Hematoma Expansion in Intracerebral Hemorrhage With Unclear Onset

P2Y12 inhibitor use predicts hematoma expansion in patients with intracerebral hemorrhage

OBJECTIVE

Hematoma expansion (HE) predicts disability and death after acute intracerebral hemorrhage (ICH). Aspirin and anticoagulants have been associated with HE. We tested the hypothesis that P2Y12 inhibitors predict subsequent HE in patients. We explored laboratory measures of P2Y12 inhibition and dual antiplatelet therapy with aspirin (DAPT).

METHODS

We prospectively identified patients with ICH. Platelet activity was measured with the VerifyNow-P2Y12 assay. Hematoma volumes for initial and follow-up CTs were calculated using a validated semi-automated technique. HE was defined as the difference between hematoma volumes on the initial and follow-up CT scans. Nonparametric statistics were performed with Kruskal-Wallis H, and correction for multiple comparisons performed with Dunn's test.

RESULTS

In 194 patients, 15 (7.7%) were known to take a P2Y12 inhibitor (clopidogrel in all but one). Patients taking a P2Y12 inhibitor had more HE compared to patients not taking a P2Y12 inhibitor (3.5 [1.2-11.9] vs. 0.1 [-0.8-1.4] mL, p = 0.004). Patients taking DAPT experienced the most HE (7.2 [2.6-13.8] vs. 0.0 [-1.0-1.1] mL, p = 0.04). The use of P2Y12 inhibitors was associated with less P2Y12 activity (178 [149-203] vs. 288 [246-319] P2Y12 reaction units, p = 0.005).

INTERPRETATION

Patients taking a P2Y12 inhibitor had more HE and less P2Y12 activity. The effect was most pronounced in patients on DAPT, suggesting a synergistic effect of P2Y12 inhibitors and aspirin with respect to HE. Acute reversal of P2Y12 inhibitors in acute ICH requires further study.

Absolute and relative iodine concentrations in the spot sign and haematoma for prediction of haematoma expansion in spontaneous intracerebral haemorrhage

AIM

To explore the predictive value of absolute and relative iodine concentrations in the spot sign (SS) and haematoma on gemstone spectral imaging (GSI) for haematoma expansion (HE).

MATERIALS AND METHODS

Patients with spontaneous intracerebral haemorrhage (ICH) who underwent computed tomography (CT) angiography using GSI were divided into an SS-positive group and an SS-negative group. In the SS-positive group, absolute and relative iodine concentrations in the SS (aICIS and rICIS, respectively) were measured. In the SS-negative group, absolute and relative iodine concentrations in haematoma (aICIH and rICIH, respectively) were measured. The area under the receiver operating characteristic curve (AUC-ROC) was used to investigate the HE predictive performance of aICIS, rICIS, and their combination in the SS-positive group, as well as the HE predictive performance of aICIH, rICIH, and their combination in the SS-negative group. The risk variables for HE in the two groups were investigated separately using logistic regression.

RESULTS

A total of 123 spontaneous ICH patients were enrolled. In the SS-positive group, the AUC of aICIS, rICIS, and their combination for predicting HE were 0.853, 0.893, and 0.922, respectively. rICIS was demonstrated to be a standalone predictor of HE via logistic regression. In the SS-negative group, aICIH, rICIH, and their combination had AUC-ROC values of 0.552, 0.783, and 0.851, respectively, to predict HE. According to multivariate analysis, rICIH was a reliable predictor of HE.

CONCLUSION

Absolute and relative iodine concentrations in the SS and haematoma can predict HE.

