1
|
Yang Q, Hu J, Luo Y, Jia D, Chen N, Yao C, Wu R. Detection of Endoleak after Endovascular Aortic Repair through Deep Learning Based on Non-contrast CT. Cardiovasc Intervent Radiol 2024:10.1007/s00270-024-03805-x. [PMID: 38977447 DOI: 10.1007/s00270-024-03805-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/28/2024] [Accepted: 06/25/2024] [Indexed: 07/10/2024]
Abstract
OBJECTIVES To develop and validate a deep learning model for detecting post-endovascular aortic repair (EVAR) endoleak from non-contrast CT. METHODS This retrospective study involved 245 patients who underwent EVAR between September 2016 and December 2022. All patients underwent both non-enhanced and enhanced follow-up CT. The presence of endoleak was evaluated based on computed tomography angiography (CTA) and radiology reports. First, the aneurysm sac was segmented, and radiomic features were extracted on non-contrast CT. Statistical analysis was conducted to investigate differences in shape and density characteristics between aneurysm sacs with and without endoleak. Subsequently, a deep learning model was trained to generate predicted segmentation of the endoleak. A binary decision was made based on whether the model produced a segmentation to detect the presence of endoleak. The absence of a predicted segmentation indicated no endoleak, while the presence of a predicted segmentation indicated endoleak. Finally, the performance of the model was evaluated by comparing the predicted segmentation with the reference segmentation obtained from CTA. Model performance was assessed using metrics such as dice similarity coefficient, sensitivity, specificity, and the area under the curve (AUC). RESULTS This study finally included 85 patients with endoleak and 82 patients without endoleak. Compared to patients without endoleak, patients with endoleak had higher CT values and greater dispersion. The AUC in validation group was 0.951, dice similarity coefficient was 0.814, sensitivity was 0.877, and specificity was 0.884. CONCLUSION This deep learning model based on non-contrast CT can detect endoleak after EVAR with high sensitivity.
Collapse
Affiliation(s)
- Qingqi Yang
- Department of Vascular Surgery, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China
| | - Jinglang Hu
- School of Medicine, Sun Yat-Sen University, Guangzhou, China
| | - Yingqi Luo
- Department of Medical Imaging, The Second Affiliated Hospital, Guangzhou Medical University, Guangzhou, China
| | - Dongdong Jia
- Department of Vascular Surgery, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China
| | - Nuo Chen
- Department of Vascular Surgery, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China
| | - Chen Yao
- Department of Vascular Surgery, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China
| | - Ridong Wu
- Department of Vascular Surgery, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China.
| |
Collapse
|
2
|
Hu G, Ding N, Wang Z, Jin Z. Unenhanced computed tomography radiomics help detect endoleaks after endovascular repair of abdominal aortic aneurysm. Eur Radiol 2024; 34:1647-1658. [PMID: 37658886 PMCID: PMC10873228 DOI: 10.1007/s00330-023-10000-y] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/08/2022] [Revised: 05/03/2023] [Accepted: 06/05/2023] [Indexed: 09/05/2023]
Abstract
OBJECTIVES To explore the feasibility of unenhanced CT images for endoleak detection of abdominal aortic aneurysm (AAA) after endovascular repair (EVAR). METHODS Patients who visited our hospital after EVAR from July 2014 to September 2021 were retrospectively collected. Two radiologists evaluated the presence or absence of endoleaks using the combination of contrast-enhanced and unenhanced CT as the referenced standard. After segmenting the aneurysm sac of the unenhanced CT, the radiomic features were automatically extracted from the region of interest. Histogram features of patients with and without endoleak were statistically analyzed to explore the differences between the two groups. Twelve common machine learning (ML) models based on radiomic features were constructed to evaluate the performance of endoleak detection with unenhanced CT images. RESULTS The study included 216 patients (69 ± 8 years; 191 men) with AAA, including 64 patients with endoleaks. A total of 1955 radiomic features of unenhanced CT were extracted. Compared with patients without endoleak, the aneurysm sac outside the stent of patients with endoleak had higher CT attenuation (41.7 vs. 33.6, p < 0.001) with smaller dispersion (51.5 vs. 58.8, p < 0.001). The average area under the curve (AUC) of the ML models constructed with unenhanced CT radiomics was 0.86 ± 0.05, the accuracy was 81% ± 4, the sensitivity was 88% ± 10, and the specificity was 78% ± 5. When fixing the sensitivity to > 90% (92% ± 2), the models retained specificity at 72% ± 10. CONCLUSIONS Unenhanced CT features exhibit significant differences between patients with and without endoleak and can help detect endoleaks in AAA after EVAR with high sensitivity. CLINICAL RELEVANCE STATEMENT Unenhanced CT radiomics can help provide an alternative method of endoleak detection in patients who have adverse reactions to contrast media. This study further exploits the value of unenhanced CT examinations in the clinical management and surveillance of postoperative abdominal aortic aneurysm. KEY POINTS • Unenhanced CT features of the aneurysm sac outside the stent exhibit significant differences between patients with and without endoleak. The endoleak group showed higher unenhanced CT attenuation (41.7 vs 33.6, p < .001) with smaller dispersion (51.5 vs 58.8, p < .001) than the nonendoleak group. • Unenhanced CT radiomics can help detect endoleaks after intervention. The average area under the curve (AUC) of twelve common machine learning models constructed with unenhanced CT radiomics was 0.86 ± 0.05, the average accuracy was 81% ± 4. • When fixing the sensitivity to > 90% (92% ± 2), the machine learning models retained average specificity at 72% ± 10.
