Prediction of Intracranial Aneurysm Risk using Machine Learning

Heo, Ji Yeon; Park, Sang Jun; Kang, Si Hyuck; Oh, Chang Wan; Bang, Jae Seung; Kim, Tackeun

doi:10.1038/s41598-020-63906-8

Cited by 33 publications

(27 citation statements)

References 36 publications

(44 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Current UIA screening guidelines in the United States and Korea are limited to two categories: (1) patients with at least two family members with UIA or SAH, and (2) patients with a history of autosomal dominant polycystic kidney disease (ADPKD), coarctation of the aorta, or microcephalic osteodysplastic primordial dwarfism. Heo et al ( 14 ) extracted data from the National Health Screening Program in Korea containing general health examinations from 2009 to 2013. Using 21 variables from this data, Logistic Regression (LR), Random Forest (RF), eXtreme Gradient Boosting (XGB), and Deep Neural Network (DNN) were trained, among which the highest area under receiver operating curve (AUROC) value was achieved by the XGB algorithm (0.765) (95% CI 0.742–0.788).…”

Section: Ai In Intracranial Aneurysm Screeningmentioning

confidence: 99%

See 1 more Smart Citation

Role of Artificial Intelligence in Unruptured Intracranial Aneurysm: An Overview

Marasini¹,

Shrestha

Phuyal

et al. 2022

Front. Neurol.

View full text Add to dashboard Cite

Intracranial aneurysms (IAs) are a significant public health concern. In populations without comorbidity and a mean age of 50 years, their prevalence is up to 3.2%. An efficient method for identifying subjects at high risk of an IA is warranted to provide adequate radiological screening guidelines and effectively allocate medical resources. Artificial intelligence (AI) has received worldwide attention for its impressive performance in image-based tasks. It can serve as an adjunct to physicians in clinical settings, improving diagnostic accuracy while reducing physicians' workload. AI can perform tasks such as pattern recognition, object identification, and problem resolution with human-like intelligence. Based on the data collected for training, AI can assist in decisions in a semi-autonomous manner. Similarly, AI can identify a likely diagnosis and also, select a suitable treatment based on health records or imaging data without any explicit programming (instruction set). Aneurysm rupture prediction is the holy grail of prediction modeling. AI can significantly improve rupture prediction, saving lives and limbs in the process. Nowadays, deep learning (DL) has shown significant potential in accurately detecting lesions on medical imaging and has reached, or perhaps surpassed, an expert-level of diagnosis. This is the first step to accurately diagnose UIAs with increased computational radiomicis. This will not only allow diagnosis but also suggest a treatment course. In the future, we will see an increasing role of AI in both the diagnosis and management of IAs.

show abstract

Section: Ai In Intracranial Aneurysm Screeningmentioning

confidence: 99%

“…This risk stratification with the help of AI models will help in improved targets for screening. This targeted screening in the future with the use of multimodal data: health status & history, family and similar population imaging analysis, genetics will eventually create new guidelines for us to follow ( 14 ).…”

Section: Ai In Intracranial Aneurysm Screeningmentioning

confidence: 99%

Role of Artificial Intelligence in Unruptured Intracranial Aneurysm: An Overview

Marasini¹,

Shrestha

Phuyal

et al. 2022

Front. Neurol.

View full text Add to dashboard Cite

show abstract

“…Categorical Anti-tTG2 IgA deposit in duodenal mucosa at time of diagnosis: low positivity www.nature.com/scientificreports/ diet starting from these features before the development of the full blown CD. ML indications can move towards precision medicine also the detection of CD, as done in other diseases with similar workflows, as shown for the evaluation of cardiometabolic risk and risk of developing diabetes [2][3][4][5][6][7][31][32][33][34] . Celiac Disease automated diagnosis is not new to computer-assisted systems, which have been explored since 2008 35 ; spatial domain, transform domain, scale-invariant and and spatio-temporal features have been applied to several aspects of CD diagnosis, especially to the subjective interpretation of the intestine small mucosal immaginery 36 .…”

Section: F2mentioning

confidence: 99%

“…morphological changes in the small bowel mucosa [1][2][3][4][5][6][7] . Only a small percentage of them showed significant clinical symptoms (and are started on a gluten free diet at time of diagnosis), while the majority progressed over several years (up to a decade) without any clinical problem or a progression of the small intestinal mucosal damage even if they continued a gluten containing diet, on long term follow up one third of them progressed to a clear pattern of CD mucosal damage.…”

mentioning

confidence: 99%

Precision medicine and machine learning towards the prediction of the outcome of potential celiac disease

