Predicting Hematoma Expansion Research Articles

Potential of Satellite Sign for Prediction of Hematoma Expansion in Small Spontaneous Hematoma within 7 Days' Follow-Up.

Background Hematoma expansion (HE) is the most important modifiable predictor that can change the clinical outcome of intracerebral hemorrhage (ICH) patients. The study aimed to investigate the potential of satellite sign for prediction of HE in spontaneous ICH patients who had follow-up non-contrast computed tomography (NCCT) within 7 days after the initial CT scan. Methods We retrospectively reviewed data and NCCT from 142 ICH patients who were treated at our hospital at Bangkok, Thailand. All included patients were treated conservatively, had baseline NCCT within 12 hours after symptom onset, and had follow-up NCCT within 168 hours after baseline NCCT. HE was initially estimated by two radiologists, and then by image analysis software. Association between satellite sign and HE was evaluated. Results HE occurred in 45 patients (31.7%). Patients with HE had significantly higher activated partial thromboplastin time ( p = 0.001) and baseline hematoma volume ( p = 0.001). The prevalence of satellite sign was 43.7%, and it was significantly independently associated with HE ( p = 0.021). The sensitivity, specificity, and accuracy of satellite sign for predicting HE was 57.8, 62.9, and 61.3%, respectively. From image analysis software, the cutoff of greater than 9% relative growth in hematoma volume on follow-up NCCT had the highest association with satellite sign ( p = 0.024), with a sensitivity of 55%, specificity of 64.6%, and accuracy of 60.5%. Conclusion Satellite sign, a new NCCT predictor, was found to be significantly associated with HE in Thai population. With different context of Thai population, HE was found in smaller baseline hematoma volume. Satellite sign was found more common in lobar hematoma. Further studies to validate satellite sign for predicting HE and to identify an optimal cutoff in Thai population that is correlated with clinical outcomes are warranted.

Abstract TP120: Noncontrast Ct, Single-phase Cta, And Multiphase Cta Intracerebral Hemorrhage Expansion Scores

Background: Several noncontrast computed tomography (NCCT), single-phase computed tomography angiography (CTA), and multiphase CTA markers of intracerebral hemorrhage (ICH) expansion have been previously proposed. We derived and validated three scores for the prediction of hematoma expansion depending on the use of NCCT, single-phase CTA, or multiphase CTA markers of hematoma expansion. Methods: We prospective studied 276 consecutive patients with ICH within 6 hours from symptom onset. After deriving NCCT, single-phase CTA, and multiphase CTA scores in a 5-year period population (n=156), we validated them in a different 3-year period population (n=120). Outcome parameters included substantial hematoma expansion >6 mL or >33% at 24 hours (primary outcome) and poor outcome (mRS score >2) at 90 days. Results: The most accurate marker of hematoma expansion was spot sign in phase 1 of multiphase CTA (80.3%). The four independent predictors of substantial hematoma expansion included in the different scores were ultraearly hematoma growth (uHG) >5 mL/h, heterogeneous density, spot sign in phase 1 of multiphase CTA, and spot sign in any phase of multiphase CTA (Table). On each of the three scores, the proportion of patients that experienced substantial hematoma expansion increased with each point increase. C-index for both substantial hematoma expansion and poor outcome in the derivation and validation cohort was lower in NCCT expansion score than in single-phase CTA expansion score which, in turn, was lower than in multiphase CTA expansion score (Table). Conclusions: This study demonstrates the added prognostic value of more advanced CT modalities in acute ICH evaluation. Single-phase CTA score and, especially, multiphase CTA score, are more robust than NCCT score in the prediction of hematoma expansion and poor outcome. These scores may help to refine the selection of patients at risk of expansion and poorest outcomes in different decision-making scenarios.