Collapse
Affiliation(s)
- Ge Hu
- Medical Research Center, State Key Laboratory of Complex Severe and Rare Disease, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Dongcheng Dist, Beijing, 100730, China
| | - Ning Ding
- Department of Radiology, State Key Laboratory of Complex Severe and Rare Disease, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Dongcheng Dist, Beijing, 100730, China
| | - Zhiwei Wang
- Department of Radiology, State Key Laboratory of Complex Severe and Rare Disease, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Dongcheng Dist, Beijing, 100730, China.
| | - Zhengyu Jin
- Department of Radiology, State Key Laboratory of Complex Severe and Rare Disease, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Dongcheng Dist, Beijing, 100730, China.
| |
Collapse
|
3
|
Li L, Wang X, Xin X, Fan M, Lu S, Wang W, Yin G. Application report of automatic unlocking baseplate in radiotherapy. J Appl Clin Med Phys 2022; 23:e13778. [PMID: 36094026 PMCID: PMC9588263 DOI: 10.1002/acm2.13778] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/31/2022] [Revised: 07/31/2022] [Accepted: 08/16/2022] [Indexed: 12/03/2022] Open
Abstract
Purpose To reduce the potential risk during radiotherapy treatment of patients with head and neck tumors, we improved upon the design of an existing immobilization device by adding a feature to improve patient safety during emergency releases, and we verified its clinical application. Method We designed an improved automatic unlocking baseplate (AUB), and conducted a dosimetry comparison with Solo Align Full Body System (SAFBS, Klarity, China). The dosimetry comparison included dose‐attenuation measurements and results from human simulation. We selected four points for measurement to allow comparison between the SAFBS and our AUB. A simulated human body model was used for CT scanning, whereby the target area and structure and simulated radiotherapy plan were conducted according to the American Academy of Pain Medicine Task Group–119 report (TG‐119), whereby the dose differences were compared. The purpose of the clinical test was to verify the reliability of the AUB system in practical clinical applications. The application tests were conducted in CT simulation (CT‐sim) and treatment rooms. The test included assessments of the stability of the system and the reliability of our device. Results The dose‐attenuation measurements of the two baseplates were as follows: The transmission values with our unlocking system were 0.10% higher at the first point and 0.67% lower at the third. The same dose was obtained at points 2 and 4. In the simulation study, the PTV of the AUB was lower than that of the SAFBS, including 0.39% lower D99 and 0.18% lower D90. Among the organ‐at‐risk doses, the average dose of the AUB in the spinal cord was 0.6% higher than that of the SAFBS, and the average dose in the left and right parotid glands was more than 1.4% lower than that of SAFBS. The clinical test results were applied in treatment room and a CT‐sim room, which show a 100% success rate after being unlocked more than 5000 times. Conclusion The AUB designed for head and neck patients had good functional versatility, the dose distribution met the requirements, and the automatic unlocking function was demonstrated to be stable and reliable.
Collapse
Affiliation(s)
- Lintao Li
- Department of Radiation Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.,Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China
| | - Xianliang Wang
- Department of Radiation Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.,Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China
| | - Xin Xin
- Department of Radiation Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.,Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China
| | - Ming Fan
- Department of Radiation Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.,Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China
| | - Shun Lu
- Department of Radiation Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.,Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China
| | - Wei Wang
- Klrity Medical&Equipment Co. Ltd., Guangzhou, China
| | - Gang Yin
- Department of Radiation Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.,Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China
| |
Collapse
|
4
|
Pediatric magnetic resonance angiography: to contrast or not to contrast. Pediatr Radiol 2022:10.1007/s00247-022-05467-8. [PMID: 35953543 DOI: 10.1007/s00247-022-05467-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/04/2022] [Revised: 06/26/2022] [Accepted: 07/21/2022] [Indexed: 10/15/2022]
Abstract
Magnetic resonance (MR) angiography and MR venography imaging with contrast and non-contrast techniques are widely used for pediatric vascular imaging. However, as with any MRI examination, imaging the pediatric population can be challenging because of patient motion, which sometimes requires sedation. There are multiple benefits of non-contrast MR angiographic techniques, including the ability to repeat sequences if motion is present, the decreased need for sedation, and avoidance of potential risks associated with gadolinium administration and radiation exposure. Thus, MR angiography is an attractive alternative to CT or conventional catheter-based angiography in pediatric populations. Contrast-enhanced MR angiographic techniques have the advantage of increased signal to noise. Blood pool imaging allows long imaging times that result in high-spatial-resolution imaging, and thus high-quality diagnostic images. This article outlines the technique details, indications, benefits and downsides of non-contrast-enhanced and contrast-enhanced MR angiographic techniques to assist in protocol decision-making.
Collapse
|