Protocol and Preliminary Results of the Establishment of Intracranial Aneurysm Database for Artificial Intelligence Application Based on CTA Images

You, Wei; Sun, Yong Han; Feng, Junqiang; Wang, Zhiliang; Li, Lin; Chen, Xiheng; Lv, Jian; Tang, Yudi; Deng, Dingwei; Wei, Dachao; Gui, Siming; Liu, Xinke; Liu, Peng; Jin, Hengwei; Ge, Huijian; Zhang, Yanling

doi:10.3389/fneur.2022.932933

Cited by 8 publications

(10 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among the 15 studies analyzed, 6 studies (40%) demonstrated a seemingly favorable trade-off between lesion-level sensitivity (greater than 0.8) and the number of false positives per image (less than 2) [9,12,19,21,22,26]. According to the FROC in Figure 4, Wang et al [19] achieved the best trade-off between the lesion-level sensitivity and false positives per image, with respective values being 0.944 and 0.6.…”

Section: Current State Of the Art Performancementioning

confidence: 99%

“…Initially, 608 studies that met the search criteria were identified, and from these, 163 full-text articles were assessed for potential eligibility. Eventually, 15 studies published between June 2019 and April 2023 were included in the analysis [8,9,[19][20][21][22][23][24][25][26][27][28][29][30][31]. The collective training and testing data encompassed 11,210 subjects, which featured a total of 13,086 IAs.…”

Section: Characteristics Of Included Studiesmentioning

confidence: 99%

“…Of these, 9012 cases were utilized for training, while 4074 cases were allocated for validation purposes, both internally (3039 cases) and externally (765 cases). For comprehensive insights, Tables 1 and 2 present detailed characteristics of the studies, divided into two groups: (i) automatic methods (13/15, 87%) [8,9,[19][20][21][22][23][24][25][26][27][28]31], and (ii) semi-automatic methods (2/15, 13%) [29,30]. All studies were retrospective in nature.…”

Section: Characteristics Of Included Studiesmentioning

confidence: 99%

“…Please note that these values were pooled across IAs of all sizes included in each study's test dataset and may be over-optimistic regarding the detection of small IAs. [8,9,[19][20][21][22][23][24][25][26][27][28][29][30]. The 95% confidence intervals (CIs) are given in square brackets.…”

Section: Sensitivity Analysismentioning

confidence: 99%

“…Eight (8/15, 53%) studies reported sufficient data to reliably determine lesion-level sensitivity based on the size of IA [8,[19][20][21]23,24,27,29]. It is important to note that while all authors defined small IAs as those with a diameter less than 3 mm, the definitions for medium and large IAs varied across the studies.…”

Section: Identificationmentioning

confidence: 99%

See 4 more Smart Citations

A Systematic Review of Deep-Learning Methods for Intracranial Aneurysm Detection in CT Angiography

Bizjak,

Špiclin

2023

Biomedicines

View full text Add to dashboard Cite

Background: Subarachnoid hemorrhage resulting from cerebral aneurysm rupture is a significant cause of morbidity and mortality. Early identification of aneurysms on Computed Tomography Angiography (CTA), a frequently used modality for this purpose, is crucial, and artificial intelligence (AI)-based algorithms can improve the detection rate and minimize the intra- and inter-rater variability. Thus, a systematic review and meta-analysis were conducted to assess the diagnostic accuracy of deep-learning-based AI algorithms in detecting cerebral aneurysms using CTA. Methods: PubMed (MEDLINE), Embase, and the Cochrane Library were searched from January 2015 to July 2023. Eligibility criteria involved studies using fully automated and semi-automatic deep-learning algorithms for detecting cerebral aneurysms on the CTA modality. Eligible studies were assessed using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines and the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool. A diagnostic accuracy meta-analysis was conducted to estimate pooled lesion-level sensitivity, size-dependent lesion-level sensitivity, patient-level specificity, and the number of false positives per image. An enhanced FROC curve was utilized to facilitate comparisons between the studies. Results: Fifteen eligible studies were assessed. The findings indicated that the methods exhibited high pooled sensitivity (0.87, 95% confidence interval: 0.835 to 0.91) in detecting intracranial aneurysms at the lesion level. Patient-level sensitivity was not reported due to the lack of a unified patient-level sensitivity definition. Only five studies involved a control group (healthy subjects), whereas two provided information on detection specificity. Moreover, the analysis of size-dependent sensitivity reported in eight studies revealed that the average sensitivity for small aneurysms (<3 mm) was rather low (0.56). Conclusions: The studies included in the analysis exhibited a high level of accuracy in detecting intracranial aneurysms larger than 3 mm in size. Nonetheless, there is a notable gap that necessitates increased attention and research focus on the detection of smaller aneurysms, the use of a common test dataset, and an evaluation of a consistent set of performance metrics.

show abstract