3D slicer-based calculation of hematoma irregularity index for predicting hematoma expansion in intracerebral hemorrhage

BackgroundIrregular hematoma is considered as a risk sign of hematoma expansion. The aim of this study was to quantify hematoma irregularity with computed tomography based on 3D Slicer.MethodsPatients with spontaneous intracerebral hemorrhage who underwent an initial and subsequent non-contrast computed tomography (CT) at a single medical center between January 2019 to January 2020 were retrospectively identified. The Digital Imaging and Communication in Medicine (DICOM) standard images were loaded into the 3D Slicer, and the surface area (S) and volume (V) of hematoma were calculated. The hematoma irregularity index (HII) was defined as frac{sqrt{S/pi}}2/sqrt[3]{3V/4pi}times100. Logistic regression analyses and receiver operating characteristic (ROC) curve analysis were performed to assess predictive performance of HII.ResultsThe enrolled patients were divided into those with hematoma enlargement (n = 36) and those without the enlargement (n = 57). HII in hematoma expansion group was 130.4 (125.1–140.0), and the index in non-enlarged hematoma group was 118.6 (113.5-122.3). There was significant difference in HII between the two groups (P < 0.01). Multivariate logistic regression analysis revealed that the HII was significantly associated with hematoma expansion before (odds ratio = 1.203, 95% confidence interval [CI], 1.115–1.298; P < 0.001) and after adjustment for age, hematoma volume, Glasgow Coma Scale score (odds ratio = 1.196, 95% CI, 1.102–1.298, P < 0.001). The area under the ROC curve was 0.86 (CI, 0.78–0.93, P < 0.01), and the best cutoff of HII for predicting hematoma growth was 123.8.ConclusionAs a quantitative indicator of irregular hematoma, HII can be calculated using the 3D Slicer. And the HII was independently correlated with hematoma expansion.

The Added Value of Intraventricular Hemorrhage on the Radiomics Analysis for the Prediction of Hematoma Expansion of Spontaneous Intracerebral Hemorrhage.

Among patients undergoing head computed tomography (CT) scans within 3 h of spontaneous intracerebral hemorrhage (sICH), 28% to 38% have hematoma expansion (HE) on follow-up CT. This study aimed to predict HE using radiomics analysis and investigate the impact of intraventricular hemorrhage (IVH) compared with the conventional approach based on intraparenchymal hemorrhage (IPH) alone. This retrospective study enrolled 127 patients with baseline and follow-up non-contrast CT (NCCT) within 4~72 h of sICH. IPH and IVH were outlined separately for performing radiomics analysis. HE was defined as an absolute hematoma growth &gt; 6 mL or percentage growth &gt; 33% of either IPH (HEP) or a combination of IPH and IVH (HEP+V) at follow-up. Radiomic features were extracted using PyRadiomics, and then the support vector machine (SVM) was used to build the classification model. For each case, a radiomics score was generated to indicate the probability of HE. There were 57 (44.9%) HEP and 70 (55.1%) non-HEP based on IPH alone, and 58 (45.7%) HEP+V and 69 (54.3%) non-HEP+V based on IPH + IVH. The majority (&gt;94%) of HE patients had poor early outcomes (death or modified Rankin Scale &gt; 3 at discharge). The radiomics model built using baseline IPH to predict HEP (RMP) showed 76.4% accuracy and 0.73 area under the ROC curve (AUC). The other model using IPH + IVH to predict HEP+V (RMP+V) had higher accuracy (81.9%) with AUC = 0.80, and this model could predict poor outcomes. The sensitivity/specificity of RMP and RMP+V for HE prediction were 71.9%/80.0% and 79.3%/84.1%, respectively. The proposed radiomics approach with additional IVH information can improve the accuracy in prediction of HE, which is associated with poor clinical outcomes. A reliable radiomics model may provide a robust tool to help manage ICH patients and to enroll high-risk ICH cases into anti-expansion or neuroprotection drug trials.

A non-contrast computed tomography-based radiomics nomogram for the prediction of hematoma expansion in patients with deep ganglionic intracerebral hemorrhage.

Background and purposeHematoma expansion (HE) is a critical event following acute intracerebral hemorrhage (ICH). We aimed to construct a non-contrast computed tomography (NCCT) model combining clinical characteristics, radiological signs, and radiomics features to predict HE in patients with spontaneous ICH and to develop a nomogram to assess the risk of early HE.Materials and methodsWe retrospectively reviewed 388 patients with ICH who underwent initial NCCT within 6 h after onset and follow-up CT within 24 h after initial NCCT, between January 2015 and December 2021. Using the LASSO algorithm or stepwise logistic regression analysis, five models (clinical model, radiological model, clinical-radiological model, radiomics model, and combined model) were developed to predict HE in the training cohort (n = 235) and independently verified in the test cohort (n = 153). The Akaike information criterion (AIC) and the likelihood ratio test (LRT) were used for comparing the goodness of fit of the five models, and the AUC was used to evaluate their ability in discriminating HE. A nomogram was developed based on the model with the best performance.ResultsThe combined model (AIC = 202.599, χ2 = 80.6) was the best fitting model with the lowest AIC and the highest LRT chi-square value compared to the clinical model (AIC = 232.263, χ2 = 46.940), radiological model (AIC = 227.932, χ2 = 51.270), clinical-radiological model (AIC = 212.711, χ2 = 55.490) or radiomics model (AIC = 217.647, χ2 = 57.550). In both cohorts, the nomogram derived from the combined model showed satisfactory discrimination and calibration for predicting HE (AUC = 0.900, sensitivity = 83.87%; AUC = 0.850, sensitivity = 80.10%, respectively).ConclusionThe NCCT-based model combining clinical characteristics, radiological signs, and radiomics features could efficiently discriminate early HE, and the nomogram derived from the combined model, as a non-invasive tool, exhibited satisfactory performance in stratifying HE risks.