Intracranial and Blood Pressure Variability and In-Hospital Outcomes in Intracranial Device-Monitored Patients with Spontaneous Intracerebral Hemorrhage

BACKGROUND

Spontaneous intracerebral hemorrhage (sICH) is a major health concern and has high mortality rates up to 52%. Despite a decrease in its incidence, fatality rates remain unchanged; understanding and preventing of factors associated with mortality and treatments for these are needed. Blood pressure variability (BPV) has been shown to be a potential modifiable factor associated with clinical outcomes in patients with traumatic intracerebral hemorrhage and sICH. Few data are available on the effect of intracranial pressure (ICP) variability (ICPV) and outcomes in patients with sICH. The goal of our study was to investigate the association between ICPV and BPV during the first 24 h of intensive care unit (ICU) admission and external ventricular drain (EVD) placement, and mortality in patients with sICH who were monitored with an EVD.

METHODS

We conducted a single-center retrospective study of adult patients admitted to an ICU with a diagnosis of sICH who required EVD placement during hospitalization. We excluded patients with ICH secondary to other pathological conditions such as trauma, underlying malignancy, or arteriovenous malformation. Blood pressure and ICP measurements were collected and recorded hourly during the first 24 h of ICU admission and EVD placement, respectively. Measures of variability used were standard deviation (SD) and successive variation (SV). Primary outcome of interest was in-hospital mortality, and secondary outcomes were hematoma expansion and discharge home (a surrogate for good functional outcome at discharge). Descriptive statistics and multivariable logistic regressions were performed.

RESULTS

We identified 179 patients with sICH who required EVD placement. Of these, 52 (29%) patients died, 121 (68%) patients had hematoma expansion, and 12 (7%) patients were discharged home. Patient's mean age (± SD) was 56 (± 14), and 87 (49%) were women. The mean opening ICP (± SD) was 21 (± 8) and median ICH score (interquartile range) was 2 (2-3). Multivariable logistic regression found an association between ICP-SV and ICP-SD and hematoma expansion (odds ratio 1.6 [1.03-2.30], p = 0.035 and odds ratio 0.77 [0.63-0.93] p = 0.009, respectively).

CONCLUSIONS

Our study found an association between ICPV and hematoma expansion in patients with sICH monitored with an EVD. Measures of ICPV relating to rapid changes in ICP (ICP-SV) were associated with a higher odds of hematoma expansion, whereas measures relating to tight control of ICP (ICP-SD) were associated with a lower odds of hematoma expansion. One measure of BPV, sytolic blood pressure maximum-minimum (SBP max-min), was found to be weakly associated with discharge home (a surrogate for good functional outcome at hospital discharge). More research is needed to support these findings.

Optimal cut-off values of haematoma volume for predicting haematoma expansion at different intracerebral haemorrhage locations

BACKGROUND

Haematoma expansion (HE) is a frequent manifestation of acute intracerebral haemorrhage (ICH) and is associated with early disease progression and poor functional status. Approximately 30 % of patients with ICH experience substantial HE within the first few hours of onset.

OBJECTIVES

This study aimed to investigate the relationship between HE and initial volume at different locations in patients with ICH.

METHODS

We investigated consecutive patients with ICH admitted to the emergency room at Xiangyang No. 1 People's Hospital between January 2018 and June 2022. Haematoma volume was calculated using a three-dimensional slicer platform. Prediction models were assessed using a logistic regression model. The Youden index was used to assess the haematoma volume cut-off values for predicting HE.

RESULTS

This study included 306 patients: 161 had basal ganglia ICH, 41 lobar ICH, and 104 thalamic ICH. The area under the ROC curve (AUC) for the thalamic ICH score in predicting intraventricular haemorrhage (IVH) expansion ≥ 1 mL or delayed IVH expansion was 0.786, and the best cut-off value was 7.05 mL (specificity, 85.3 %; sensitivity, 62.8 %; and accuracy, 76.0 %). The AUC for the thalamic ICH and lobar ICH scores in predicting haematoma or IVH expansion were 0.756 and 0.653, respectively; the best cut-offs were 7.05 mL for the thalamus (specificity, 84.8 %; sensitivity, 60.0 %; and accuracy, 74.0 %) and 31.89 mL in the lobar area (specificity, 81.8 %; sensitivity, 52.3 %; and accuracy, 68.3 %).

