Based on hematoma and perihematomal tissue NCCT imaging radiomics predicts early clinical outcome of conservatively treated spontaneous cerebral hemorrhage.

Spontaneous intracerebral hemorrhage (ICH) is a very serious kind of stroke. If the outcome of patients can be accurately assessed at the early stage of disease occurrence, it will be of great significance to the patients and clinical treatment. The present study was conducted to investigate whether non-contrast computer tomography (NCCT) models of hematoma and perihematomal tissues could improve the accuracy of short-term prognosis prediction in ICH patients with conservative treatment. In this retrospective analysis, a total of 166 ICH patients with conservative treatment during hospitalization were included. Patients were randomized into a training group (N = 132) and a validation group (N = 34) in a ratio of 8:2, and the functional outcome at 90 days after clinical treatment was assessed by the modified Rankin Scale (mRS). Radiomic features of hematoma and perihematomal tissues of 5 mm, 10 mm, 15 mm were extracted from NCCT images. Clinical factors were analyzed by univariate and multivariate logistic regression to identify independent predictive factors. In the validation group, the mean area under the ROC curve (AUC) of the hematoma was 0.830, the AUC of the perihematomal tissue within 5 mm, 10 mm, 15 mm was 0.792, 0.826, 0.774, respectively, and the AUC of the combined model of hematoma and perihematomal tissue within 10 mm was 0.795. The clinical-radiomics nomogram consisting of five independent predictors and radiomics score (Rad-score) of the hematoma model were used to assess 90-day functional outcome in ICH patients with conservative treatment. Our findings found that the hematoma model had better discriminative efficacy in evaluating the early prognosis of conservatively managed ICH patients. The visual clinical-radiomics nomogram provided a more intuitive individualized risk assessment for 90-day functional outcome in ICH patients with conservative treatment. The hematoma could remain the primary therapeutic target for conservatively managed ICH patients, emphasizing the need for future clinical focus on the biological significance of the hematoma itself.

Critical illness-associated cerebral microbleeds involving the corpus callosum following cardiac arrest: A case report.

RATIONALE: Critical illness-associated cerebral microbleeds (CI-aCMBs) are emerging as significant radiographic findings in patients with hypoxic ischemic injuries. Their occurrence, particularly in the corpus callosum, warrants a closer examination due to the potential implications for neurological outcomes in critically ill patients. We aim to describe a rare case of CI-aCMBs within the corpus callosum following cardiac arrest with the goal of bolstering the scientific literature on this topic. PATIENT CONCERNS: A 34-year-old man with a history of polysubstance abuse was found unconscious and experienced a pulseless electrical activity (PEA) cardiac arrest after a suspected drug overdose. Post-resuscitation, the patient exhibited severe respiratory distress, acute kidney injury, and profound neurological deficits. DIAGNOSES: Initial magnetic resonance imaging scans post-cardiac arrest showed no acute brain abnormalities. However, subsequent imaging revealed extensive cerebral microbleeds predominantly in the corpus callosum, diagnosed as CI-aCMBs. These findings were made in the absence of high signal intensity on T2-weighted images, suggesting a unique pathophysiological profile of microhemorrhages. INTERVENTIONS: The patient underwent targeted temperature management (TTM) and supportive care in the intensive care unit after cardiac arrest. OUTCOMES: He was subsequently extubated and had significant recovery without any neurological deficits. LESSONS: CI-aCMBs is a rare radiographic finding after cardiac arrest. These lesions may be confined to the corpus callosum and the long-term clinical and radiographic sequelae are still largely unknown.

Temporal and Spatial Clustering of Intracerebral Hemorrhage in Cerebral Amyloid Angiopathy.

OBJECTIVES: Cerebral amyloid angiopathy (CAA)-associated lobar intracerebral hemorrhage (ICH) has a high risk of recurrence, but the underlying mechanisms remain uncertain. We, therefore, aimed to characterize patterns of recurrent ICH. METHODS: We investigated early recurrent ICH (≥1 recurrent ICH event within 90 days of the index event) and ICH clusters (≥2 ICH events within 90 days at any time point) in 2 large cohorts of consecutive patients with first-ever ICH and available MRI. RESULTS: In 682 included patients (median age 68 years, 40.3% female, median follow-up time 4.1 years), 18 (2.6%) had an early recurrent ICH, which was associated with higher age and CAA. In patients with probable CAA, the risk of early recurrent ICH was increased 5-fold within the first 3 months compared with during months 4-12 (hazard ratio 5.41, 95% CI 2.18-13.4) while no significant difference was observed in patients without CAA. In patients with an ICH cluster, we observed spatial clustering (recurrent ICH within close proximity of index ICH in 63.0%) and a tendency for multiple sequential hemorrhages (≥3 ICH foci within 3 months in 44.4%). DISCUSSION: Our data provide evidence of both temporal and spatial clustering of ICH in CAA, suggesting a transient and localized active bleeding-prone process.