Intracerebral hemorrhage markers on non-contrast computed tomography as predictors of the dynamic spot sign on CT perfusion and associations with hematoma expansion and outcome.

To assess the association between non-contrast computed tomography (NCCT) hematoma markers and the dynamic spot sign on computed tomography perfusion (CTP), and their associations with hematoma expansion, clinical outcome, and in-hospital mortality. Patients who presented with intracerebral hemorrhage (ICH) to a stroke center over an 18-month period and underwent baseline NCCT and CTP, and a follow-up NCCT within 24h after the baseline scan were included. The initial and follow-up hematoma volumes were calculated. Two raters independently assessed the baseline NCCT for hematoma markers and concurrently assessed the CTP for the dynamic spot sign. Univariate and multivariate logistic regression analyses were performed to assess the association between the hematoma markers and the dynamic spot sign, adjusting for known ICH expansion predictors. Eighty-five patients were included in our study and 55 patients were suitable for expansion analysis. Heterogeneous density was the only NCCT hematoma marker to be associated with the dynamic spot sign after multivariate analysis (odds ratio, 58.61; 95% confidence interval, 9.13-376.05; P < 0.001). The dynamic spot sign was present in 22 patients (26%) and significantly predicted hematoma expansion (odds ratio, 36.6; 95% confidence interval, 2.51-534.2; P = 0.008). All patients with a spot sign had a swirl sign. A co-located hypodensity and spot sign was significantly associated with in-hospital mortality (odds ratio, 6.17; 95% confidence interval, 1.09-34.78; P = 0.039). Heterogeneous density and swirl sign are associated with the dynamic spot sign. The dynamic spot sign is a stronger predictor than NCCT hematoma markers of significant hematoma expansion. A co-located spot sign and hypodensity predicts in-hospital mortality.

Added value of non-contrast CT and CT perfusion markers for prediction of intracerebral hemorrhage expansion and outcome.

To test the hypothesis that the combined analysis of non-contrast CT (NCCT) and CT perfusion (CTP) imaging markers improves prediction of hematoma expansion (HE) and outcome in intracerebral hemorrhage (ICH). Retrospective, single-center analysis of patients with primary ICH undergoing NCCT and CTP within 6 h from onset. NCCT images were assessed for the presence of intrahematomal hypodensity and shape irregularity. Perihematomal cerebral blood volume and spot sign were assessed on CTP. The main outcomes of the analysis were HE (growth > 6 mL and/or > 33%) and poor functional prognosis (90 days modified Rankin Scale 3-6). Predictors of HE and outcome were explored with logistic regression. A total of 150 subjects were included (median age 68, 47.1% males) of whom 54 (36%) had HE and 52 (34.7%) had poor outcome. The number of imaging markers on baseline imaging was independently associated with HE (odds ratio 2.66, 95% confidence interval 1.70-4.17, p < 0.001) and outcome (odds ratio 1.64, 95% CI 1.06-2.56, p = 0.027). Patients with the simultaneous presence of all the four markers had the highest risk of HE and unfavorable prognosis (mean predicted probability of 91% and 79% respectively). The combined-markers analysis outperformed the sensitivity of the single markers analyzed separately. In particular, the presence of at least one marker identified patients with HE and poor outcome with 91% and 87% sensitivity respectively. NCCT and CTP markers provide additional yield in the prediction of HE and ICH outcome. • Perihematomal hypoperfusion is associated with hematoma expansion and poor outcome in acute intracerebral hemorrhage. • Non-contrast CT and CT perfusion markers improve prediction of hematoma expansion and unfavorable prognosis. • A multimodal CT protocol including CT perfusion will help the identification of patients at high risk of clinical deterioration and poor outcome.