CONCLUSIONS

Initial ICH volume predicted haematoma or IVH expansion at different locations. Moreover, it can assist clinicians in determining whether patients are suitable for future surgical interventions or other procedures.

Prognostic Neuroimaging Biomarkers in Acute Vascular Brain Injury and Traumatic Brain Injury

Prognostic imaging biomarkers after acute brain injury inform treatment decisions, track the progression of intracranial injury, and can be used in shared decision-making processes with families. Herein, key established biomarkers and prognostic scoring systems are surveyed in the literature, and their applications in clinical practice and clinical trials are discussed. Biomarkers in acute ischemic stroke include computed tomography (CT) hypodensity scoring, diffusion-weighted lesion volume, and core infarct size on perfusion imaging. Intracerebral hemorrhage biomarkers include hemorrhage volume, expansion, and location. Aneurysmal subarachnoid biomarkers include hemorrhage grading, presence of diffusion-restricting lesions, and acute hydrocephalus. Traumatic brain injury CT scoring systems, contusion expansion, and diffuse axonal injury grading are reviewed. Emerging biomarkers including white matter disease scoring, diffusion tensor imaging, and the automated calculation of scoring systems and volumetrics are discussed.

NCCT Markers of Intracerebral Hemorrhage Expansion Using Revised Criteria: An External Validation of Their Predictive Accuracy

BACKGROUND AND PURPOSE

Several NCCT expansion markers have been proposed to improve the prediction of hematoma expansion. We retrospectively evaluated the predictive accuracy of 9 expansion markers.

MATERIALS AND METHODS

Patients admitted for intracerebral hemorrhage within 24 hours of last seen well were retrospectively included from April 2016 to April 2020. The primary outcome was revised hematoma expansion, defined as any of a ≥6-mL or ≥33% increase in intracerebral hemorrhage volume, a ≥ 1-mL increase in intraventricular hemorrhage volume, or de novo intraventricular hemorrhage. We assessed the predictive accuracy of expansion markers and determined their association with revised hematoma expansion.

RESULTS

We included 124 patients, of whom 51 (41%) developed revised hematoma expansion. The sensitivity of each marker for the prediction of revised hematoma expansion ranged from 4% to 78%; the specificity, 37%-97%; the positive likelihood ratio, 0.41-7.16; and the negative likelihood ratio, 0.49-1.06. By means of univariable logistic regressions, 5 markers were significantly associated with revised hematoma expansion: black hole (OR = 8.66; 95% CI, 2.15-58.14; P = .007), hypodensity (OR = 3.18; 95% CI, 1.49-6.93; P = .003), blend (OR = 2.90; 95% CI, 1.08-8.38; P = .04), satellite (OR = 2.84; 95% CI, 1.29-6.61; P = .01), and Barras shape (OR = 2.41, 95% CI; 1.17-5.10; P = .02). In multivariable models, only the black hole marker remained independently associated with revised hematoma expansion (adjusted OR = 5.62; 95% CI, 1.23-40.23; P = .03).

CONCLUSIONS

No single NCCT expansion marker had both high sensitivity and specificity for the prediction of revised hematoma expansion. Improved image-based analysis is needed to tackle limitations associated with current NCCT-based expansion markers.

Noncontrast Computed Tomography Markers Associated with Hematoma Expansion: Analysis of a Multicenter Retrospective Study

BACKGROUND

Hematoma expansion (HE) is a significant predictor of poor outcomes in patients with intracerebral hemorrhage (ICH). Non-contrast computed tomography (NCCT) markers in ICH are promising predictors of HE. We aimed to determine the association of the NCCT markers with HE by using different temporal HE definitions.

METHODS

We utilized Risa-MIS-ICH trial data (risk stratification and minimally invasive surgery in acute intracerebral hemorrhage). We defined four HE types based on the time to baseline CT (BCT) and the time to follow-up CT (FCT). Hematoma volume was measured by software with a semi-automatic edge detection tool. HE was defined as a follow-up CT hematoma volume increase of >6 mL or a 33% hematoma volume increase relative to the baseline CT. Multivariable regression analyses were used to determine the HE parameters. The prediction potential of indicators for HE was evaluated using receiver-operating characteristic analysis.

