Automatic risk prediction of intracranial aneurysm on CTA image with convolutional neural networks and radiomics analysis

Abstract

BackgroundIntracranial aneurysm (IA) is a nodular protrusion of the arterial wall caused by the localized abnormal enlargement of the lumen of a brain artery, which is the primary cause of subarachnoid hemorrhage. Accurate rupture risk prediction can effectively aid treatment planning, but conventional rupture risk estimation based on clinical information is subjective and time-consuming.MethodsWe propose a novel classification method based on the CTA images for differentiating aneurysms that are prone to rupture. The main contribution of this study is that the learning-based method proposed in this study leverages deep learning and radiomics features and integrates clinical information for a more accurate prediction of the risk of rupture. Specifically, we first extracted the provided aneurysm regions from the CTA images as 3D patches with the lesions located at their centers. Then, we employed an encoder using a 3D convolutional neural network (CNN) to extract complex latent features automatically. These features were then combined with radiomics features and clinical information. We further applied the LASSO regression method to find optimal features that are highly relevant to the rupture risk information, which is fed into a support vector machine (SVM) for final rupture risk prediction.ResultsThe experimental results demonstrate that our classification method can achieve accuracy and AUC scores of 89.78% and 89.09%, respectively, outperforming all the alternative methods.DiscussionOur study indicates that the incorporation of CNN and radiomics analysis can improve the prediction performance, and the selected optimal feature set can provide essential biomarkers for the determination of rupture risk, which is also of great clinical importance for individualized treatment planning and patient care of IA.

Classification, detection, and segmentation performance of image-based AI in intracranial aneurysm: a systematic review

“…(2) Based on imaging features [ 22 – 24 , 4 , 9 , 25 – 35 ]: using deep learning techniques, such as convolutional neural networks (CNNs), to learn feature representations from intracranial aneurysm images and apply them to classification tasks. This method does not require manual feature extraction but learns feature representations from raw image data through end-to-end learning, which can better capture the complex features of intracranial aneurysms and improve classification performance.…”

“…A total of 21 studies implemented IA recognition based on classification models [ 4 , 9 , 17 – 35 ], as shown in Table 1 . Combined with Tables 2 , 3 , and 4 , it can be seen that classification is currently the predominant method for IA recognition.…”

“…Medical imaging techniques play a significant role in the detection, risk assessment, treatment, and prognosis of intracranial aneurysms. Key imaging techniques used include computed tomography angiography (CTA) [ 4 , 5 ], magnetic resonance angiography (MRA) [ 6 , 7 ], digital subtraction angiography (DSA) [ 8 , 9 ], and the examples are shown in Fig. 1 .…”

Advances in research and application of artificial intelligence and radiomic predictive models based on intracranial aneurysm images

“…The results demonstrated that all three models exhibited high accuracy and were effective in discerning the status of aneurysms. Xie et al ( 36 ) combined features extracted by CNN with radiomics features and patient clinical information, employing LASSO regression to select important feature variables for constructing an SVM-based aneurysm rupture risk prediction model. The accuracy and AUC were 89.78 and 89.09%, respectively.…”

“…Although clinical outcomes of aneurysms are influenced by various factors such as individual patient characteristics, severity of hemorrhage, different treatment modalities, and preoperative, intraoperative, and postoperative management ( 36 ), the application of radiomics and artificial intelligence in predicting the prognosis of patients with intracranial aneurysms offers new personalized treatment methods and holds great potential.…”