Machine learning prediction of hematoma expansion in acute intracerebral hemorrhage

To examine whether machine learning (ML) approach can be used to predict hematoma expansion in acute intracerebral hemorrhage (ICH) with accuracy and widespread applicability, we applied ML algorithms to multicenter clinical data and CT findings on admission. Patients with acute ICH from three hospitals (n = 351) and those from another hospital (n = 71) were retrospectively assigned to the development and validation cohorts, respectively. To develop ML predictive models, the k-nearest neighbors (k-NN) algorithm, logistic regression, support vector machines (SVMs), random forests, and XGBoost were applied to the patient data in the development cohort. The models were evaluated for their performance on the patient data in the validation cohort, which was compared with previous scoring methods, the BAT, BRAIN, and 9-point scores. The k-NN algorithm achieved the highest area under the receiver operating characteristic curve (AUC) of 0.790 among all ML models, and the sensitivity, specificity, and accuracy were 0.846, 0.733, and 0.775, respectively. The BRAIN score achieved the highest AUC of 0.676 among all previous scoring methods, which was lower than the k-NN algorithm (p = 0.016). We developed and validated ML predictive models of hematoma expansion in acute ICH. The models demonstrated good predictive ability, showing better performance than the previous scoring methods.

Prediction of Intraparenchymal Hemorrhage Progression and Neurologic Outcome in Traumatic Brain Injury Patients Using Radiomics Score and Clinical Parameters.

(1) Background: Radiomics analysis of spontaneous intracerebral hemorrhages on computed tomography (CT) images has been proven effective in predicting hematoma expansion and poor neurologic outcome. In contrast, there is limited evidence on its predictive abilities for traumatic intraparenchymal hemorrhage (IPH). (2) Methods: A retrospective analysis of 107 traumatic IPH patients was conducted. Among them, 45 patients (42.1%) showed hemorrhagic progression of contusion (HPC) and 51 patients (47.7%) had poor neurological outcome. The IPH on the initial CT was manually segmented for radiomics analysis. After feature extraction, selection and repeatability evaluation, several machine learning algorithms were used to derive radiomics scores (R-scores) for the prediction of HPC and poor neurologic outcome. (3) Results: The AUCs for R-scores alone to predict HPC and poor neurologic outcome were 0.76 and 0.81, respectively. Clinical parameters were used to build comparison models. For HPC prediction, variables including age, multiple IPH, subdural hemorrhage, Injury Severity Score (ISS), international normalized ratio (INR) and IPH volume taken together yielded an AUC of 0.74, which was significantly (p = 0.022) increased to 0.83 after incorporation of the R-score in a combined model. For poor neurologic outcome prediction, clinical variables of age, Glasgow Coma Scale, ISS, INR and IPH volume showed high predictability with an AUC of 0.92, and further incorporation of the R-score did not improve the AUC. (4) Conclusion: The results suggest that radiomics analysis of IPH lesions on initial CT images has the potential to predict HPC and poor neurologic outcome in traumatic IPH patients. The clinical and R-score combined model further improves the performance of HPC prediction.

Deep Learning-Based Prediction of Hematoma Expansion Using a Single Brain Computed Tomographic Slice in Patients With Spontaneous Intracerebral Hemorrhages

We aimed to predict hematoma expansion in intracerebral hemorrhage (ICH) patients by using the deep learning technique. We retrospectively collected data from ICH patients treated between May 2015 and May 2019. Head computed tomography (CT) scans were performed at admission, and 6 hours, 24 hours, and 72 hours after admission. CT scans were mandatory when neurologic deficits occurred. Univariate and multivariate analyses were conducted to illustrate the association between clinical variables and hematoma expansion. Convolutional neural network (CNN) was adopted to predict hematoma expansion based on brain CT slices. In addition, 5 machine learning methods, including support vector machine, multi-layer perceptron, naive Bayes, decision tree, and random forest, were also performed to predict hematoma expansion based on clinical variables for comparisons. A total of 223 patients were included. It was revealed that patients' older age (odds ratio [95% confidence interval]: 1.783 [1.417-1.924]), cerebral hemorrhage and breaking into the ventricle (2.524 [1.291-1.778]), coagulopathy (2.341 [1.677-3.454]), and baseline National Institutes of Health Stroke Scale (1.545 [1.132-3.203]) and Glasgow Coma Scale scores (0.782 [0.432-0.918]) independently associated with hematoma expanding. After 4-5 epochs, the CNN framework was well trained. The average sensitivity, specificity, and accuracy of CNN prediction are 0.9197, 0.8837, and 0.9058, respectively. Compared with 5 machine learning methods based on clinical variables, CNN can also achieve better performance. More than 90% of hematomas with or without expansion can be precisely classified by deep learning technology within this study, which is better than other methods based on clinical variables only. Deep learning technology could favorably predict hematoma expansion from non-contrast CT scan images.