RESULTS

The study enrolled 158 patients in total. The time to baseline CT was independently associated with HE in one type (odds ratio (OR) 0.234, 95% confidence interval (CI) 0.077-0.712, p = 0.011), and the blend sign was independently associated with HE in two types (OR, 6.203-6.985, both p < 0.05). Heterogeneous density was independently associated with HE in all types (OR, 6.465-88.445, all p < 0.05) and was the optimal type for prediction, with an area under the curve of 0.674 (p = 0.004), a sensitivity of 38.9%, and specificity of 96.0%.

CONCLUSION

In specific subtypes, the time to baseline CT, blend sign, and heterogeneous density were independently associated with HE. The association between NCCT markers and HE is influenced by the temporal definition of HE. Heterogeneous density is a stable and robust predictor of HE in different subtypes of hematoma expansion.

Using Noncontrast Computed Tomography to Improve Prediction of Intracerebral Hemorrhage Expansion

BACKGROUND

Noncontrast computed tomography hypodensities are a validated predictor of hematoma expansion (HE) in intracerebral hemorrhage and a possible alternative to the computed tomography angiography (CTA) spot sign but their added value to available prediction models remains unclear. We investigated whether the inclusion of hypodensities improves prediction of HE and compared their added value over the spot sign.

METHODS

Retrospective analysis of patients admitted for primary spontaneous intracerebral hemorrhage at the following 8 university hospitals in Boston, US (1994-2015, prospective), Hamilton, Canada (2010-2016, retrospective), Berlin, Germany (2014-2019, retrospective), Chongqing, China (2011-2015, retrospective), Pavia, Italy (2017-2019, prospective), Ferrara, Italy (2010-2019, retrospective), Brescia, Italy (2020-2021, retrospective), and Bologna, Italy (2015-2019, retrospective). Predictors of HE (hematoma growth >6 mL and/or >33% from baseline to follow-up imaging) were explored with logistic regression. We compared the discrimination of a simple prediction model for HE based on 4 predictors (antitplatelet and anticoagulant treatment, baseline intracerebral hemorrhage volume, and onset-to-imaging time) before and after the inclusion of noncontrast computed tomography hypodensities, using receiver operating characteristic curve and De Long test for area under the curve comparison.

RESULTS

A total of 2465 subjects were included, of whom 664 (26.9%) had HE and 1085 (44.0%) had hypodensities. Hypodensities were independently associated with HE after adjustment for confounders in logistic regression (odds ratio, 3.11 [95% CI, 2.55-3.80]; P<0.001). The inclusion of noncontrast computed tomography hypodensities improved the discrimination of the 4 predictors model (area under the curve, 0.67 [95% CI, 0.64-0.69] versus 0.71 [95% CI, 0.69-0.74]; P=0.025). In the subgroup of patients with a CTA available (n=895, 36.3%), the added value of hypodensities remained statistically significant (area under the curve, 0.68 [95% CI, 0.64-0.73] versus 0.74 [95% CI, 0.70-0.78]; P=0.041) whereas the addition of the CTA spot sign did not provide significant discrimination improvement (area under the curve, 0.74 [95% CI, 0.70-0.78]).

CONCLUSIONS

Noncontrast computed tomography hypodensities provided a significant added value in the prediction of HE and appear a valuable alternative to the CTA spot sign. Our findings might inform future studies and suggest the possibility to stratify the risk of HE with good discrimination without CTA.

