algorithm for detection of aortic dissection on non-contrast-enhanced CT［J］. Eur Radiol， 2021， 31： 1151-1159. doi： 10.1007/s00330-020-07213-w .
[4] CAO L， SHI R， GE Y， et al. Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning［J］. Eur J Radiol， 2019， 121： 108713. doi： 10.1016/j.ejrad. 2019.108713 .
[5] 张亚楠， 崔磊， 薛颖， 等. 基于胸部CT平扫和临床参数的主动脉夹层预测模型的建立与初步验证［J］. 国际医学放射学杂志， 2023， 46（5）： 516-522. doi： 10.19300/j.2023.L20359 .
[6] 张云芳， 李铮， 聂晓改， 等. D-二聚体/白蛋白、白细胞介素-6、纤维蛋白降解产物联合预测A型主动脉夹层术后不良结局［J］. 实用医学杂志，2025， 41（17）： 2755-2760. doi： 10.3969/j.issn.1006-5725.2025.17.021 .
[7] SAKAMOTO S， TANIGUCHI N， NAKAJIMA S， et al. Diagnostic value of nonenhanced multidetector computed tomography for ruling out acute aortic dissection in patients presenting with chest or back pain［J］. Int J Cardiol， 2013， 168（2）： 734-738. doi： 10.1016/j.ijcard.2012.09.219 .
[8] XIONG X， DING Y， SUN C， et al. A Cascaded Multi-Task Generative Framework for Detecting Aortic Dissection on 3-D Non-Contrast-Enhanced Computed Tomography［J］. IEEE J Biomed Health Inform， 2022， 26（10）： 5177-5188. doi： 10.1109/JBHI.2022.3190297 .
[9] CLOUSE W D， HALLETT J W J R， SCHAFF H V， et al. Acute aortic dissection： population-based incidence compared with degenerative aortic aneurysm rupture［J］. Mayo Clin Proc， 2004， 79（2）： 176-180. doi： 10.4065/79.2.176 .
[10] GOLLEDGE J， EAGLE K A. Acute aortic dissection［J］. Lancet， 2008， 372（9632）： 55-66. doi： 10.1016/S0140-6736（08）60994-0 .
[11] 刘冰， 陈瀚熙， 刘汶佶， 等. Stanford A型主动脉夹层术后患者疾病感知及应对方式对社会参与的影响［J］. 实用医学杂志， 2024， 40（8）： 1160-1166. doi： 10.3969/j.issn.1006-5725.2024.08.024 .
[12] HAGAN P G， NIENABER C A， ISSELBACHER E M， et al. The International Registry of Acute Aortic Dissection （IRAD）： New insights into an old disease［J］. JAMA， 2000， 283（7）： 897-903. doi： 10.1001/jama.283.7.897 .
[13] IBRAHIM I， CHUA M T， TAN D W， et al. Impact of 24-hour specialist coverage and an on-site CT scanner on the timely diagnosis of acute aortic dissection［J］. Singapore Med J， 2020， 61（2）： 86-91. doi： 10.11622/smedj.2019039 .
[14] OTANI T， ICHIBA T， KASHIWA K， et al. Potential of unenhanced computed tomography as a screening tool for acute aortic syndromes［J］. Eur Heart J Acute Cardiovasc Care， 2021， 10（9）： 967-975. doi： 10.1093/ehjacc/zuab069 .
[15] 陈丽萍. CT平扫、胸腹部CT血管造影技术鉴别主动脉夹层的结果研究［J］. 影像研究与医学应用， 2024， 8（24）： 111-113. doi： 10.3969/j.issn.2096-3807. 2024. 24.033 .
[16] DI GIOIA C R T， ASCIONE A， CARLETTI R， et al. Thoracic Aorta： Anatomy and Pathology［J］. Diagnostics， 2023， 13（13）： 2166.doi：10.3390/diagnostics 13132166 .
[17] CHAO C P， WALKER T G， KALVA S P. Natural history and CT appearances of aortic intramural hematoma［J］. Radiographics， 2009， 29（3）： 791-804. doi： 10.1148/rg.293085122 .
[18] KURABAYASHI M， OKISHIGE K， UESHIMA D， et al. Diagnostic utility of unenhanced computed tomography for acute aortic syndrome［J］. Circ J， 2014， 78（8）： 1928-1934. doi： 10.1253/circj.cj-14-0198 .
[19] SPANGENBERG A， RAO S J， MACKRELL J， et al. Type A Aortic Dissection and Non-Contrast Computed Tomography［J］. J Community Hosp Intern Med Perspect， 2023， 13（3）： 118-120. doi：10.55729/2000-9666.1178 .
[20] 张雪群， 许金坨， 曾森斌. 主动脉夹层在螺旋CT平扫中的特征性征象总结分析［J］. 现代医用影像学， 2024， 33（3）： 408-410. doi： 10.3969/j.issn.1006-7035. 2024.03.004
[21] TANG Z， HUANG Y， HU S， et al. Deep Learning Application of YOLOv8 for Aortic Dissection Screening using Non-contrast Computed Tomography［J］. Eur J Vasc Endovasc Surg， 2025： S1078-5884（25）00817-2. doi： 10.1016/j.ejvs. 2025.08.054 .
[22] KIM Y S， KIM J G， CHOI H Y， et al. Detection of Aortic Dissection and Intramural Hematoma in Non-Contrast Chest Computed Tomography Using a You Only Look Once-Based Deep Learning Model［J］. J Clin Med， 2024， 13（22）：6868. doi：10.3390/jcm13226868 .
[23] DONG F， SONG J， CHEN B， et al. Improved detection of aortic dissection in non-contrast-enhanced chest CT using an attention-based deep learning model［J］. Heliyon， 2024， 10（2）： e24547. doi：10.1016/j.heliyon.2024.e24547 .
[24] LI Z， CHEN L， ZHANG S， et al. An innovative bimodal computed tomography data-driven deep learning model for predicting aortic dissection： A multi-center study［J］. Quant Imaging Med Surg， 2025， 15（9）： 8359-8371. doi：10.21037/ qims-2024-2807 .
[25] TANG J， CHEN F， WU D. Early auxiliary diagnosis model for chest pain triad based on artificial intelligence multimodal fusion［J］. JAMIA Open， 2025， 8（5）： ooaf114. doi：10.1093/jamiaopen/ooaf114 .
[26] YI Y， MAO L， WANG C， et al. Advanced Warning of Aortic Dissection on Non-Contrast CT： The Combination of Deep Learning and Morphological Characteristics［J］. Front Cardiovasc Med， 2022， 8： 762958. doi： 10.3389/fcvm.2021.762958 .
[27] HU Y， XIANG Y， ZHOU Y J， et al. AI-based diagnosis of acute aortic syndrome from noncontrast CT［J］. Nat Med， 2025， 31（11）： 3832-3844. doi： 10.1038 /s41591-025-03916-z .
Options
Outlines

/

〈 〉