Automatic and Efficient Prediction of Hematoma Expansion in Patients with Hypertensive Intracerebral Hemorrhage Using Deep Learning Based on CT Images.

Patients with hypertensive intracerebral hemorrhage (ICH) have a high hematoma expansion (HE) incidence. Noninvasive prediction HE helps doctors take effective measures to prevent accidents. This study retrospectively analyzed 253 cases of hypertensive intraparenchymal hematoma. Baseline non-contrast-enhanced CT scans (NECTs) were collected at admission and compared with subsequent CTs to determine the presence of HE. An end-to-end deep learning method based on CT was proposed to automatically segment the hematoma region, region of interest (ROI) feature extraction, and HE prediction. A variety of algorithms were employed for comparison. U-Net with attention performs best in the task of segmenting hematomas, with the mean Intersection overUnion (mIoU) of 0.9025. ResNet-34 achieves the most robust generalization capability in HE prediction, with an area under the receiver operating characteristic curve (AUC) of 0.9267, an accuracy of 0.8827, and an F1 score of 0.8644. The proposed method is superior to other mainstream models, which will facilitate accurate, efficient, and automated HE prediction.

Blood Pressure and Spot Sign in Spontaneous Supratentorial Subcortical Intracerebral Hemorrhage

BackgroundSpontaneous intracerebral hemorrhage is a potentially devastating cause of brain injury, often occurring secondary to hypertension. Contrast extravasation on computed tomography angiography (CTA), known as the spot sign, has been shown to predict hematoma expansion and worse outcomes. Although hypertension has been associated with an increased rate of the spot sign being present, the relationship between spot sign and blood pressure has not been fully explored.MethodsWe retrospectively analyzed data from 134 patients (40 women and 94 men, mean age 62.3 ± 15.73 years) presenting to a tertiary academic medical center with spontaneous supratentorial subcortical intracerebral hemorrhage from 1/1/2018 to 1/4/2021.ResultsA spot sign was demonstrated in images of 18 patients (13.43%) and correlated with a higher intracerebral hemorrhage score (2.61 ± 1.42 vs. 1.31 ± 1.25, p = 0.002), larger hematoma volume (53.49cm3 ± 32.08 vs. 23.45cm3 ± 25.65, p = 0.001), lower Glasgow Coma Scale on arrival (9.06 ± 4.56 vs. 11.74 ± 3.65, p = 0.027), increased risk of hematoma expansion (16.67% vs. 5.26%, p = 0.042), and need for surgical intervention (66.67% vs. 15.52%, p < 0.001). We did not see a correlation with age, sex, or underlying comorbidities. The presence of spot sign correlated with higher modified Rankin scores at discharge (4.94 ± 1.00 vs. 3.92 ± 1.64, p < 0.001). We saw significantly higher systolic blood pressure at the time of CTA in patients with a spot sign (184 mm Hg ± 43.11 vs. 153 mm Hg ± 36.99, p = 0.009) and the highest recorded blood pressure (p = 0.019), although not blood pressure on arrival (p = 0.081). Performing CTA early in the process of blood pressure lowering was associated with a spot sign (p < 0.001).ConclusionsThe presence of spot sign correlates with larger hematomas, worse outcomes, and increased surgical intervention. There is a significant association between spot sign and systolic blood pressure at the time of CTA, with the highest systolic blood pressure being recorded prior to CTA. Although the role of intensive blood pressure management in spontaneous intracerebral hemorrhage remains a subject of debate, patients with a spot sign may be a subgroup that could benefit from this.

The Attenuation Value Within the Non-hypodense Region on Non-contrast Computed Tomography of Spontaneous Cerebral Hemorrhage: A Long-Neglected Predictor of Hematoma Expansion.