Intracerebral haemorrhage expansion: definitions, predictors, and prevention

Haematoma expansion affects a fifth of patients within 24 h of the onset of acute intracerebral haemorrhage and is associated with death and disability, which makes it an appealing therapeutic target. The time in which active intervention can be done is short as expansion occurs mostly within the first 3 h after onset. Baseline haemorrhage volume, antithrombotic treatment, and CT angiography spot signs are each associated with increased risk of haematoma expansion. Non-contrast CT features are promising predictors of haematoma expansion, but their potential contribution to current models is under investigation. Blood pressure lowering and haemostatic treatment minimise haematoma expansion but have not led to improved functional outcomes in randomised clinical trials. Ultra-early enrolment and selection of participants on the basis of non-contrast CT imaging markers could focus future clinical trials to show clinical benefit in people at high risk of expansion or investigate heterogeneity of treatment effects in clinical trials with broad inclusion criteria.

Time to hospital presentation following intracerebral haemorrhage: Proportion of patients presenting within eight hours and factors associated with delayed presentation

PURPOSE

Prolonged time to diagnosis of primary intracerebral haemorrhage (ICH) can result in delays in obtaining appropriate blood pressure control, reversal of coagulopathy or surgical intervention in select cases. We sought to characterise the time to diagnosis in a cohort of patients with ICH and identify factors associated with delayed diagnosis.

METHODOLOGY

The stroke database of our hospital was retrospectively reviewed to identify patients presenting to our hospitals emergency department with ICH over two years (January 2017-December 2018.) Data collected included demographics (age and sex), comorbidities, anticoagulation status, clinical scores (NIHSS, GCS, ICH score), and imaging (anatomical site, haematoma size). Time from symptom onset to diagnosis and hospital presentation were recorded. Factors associated with diagnosis >8 h post ictus were assessed using a univariate and then multivariable analysis.

RESULTS

235 patients were identified with 125 males (53%) and a median age of 76 (range 40-98). For the 200 patients that initially presented to our hospital, median time to presentation was 179 min (IQR 77-584 min), and median time from ictus to imaging diagnosis was 268 min (IQR 114-717 min). 139 (70%) presented within 8 h of symptom onset, and 129 (65%) patients had imaging of the brain performed within 8 h of symptom onset. Factors associated with presentation >8 h post symptom onset included wake up stroke (OR 5.31, 95% confidence interval (CI) 2.36-11.96, p < 0.0001) and age (OR 1.04, 95% CI 1.01-1.08, p = 0.01). Patients with hemiplegia were less likely to present >8 h following ictus (OR 0.41, 95% CI 0.21-0.84, p = 0.01).

CONCLUSIONS

The majority of patients with ICH presented within 8 h of ictus. Cases of delayed diagnosis involved patients who had not incurred hemiplegia.

European Stroke Organisation (ESO) guidelines on mobile stroke units for prehospital stroke management

The safety and efficacy of mobile stroke units (MSUs) in prehospital stroke management has recently been investigated in different clinical studies. MSUs are ambulances equipped with a CT scanner, point-of-care lab, telemedicine and are staffed with a stroke specialised medical team. This European Stroke Organisation (ESO) guideline provides an up-to-date evidence-based recommendation to assist decision-makers in their choice on using MSUs for prehospital management of suspected stroke, which includes patients with acute ischaemic stroke (AIS), intracranial haemorrhage (ICH) and stroke mimics. The guidelines were developed according to the ESO standard operating procedure and Grading of Recommendations, Assessment, Development and Evaluation (GRADE) methodology. The working group identified relevant clinical questions, performed systematic reviews and aggregated data meta-analyses of the literature, assessed the quality of the available evidence and made specific recommendations. Expert consensus statements are provided where sufficient evidence was not available to provide recommendations based on the GRADE approach. We found moderate evidence for suggesting MSU management for patients with suspected stroke. The patient group diagnosed with AIS shows an improvement of functional outcomes at 90 days, reduced onset to treatment times and increased proportion receiving IVT within 60 min from onset. MSU management might be beneficial for patients with ICH as MSU management was associated with a higher proportion of ICH patients being primarily transported to tertiary care stroke centres. No safety concerns (all-cause mortality, proportion of stroke mimics treated with IVT, symptomatic intracranial bleeding and major extracranial bleeding) could be identified for all patients managed with a MSU compared to conventional care. We suggest MSU management to improve prehospital management of suspected stroke patients.

