"Pilot randomized prospective clinical study of the effectiveness of the use of artificial intelligence in determining "safe" clamping zones in the surgical treatment of abdominal aortic aneurysms."

Abdominal aortic aneurysm is a life-threatening disease, a formidable complication of which is an aneurysm rupture (Editor's Choice - European Society for Vascular Surgery). The main method of treating aneurysms is surgical reconstruction, including open or endovascular intervention ((ESVS) 2019 Clinical Practice Guidelines on the Management of Abdominal Aorto-iliac Artery Aneurysms). Anatomical features of aneurysms and the presence of intraluminal thrombomass are among the criteria in deciding on the tactics of surgical treatment. These factors carry additional technical difficulties and lead to the development of intraoperative complications, including ischemic ones. Ischemia of the lower extremities is the most common complication and can be caused by thrombosis, embolism or dissection of the aortic wall (occurs in 7% of patients) (Complications Associated with Aortic Aneurysm Repair).

Thus, in order to reduce the frequency of embolic complications, it is important for the surgeon to determine a "safe" zone for applying a clamp to the aorta and main vessels. Thus, artificial intelligence (AI) can be used to interpret and analyze images of aneurysms that allow automatic quantitative measurements and determination of the exact characteristics of morphology and hydrodynamics, as well as the presence of intraluminal blood clots and calcifications. Analysis based on artificial intelligence can lead to the development of computational programs for predicting the development of aneurysms and the risk of their rupture, as well as postoperative outcomes. Artificial intelligence can also be used to determine the "safe" areas of aortic clamping. (Artificial intelligence in abdominal aortic aneurysm).

Adam and co-authors trained a neural network to detect and estimate the maximum outer diameter of aneurysms using a database of 489 CT angiographs of abdominal aortic aneurysms. AI has achieved a level of performance and accuracy suitable for clinical practice, and with the use of more CT images, further improvement in accuracy is expected (Pre-surgical and Post-surgical Aortic Aneurysm Maximum Diameter Measurement: Full Automation by Artificial Intelligence). In a study by Fujiwara et al. 145 non-contrast CT scans with suspected aneurysm were retrospectively collected. Initially, AI was trained by manually segmenting CT images. Image processing was used to determine the abdominal aortic aneurysm area and to automatically measure the size. This method has shown that AI is a useful tool for fully automatic detection and measurement of aneurysm diameter. (Fully automatic detection and measurement of abdominal aortic aneurysm using artificial intelligence). Florent Lalys and his coauthor. an automatic fast and universal algorithm for determining an intraluminal thrombus was developed. The method was tested on pre- and postoperative CT scans of the abdominal aorta and iliac artery of 145 patients and consists in determining the central line and segmentation of the aortic lumen, an optimized stage of pretreatment and the use of a 3D model (Generic thrombus segmentation from pre- and post-operative CTA).

Taking into account the references already available in some studies of the use of artificial intelligence for the treatment of cardiovascular diseases, its use is seen as a promising method for making decisions in determining "safe" clamping zones in the surgical treatment of abdominal aortic aneurysms, which will reduce the frequency of postoperative complications.

Inclusion Criteria:

• Patients with aneurysmal dilation of the abdominal aorta, who are shown surgery.
• Patients who have agreed to participate in this study

Exclusion Criteria:

• Chronic heart failure of functional class III -IV according to NYHA classification;
• Chronic decompensated "pulmonary" heart;
• Severe hepatic or renal insufficiency (bilirubin >35 mmol/l, glomerular filtration rate <60 ml/min);
• Polyvalent drug allergy;
• Malignant oncological diseases in the terminal stage with a predicted life span of up to 6 months;
• Acute cerebrovascular accident;

after the end of the study, receipt of the results and publication of the first data, these sections will be available to other researchers upon personal request.

1 year after the first publication

access to data on a personal request