Piccialli

Calabrò

Crisci

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Potential Celiac Patients (PCD) bear the Celiac Disease (CD) genetic predisposition, a significant production of antihuman transglutaminase antibodies, but no morphological changes in the small bowel mucosa. A minority of patients (17%) showed clinical symptoms and need a gluten free diet at time of diagnosis, while the majority progress over several years (up to a decade) without any clinical problem neither a progression of the small intestine mucosal damage even when they continued to assume gluten in their diet. Recently we developed a traditional multivariate approach to predict the natural history, on the base of the information at enrolment (time 0) by a discriminant analysis model. Still, the traditional multivariate model requires stringent assumptions that may not be answered in the clinical setting. Starting from a follow-up dataset available for PCD, we propose the application of Machine Learning (ML) methodologies to extend the analysis on available clinical data and to detect most influent features predicting the outcome. These features, collected at time of diagnosis, should be capable to classify patients who will develop duodenal atrophy from those who will remain potential. Four ML methods were adopted to select features predictive of the outcome; the feature selection procedure was indeed capable to reduce the number of overall features from 85 to 19. ML methodologies (Random Forests, Extremely Randomized Trees, and Boosted Trees, Logistic Regression) were adopted, obtaining high values of accuracy: all report an accuracy above 75%. The specificity score was always more than 75% also, with two of the considered methods over 98%, while the best performance of sensitivity was 60%. The best model, optimized Boosted Trees, was able to classify PCD starting from the selected 19 features with an accuracy of 0.80, sensitivity of 0.58 and specificity of 0.84. Finally, with this work, we are able to categorize PCD patients that can more likely develop overt CD using ML. ML techniques appear to be an innovative approach to predict the outcome of PCD, since they provide a step forward in the direction of precision medicine aimed to customize healthcare, medical therapies, decisions, and practices tailoring the clinical management of PCD children.

show abstract

“…Recently, owing to the development of artificial intelligence, technologies such as machine learning and deep learning have been applied to medical fields 3–6. Artificial intelligence has been used to aid the detection of cerebral aneurysms and also to predict ruptures 5 6.…”

Section: Introductionmentioning

confidence: 99%

Rupture risk prediction of cerebral aneurysms using a novel convolutional neural network-based deep learning model

Yang

Cho

Kim

et al. 2022

J NeuroIntervent Surg

View full text Add to dashboard Cite

BackgroundCerebral aneurysms should be treated before rupture because ruptured aneurysms result in serious disability. Therefore, accurate prediction of rupture risk is important and has been estimated using various hemodynamic factors.ObjectiveTo suggest a new way to predict rupture risk in cerebral aneurysms using a novel deep learning model based on hemodynamic parameters for better decision-making about treatment.MethodsA novel convolutional neural network (CNN) model was used for rupture risk prediction retrospectively of 123 aneurysm cases. To include the effect of hemodynamic parameters into the CNN, the hemodynamic parameters were first calculated using computational fluid dynamics and fluid–structure interaction. Then, they were converted into images for training the CNN using a novel approach. In addition, new data augmentation methods were devised to obtain sufficient training data. A total of 53,136 images generated by data augmentation were used to train and test the CNN.ResultsThe CNNs trained with wall shear stress (WSS), strain, and combination images had area under the receiver operating characteristics curve values of 0.716, 0.741, and 0.883, respectively. Based on the cut-off values, the CNN trained with WSS (sensitivity: 0.5, specificity: 0.79) or strain (sensitivity: 0.74, specificity: 0.71) images alone was not highly predictive. However, the CNN trained with combination images of WSS and strain showed a sensitivity and specificity of 0.81 and 0.82, respectively.ConclusionCNN-based deep learning algorithm using hemodynamic factors, including WSS and strain, could be an effective tool for predicting rupture risk in cerebral aneurysms with good predictive accuracy.

show abstract

Prediction of Intracranial Aneurysm Risk using Machine Learning

Cited by 33 publications

References 36 publications

Role of Artificial Intelligence in Unruptured Intracranial Aneurysm: An Overview

Role of Artificial Intelligence in Unruptured Intracranial Aneurysm: An Overview

Precision medicine and machine learning towards the prediction of the outcome of potential celiac disease

Rupture risk prediction of cerebral aneurysms using a novel convolutional neural network-based deep learning model

Contact Info

Product

Resources

About