Early Deterioration and Long-Term Prognosis of Patients With Intracerebral Hemorrhage Along With Hematoma Volume More Than 20 ml: Who Needs Surgery?

Background and Purpose: The treatment of patients with intracerebral hemorrhage along with moderate hematoma and without cerebral hernia is controversial. This study aimed to explore risk factors and establish prediction models for early deterioration and poor prognosis.

Methods: We screened patients from the prospective intracerebral hemorrhage (ICH) registration database (RIS-MIS-ICH, ClinicalTrials.gov Identifier: NCT03862729). The enrolled patients had no brain hernia at admission, with a hematoma volume of more than 20 ml. All patients were initially treated by conservative methods and followed up ≥ 1 year. A decline of Glasgow Coma Scale (GCS) more than 2 or conversion to surgery within 72 h after admission was defined as early deterioration. Modified Rankin Scale (mRS) ≥ 4 at 1 year after stroke was defined as poor prognosis. The independent risk factors of early deterioration and poor prognosis were determined by univariate and multivariate regression analysis. The prediction models were established based on the weight of the independent risk factors. The accuracy and value of models were tested by the receiver operating characteristic (ROC) curve.

Results: After screening 632 patients with ICH, a total of 123 legal patients were included. According to statistical analysis, admission GCS (OR, 1.43; 95% CI, 1.18–1.74; P < 0.001) and hematoma volume (OR, 0.9; 95% CI, 0.84–0.97; P = 0.003) were the independent risk factors for early deterioration. Hematoma location (OR, 0.027; 95% CI, 0.004–0.17; P < 0.001) and hematoma volume (OR, 1.09; 95% CI, 1.03–1.15; P < 0.001) were the independent risk factors for poor prognosis, and island sign had a trend toward significance (OR, 0.5; 95% CI, 0.16-1.57; P = 0.051). The admission GCS and hematoma volume score were combined for an early deterioration prediction model with a score from 2 to 5. ROC curve showed an area under the curve (AUC) was 0.778 and cut-off point was 3.5. Combining the score of hematoma volume, island sign, and hematoma location, a long-term prognosis prediction model was established with a score from 2 to 6. ROC curve showed AUC was 0.792 and cutoff point was 4.5.

Conclusions: The novel early deterioration and long-term prognosis prediction models are simple, objective, and accurate for patients with ICH along with a hematoma volume of more than 20 ml.

Imaging markers of intracerebral hemorrhage expansion in patients with unclear symptom onset

Background:

Hematoma expansion (HE) is common and associated with poor outcome in intracerebral hemorrhage (ICH) with unclear symptom onset (USO).

Aims:

We tested the association between non-contrast computed tomography (NCCT) markers and HE in this population.

Methods:

Retrospective analysis of patients with primary spontaneous ICH admitted at five centers in the United States and Italy. Baseline NCCT was analyzed for presence of the following markers: intrahematoma hypodensities, heterogeneous density, blend sign, and irregular shape. Variables associated with HE (hematoma growth > 6 mL and/or > 33% from baseline to follow-up imaging) were explored with multivariable logistic regression.

Results:

Of 2074 patients screened, we included 646 subjects (median age = 75, 53.9% males), of whom 178 (27.6%) had HE. Hypodensities (odds ratio (OR) = 2.67, 95% confidence interval (CI) = 1.79–3.98), heterogeneous density (OR = 2.16, 95% CI = 1.46–3.21), blend sign (OR = 2.28, 95% CI = 1.38–3.75) and irregular shape (OR = 1.82, 95% CI = 1.21–2.75) were independently associated with a higher risk of HE, after adjustment for confounders (ICH volume, anticoagulation, and time from last seen well (LSW) to NCCT). Hypodensities had the highest sensitivity for HE (0.69), whereas blend sign was the most specific marker (0.90). All NCCT markers were more frequent in early presenters (time from LSW to NCCT ⩽ 6 h, n = 189, 29.3%), and more sensitive in this population as well (hypodensities had 0.77 sensitivity).