Background and PurposeThe ability of attenuation value of the non-hypodense region of hematoma in non-contrast computed tomography (NCCT) for predicting hematoma expansion (HE) remains unclear. Our purpose is to explore this relationship.MethodsTwo cohorts of patients were collected for analysis. The region where we measured hematoma attenuation values was limited to the non-hypodense region that was not adjacent to the normal brain tissue on NCCT. The critical attenuation value was derived via receiver operating characteristic (ROC) curve analysis in the derivation cohort and its predictive ability was validated in the validation cohort. Independent relationships between predictors, such as critical attenuation value of the non-hypodense region and HE were analyzed using the least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic analysis.ResultsThe results showed that the attenuation value <64 Hounsfield units (HU) was independently associated with HE [odds ratio (OR), 4.118; 95% confidential interval (CI), 1.897–9.129, p < 0.001] and the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and area under the curve (AUC) for predicting HE were 36.11%, 81.71%, 1.97, 0.78, 44.8%, 75.7%, and 0.589, respectively.ConclusionsOur research explored and validated the relationship between the attenuation value of the non-hypodense region of hematoma and HE. The attenuation value < 64 HU was an appropriate indicator of early HE.

Predicting Hematoma Expansion after Spontaneous Intracranial Hemorrhage Through a Radiomics Based Model.

Intracranial hemorrhage (ICH) is characterized as bleeding into the brain tissue, intracranial space, and ventricles and is the second most disabling form of stroke. Hematoma expansion (HE) following ICH has been correlated with significant neurological decline and death. For early detection of patients at risk, deep learning prediction models were developed to predict whether hematoma due to ICH will expand. This study aimed to explore the feasibility of HE prediction using a radiomic approach to help clinicians better stratify HE patients and tailor intensive therapies timely and effectively. Two hundred ICH patients with known hematoma evolution, were enrolled in this study. An open-source python package was utilized for the extraction of radiomic features from both non-contrast computed tomography (NCCT) and magnetic resonance imaging (MRI) scans through characterization algorithms. A total of 99 radiomic features were extracted and different features were selected for network inputs for the NCCT and MR models. Seven supervised classifiers: Support Vector Machine, Naïve Bayes, Decision Tree, Random Forest, Logistic Regression, K-Nearest Neighbor and Multilayer Perceptron were used to build the models. A training:testing split of 80:20 and 20 iterations of Monte Carlo cross validation were performed to prevent overfitting and assess the variability of the networks, respectively. The models were fed training datasets from which they learned to classify the data based on pre-determined radiomic categories. The highest sensitivity among the NCCT classifier models was seen with the support vector machine (SVM) and logistic regression (LR) of 72 ± 0.3% and 73 ± 0.5%, respectively. The MRI classifier models had the highest sensitivity of 68 ± 0.5% and 72 ± 0.5% for the SVM and LR models, respectively. This study indicates that the NCCT radiomics model is a better predictor of HE and that SVM and LR classifiers are better predictors of HE due to their more cautious approach indicated by a higher sensitivity metric.

Initial investigation of predicting hematoma expansion for intracerebral hemorrhage using imaging biomarkers and machine learning.

Intracerebral Hemorrhage (ICH) is one of the most devastating types of strokes with mortality and morbidity rates ranging from about 51%-65% one year after diagnosis. Early hematoma expansion (HE) is a known cause of worsening neurological status of ICH patients. The goal of this study was to investigate whether non-contrast computed tomography imaging biomarkers (NCCT-IB) acquired at initial presentation can predict ICH growth in the acute stage. We retrospectively collected NCCT data from 200 patients with acute (<6 hours) ICH. Four NCCT-IBs (blending region, dark hole, island, and edema) were identified for each hematoma, respectively. HE status was recorded based on the clinical observation reported in the patient chart. Supervised machine learning models were developed, trained, and tested for 15 different input combinations of the NCCT-IBs to predict HE. Model performance was assessed using area under the receiver operating characteristic curve and probability for accurate diagnosis (PAD) was calculated. A 20-fold Monte-Carlo cross validation was implemented to ensure model reliability on a limited sample size of data, by running a myriad of random training/testing splits. The developed algorithm was able to predict expansion utilizing all four inputs with an accuracy of 70.17%. Further testing of all biomarker combinations yielded P AD ranging from 0.57, to 0.70. Specific attributes of ICHs may influence the likelihood of HE and can be evaluated via a machine learning algorithm. However, certain parameters may differ in importance to reach accurate conclusions about potential expansion.

