Discrimination of Tumorous Intracerebral Hemorrhage from Benign Causes Using CT Densitometry

Choi, Yoon Seong; Rim, Tyler Hyungtaek; Ahn, Sung Soo; Lee, Seung Koo

doi:10.3174/ajnr.a4233

Cited by 14 publications

(14 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As ICH density decreases over time, this might have biased results. However, the difference in onset to imaging times was statistically not significant and in line with current literature (4).…”

Section: Discussionsupporting

confidence: 90%

“…Preliminary studies underline the CTbased diagnostic importance of this pathophysiological process, as recently published (23). In line with this, Choi et al (4) have described that a reduced hematoma attenuation in ICH can differentiate neoplastic from non-neoplastic lesions with high diagnostic accuracy. Accordingly, our analysis of the 100 most important features demonstrates that intensity distributionbased predictors (first-order histogram) contribute 51.9% of the cumulated feature importance (Figure 3A).…”

Section: Discussionmentioning

confidence: 70%

“…About 10% of intracerebral neoplastic lesions initially present as spontaneous hemorrhagic stroke (4). Acute non-contrastenhanced computed tomography (NECT) imaging is the preferred screening method when intracerebral hemorrhage (ICH) is suspected; however, follow-up imaging is required for final diagnosis (4,5). Interpretive challenges emerge from intra-hemorrhage and spatial heterogeneity as well as from the wide variety of different encompassing entities (6).…”

Section: Introductionmentioning

confidence: 99%

“…Interpretive challenges emerge from intra-hemorrhage and spatial heterogeneity as well as from the wide variety of different encompassing entities (6). Hence, initial radiological evaluation may be unreliable (4,7). A recent pooled analysis identified 18 reported cases of glioblastomainduced hemorrhage that were misdiagnosed as hypertensive ICHs, leading to significant diagnostic delays in two-thirds of the cases (7).…”

Section: Introductionmentioning

confidence: 99%

“…Radiomic analysis is built on the hypothesis that imaging data reflect the underlying morphology and dynamics of smallerscale biologic phenomena (8,9). In this context, two important imaging markers of ICH have been described: firstly, the presence and extent of perihematomal edema (PHE), and secondly, the dynamics of hemorrhage attenuation (4,10). However, radiomic analysis aims to capture also image information not assessable by human eyes, such as texture metrics or the evaluation of filtered images.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features

et al. 2020

View full text Add to dashboard Cite

Background: Early differentiation of neoplastic and non-neoplastic intracerebral hemorrhage (ICH) can be difficult in initial radiological evaluation, especially for extensive ICHs. The aim of this study was to evaluate the potential of a machine learning-based prediction of etiology for acute ICHs based on quantitative radiomic image features extracted from initial non-contrast-enhanced computed tomography (NECT) brain scans. Methods: The analysis included NECT brain scans from 77 patients with acute ICH (n = 50 non-neoplastic, n = 27 neoplastic). Radiomic features including shape, histogram, and texture markers were extracted from non-, wavelet-, and log-sigma-filtered images using regions of interest of ICH and perihematomal edema (PHE). Six thousand and ninety quantitative predictors were evaluated utilizing random forest algorithms with five-fold model-external cross-validation. Model stability was assessed through comparative analysis of 10 randomly drawn cross-validation sets. Classifier performance was compared with predictions of two radiologists employing the Matthews correlation coefficient (MCC). Results: The receiver operating characteristic (ROC) area under the curve (AUC) of the test sets for predicting neoplastic vs. non-neoplastic ICHs was 0.89 [95% CI (0.70; 0.99); P < 0.001], and specificities and sensitivities reached >80%. Compared to the radiologists' predictions, the machine learning algorithm yielded equal or superior results for all evaluated metrics. The MCC of the proposed algorithm at its optimal operating point (0.69) was significantly higher than the MCC of the radiologist readers (0.54); P = 0.01. Conclusion: Evaluating quantitative features of acute NECT images in a machine learning algorithm provided high discriminatory power in predicting non-neoplastic vs. neoplastic ICHs. Utilized in the clinical routine, the proposed approach could improve patient care at low risk and costs.

show abstract

Section: Discussionsupporting

confidence: 90%

Section: Discussionmentioning

confidence: 70%