Conclusion:

NCCT markers are associated with HE in ICH with USO. These findings require prospective replication and suggest that NCCT features may help the stratification of HE in future studies on USO patients.

Noncontrast Computed Tomography Markers of Cerebral Hemorrhage Expansion: Diagnostic Accuracy Meta-Analysis

BACKGROUND AND PURPOSE

Assess the diagnostic accuracy of noncontrast computed tomography (NCCT) markers of hematoma expansion in patients with primary intracerebral hemorrhage.

METHODS

We performed a meta-analysis of observational studies and randomized controlled trials with available data for calculation of sensitivity and specificity of NCCT markers for hematoma expansion (absolute growth >6 or 12.5 mL and/or relative growth >33%). The following NCCT markers were analyzed: irregular shape, island sign (shape-related features); hypodensity, heterogeneous density, blend sign, black hole sign, and swirl sign (density-related features). Pooled accuracy values for each marker were derived from hierarchical logistic regression models.

RESULTS

A total of 10,363 subjects from 23 eligible studies were included. Significant risk of bias of included studies was noted. Hematoma expansion frequency ranged from 7% to 40%, mean intracerebral hemorrhage volume from 9 to 27.8 ml, presence of NCCT markers from 9% (island sign) to 82% (irregular shape). Among shape features, sensitivity ranged from 0.32 (95%CI = 0.20-0.47) for island sign to 0.68 (95%CI = 0.57-0.77) for irregular shape, specificity ranged from 0.47 (95%CI = 0.36-0.59) for irregular shape to 0.92 (95%CI = 0.85-0.96) for island sign; among density features sensitivity ranged from 0.28 (95%CI = 0.21-0.35) for black hole sign to 0.63 (95%CI = 0.44-0.78) for hypodensity, specificity ranged from 0.65 (95%CI = 0.56-0.73) for heterogeneous density to 0.89 (95%CI = 0.85-0.92) for blend sign.

CONCLUSION

Diagnostic accuracy of NCCT markers remains suboptimal for implementation in clinical trials although density features performed better than shape-related features. This analysis may help in better tailoring patients' selection for hematoma expansion targeted trials.

Weakly supervised multitask learning models to identify symptom onset time of unclear-onset intracerebral hemorrhage

BACKGROUND

Approximately one-third of spontaneous intracerebral hemorrhage patients did not know the onset time and were excluded from studies about time-dependent treatments for hyperacute spontaneous intracerebral hemorrhage.

AIMS

To help clinicians explore the benefit of time-dependent treatments for unclear-onset patients, we presented artificial intelligence models to identify onset time using non-contrast computed tomography (NCCT) based on weakly supervised multitask learning (WS-MTL) structure.

METHODS

The patients with reliable symptom onset time (strong label) or repeat CT (weak label) were included and split into training set and test set (internal and external). The WS-MTL structure utilized strong and weak labels simultaneously to improve performance. The models included three binary classification models for classifying whether NCCT acquired within 6, 8 or 12 h for different treatments measured by area under curve, and a regression model for determining the exact onset time measured by mean absolute error. The generalizability of models was also explored in comprehensive analysis.

RESULTS

A total of 4004 patients with 10,780 NCCT scans were included. The performance of WS-MTL classification model showed high accuracy, and that of regression model was satisfactory in ≤6 h subgroup. In comprehensive analysis, the WS-MTL showed better performance for larger hematomas and thinner scans. And the performance improved effectively as training amounts increasing and could be improved steadily through retraining.

CONCLUSIONS

The WS-MTL models showed good performance and generalizability. Considering the large number of unclear-onset spontaneous intracerebral hemorrhage patients, it may be worth to integrate the WS-MTL model into clinical practice to identify the onset time.