Predicting hematoma expansion in intracerebral hemorrhage from brain CT scans via K-nearest neighbors matting and deep residual network

Hematoma Expansion (HE) in spontaneous IntraCerebral Hemorrhage (ICH) is one of the highest mortality rates in neurosurgery. It is also influenced by various factors such as hematoma volume, location, shape, and so on. Although it’s crucial to present a method to predict HE, there is no work on the prediction model using brain imaging for this task to our knowledge. Inspired by the above, this work proposes a method including preprocessing and classification to predict HE in ICH from brain computed tomography (CT) scans. A k-nearest neighbors matting method is adopted in the preprocessing of brain CT scans to remove the outer part of the skull and retain brain tissue features. A deep residual network is then presented to give classification results, which helps to learn from features of hematoma and other parts in brain tissue. Experimental results on 223 patients including 137 HE patients show that the proposed framework can achieve 0.890±0.020 accuracy, 0.880±0.033 specificity, 0.925±0.041 sensitivity, and 0.867±0.025 F1-Score. Compared with other state-of-the-art works that need a large number of clinic data, the proposed method can obtain better prediction performance using brain CT scans only.

Baseline neutrophil\u2013lymphocyte ratio can be associated with hematoma expansion in patients with intracerebral hemorrhage: a retrospective observational study

BackgroundHematoma expansion can be related to increased mortality and poor clinical outcomes in patients with intracerebral hemorrhage (ICH). So, early identification and prevention of hematoma expansion can be considered as an important therapeutic aim. This study aimed to evaluate the hypothesis that the neutrophil to lymphocyte ratio (NLR) is associated with hematoma expansion in ICH patients.MethodsWe retrospectively evaluated the clinical data of a total of 221 patients with ICH who were treated in our department between April 2018 and April 2021. The demographic, clinical, radiological, and laboratory test data including the NLR upon admission were investigated. A binary logistic regression analysis was used to assess the independent associations between different variables and hematoma expansion.ResultsA total of 221 patients with ICH were included. There were 122 (55.2%) males and 99 (44.8%) females. The mean age (years) at admission was 66.43 ± 8.28.The hematoma expansion occurred in 57 (25.8%) cases. The results of the multivariate analysis showed that hematoma volume at baseline (OR, 3.12; 95% CI 1.78–5.02; P < 0.001), admission systolic blood pressure (OR, 2.87; 95% CI 1.79–4.34; P = 0.013), Glasgow Coma Scale (GCS) (OR, 1.94; 95% CI 1.45–2.93; P = 0.020), and NLR (OR, 1.74; 95% CI 1.16–2.60; P = 0.032) were correlated with hematoma expansion in these patients.ConclusionsOur findings suggest that NLR can be a predictor of hematoma expansion in patients with ICH. This cost-effective and easily available biomarker could be used to early prediction of hematoma expansion in these patients.

External validation study on the value of deep learning algorithm for the prediction of hematoma expansion from noncontrast CT scans

BackgroundHematoma expansion is an independent predictor of patient outcome and mortality. The early diagnosis of hematoma expansion is crucial for selecting clinical treatment options. This study aims to explore the value of a deep learning algorithm for the prediction of hematoma expansion from non-contrast computed tomography (NCCT) scan through external validation.Methods102 NCCT images of hypertensive intracerebral hemorrhage (HICH) patients diagnosed in our hospital were retrospectively reviewed. The initial computed tomography (CT) scan images were evaluated by a commercial Artificial Intelligence (AI) software using deep learning algorithm and radiologists respectively to predict hematoma expansion and the corresponding sensitivity, specificity and accuracy of the two groups were calculated and compared. Comparisons were also conducted among gold standard hematoma expansion diagnosis time, AI software diagnosis time and doctors’ reading time.ResultsAmong 102 HICH patients, the sensitivity, specificity, and accuracy of hematoma expansion prediction in the AI group were higher than those in the doctor group(80.0% vs 66.7%, 73.6% vs 58.3%, 75.5% vs 60.8%), with statistically significant difference (p < 0.05). The AI diagnosis time (2.8 ± 0.3 s) and the doctors’ diagnosis time (11.7 ± 0.3 s) were both significantly shorter than the gold standard diagnosis time (14.5 ± 8.8 h) (p < 0.05), AI diagnosis time was significantly shorter than that of doctors (p < 0.05).ConclusionsDeep learning algorithm could effectively predict hematoma expansion at an early stage from the initial CT scan images of HICH patients after onset with high sensitivity and specificity and greatly shortened diagnosis time, which provides a new, accurate, easy-to-use and fast method for the early prediction of hematoma expansion.

Increased Prognostic Yield by Combined Assessment of Non-Contrast Computed Tomography Markers of Antithrombotic-Related Spontaneous Intracerebral Hemorrhage Expansion.

Background and aims: The utility of proposed non-contrast computed tomography (NCCT) markers for the prediction of hematoma expansion in patients with antithrombotic-related spontaneous intracerebral hemorrhage (ICH) is limited. Additionally, there is significant overlap between different suggested ICH shape and density markers. Methods: We assessed the prognostic yield for hematoma expansion of a combined score incorporating features of ICH shape irregularity (satellite sign and/or Barras score ≥ 3), heterogeneous ICH density (swirl sign and/or Barras score ≥ 3) on baseline NCCT and timing from ICH onset to NCCT. Results: We evaluated data from 79 patients with antithrombotic-related spontaneous ICH (32% with hematoma expansion). Swirl (84% vs. 39%) and satellite signs (20% vs. 7%) on baseline NCCT were significantly more prevalent (p < 0.001) in patients with hematoma expansion. Patients with hematoma expansion had more irregular and heterogeneous bleeds on baseline NCCT scans, as quantified by higher (p < 0.001) Barras shape (4 (4–5) vs. 3 (2–4)) and density scores (4 (3–5) vs. 2 (1–3)), respectively. The overall diagnostic yield of the combined score (area under the curve: 0.86, 95%CI: 0.78–0.94) significantly outperformed (p < 0.001) the diagnostic yield of each individual marker. Scores of 4 or 5 in the combined score were associated with a sensitivity of 60.0%, specificity of 90.7%, overall diagnostic accuracy of 81.0%, positive likelihood ratio (LR) of 6.48, negative LR of 0.44, positive predictive value (PV) of 0.76 and negative PV of 0.83. Conclusion: Combined NCCT marker assessment seems to increase the prognostic accuracy for hematoma expansion in antithrombotic-related spontaneous ICH patients.

417 Predicting Hematoma Expansion after Spontaneous Intracerebral Hemorrhage Through a Noncontrast Computed Tomography Based Model

INTRODUCTION: Spontaneous Intracerebral Hemorrhage (ICH) accounts for 15% of all acute strokes and is the deadliest stroke subtype with a 40% mortality rate after one month. More than 75% of all ICH patients are severely disabled or deceased after the first year. 30% of patients continue to bleed and demonstrate significant hematoma expansion (HE). Once HE occurs in ICH patients, treatment options are limited and the patient prognosis is significantly decreased. METHODS: NCCT scans for 200 ICH patients (70 expansion, 130 non-expansion) who presented with stroke-like symptoms between August 2016 and December 2019 were collected for this study. The patients had at least one of the following hemorrhages: intraparenchymal (n = 181), intraventricular (n = 45), subdural (n = 13), or subarachnoid (n = 19). Data augmentation and skull-stripping was conducted. A deep neural network was used to identify and segment the hemorrhagic region. Two networks were trained using two-dimensional images created by taking the standard deviation and mean of the voxel intensities over the entire volume to classify hematoma expansion using a training:testing split of 80:20 and 20 iterations of Monte Carlo cross validation. RESULTS: Hematoma expansion metrics using st.dev images are seen with a 95% confidence interval as: accuracy = 0.65 ± 0.04, sensitivity = 0.79 ± 0.06, specificity = 0.58 ± 0.06, precision = 0.51 ± 0.04, negative predictive value (NPV) = 0.84 ± 0.03. Metrics using mean images are: accuracy = 0.64 ± 0.03, sensitivity = 0.68 ± 0.08, specificity = 0.63 ± 0.07, precision = 0.49 ± 0.02, NPV = 0.78 ± 0.03. CONCLUSION: CNNs have the ability to predict hematoma expansion by utilizing two-dimensional images created from flattening NCCT volumes. Taking the st.dev over the NCCT volume within the hemorrhagic region is a better predictor of HE due to its more cautious approach indicated by a higher sensitivity metric.