Integrating Clinical and Neuroimaging Markers to Predict the Onset of Posthemorrhagic Ventricular Dilatation in Preterm Neonates.

BACKGROUND: Posthemorrhagic ventricular dilatation (PHVD) is a major complication of intraventricular hemorrhage (IVH); it is associated with high risks of cerebral palsy and cognitive deficits compared with infants without PHVD. This study aims to explore the early perinatal risk factors-associated with the risk of progressive PHVD. METHODS: Neonates ≤29 weeks gestational age (GA) with Grade II-III IVH and periventricular hemorrhagic infarct (PVHI) between 2015 and 2021 were retrospectively reviewed. All cranial ultrasounds done within 14 days postnatal age (PNA) were assessed for grade of IVH, anterior horn width (AHW), ventricular index (VI), and thalamo-occipital index (TOD). The outcome was defined as death of any cause or VI and/or AHW and/or TOD ≥ moderate-risk zone based on an ultrasound done beyond two weeks PNA. RESULTS: A total of 146 infants with a mean GA of 26 ± 1.8 weeks, birth weight 900 ± 234 g were included, 46% were females. The primary outcome occurred in 56 (39%) infants; among them 17 (30%) and 11 (20%) needed ventricular reservoir and shunt insertion, respectively. The risk factors present within 14 days PNA that significantly increased the odds of developing PHVD were hemodynamically significant patent ductus arteriosus (odds ratio [OR] 6.1, 95% confidence interval [CI] 1.9 to 22), culture-proven sepsis (OR 5.4, 95% CI 1.8 to 18), Grade III IVH (OR 4.6, 95% CI 1.1 to 22), PVHI (OR 3.0, 95% CI 0.9 to 10), and VI (OR 2.1, 95% CI 1.6 to 2.9). CONCLUSIONS: Clinical predictors such as significant ductus arteriosus and bacterial septicemia, along with risk levels of AHW and VI measured with early cranial ultrasounds, are potential predictors of subsequent onset of PHVD.

Evaluating the interaction between hemorrhagic transformation and cerebral edema on functional outcome after ischemic stroke.

BACKGROUND: Hemorrhagic transformation (HT) and cerebral edema (CED) are both major complications following ischemic stroke, but few studies have evaluated their overlap. We evaluated the frequency and predictors of CED/HT overlap and whether their co-occurrence impacts functional outcome more than each in isolation. METHODS: 892 stroke patients enrolled in a prospective study had follow-up CT imaging evaluated for HT and CED; the latter was quantified using the ratio of hemispheric CSF volumes (with hemispheric CSF ratio < 0.90 used as the CED threshold). The interaction between HT and CED on functional outcome (using modified Rankin Scale at 3 months) was compared to that for each condition separately. RESULTS: Among the 275 (31%) who developed HT, 233 (85%) manifested hemispheric CSF ratio < 0.9 (CED/HT), with this overlap group representing half of the 475 with measurable CED. Higher baseline NIHSS scores and larger infarct volumes were observed in the CED/HT group compared with those with CED or HT alone. Functional outcome was worse in those with CED/HT [median mRS 3 (IQR 2-5)] than those with CED [median 2 (IQR 1-4)] or HT alone [median 1 (IQR 0-2), p < 0.0001]. Overlap of CED/HT independently predicted worse outcome [OR 1.89 (95% CI: 1.12-3.18), p = 0.02] while HT did not; however, CED/HT was no longer associated with worse outcome after adjusting for severity of CED [adjusted OR 0.35 (95% CI: 0.23, 0.51) per 0.21 lower hemispheric CSF ratio, p < 0.001]. CONCLUSIONS: Most stroke patients with HT also have measurable CED. The co-occurrence of CED and HT occurs in larger and more severe strokes and is associated with worse functional outcome, although this is driven by greater severity of stroke-related edema in those with HT.

A comprehensive prediction model predicts perihematomal edema growth in the acute stage after intracerebral hemorrhage.

BACKGROUND: Perihematomal edema (PHE) is regarded as a potential intervention indicator of secondary injury following intracerebral hemorrhage (ICH). But it still lacks a comprehensive prediction model for early PHE formation. METHODS: The included ICH patients have received an initial Computed Tomography scan within 6 hours of symptom onset. Hematoma volume and PHE volume were computed using semiautomated computer-assisted software. The volume of the hematoma, edema around the hematoma, and surface area of the hematoma were calculated. The platelet-to-lymphocyte ratio (PLR) was calculated by dividing the platelet count by the lymphocyte cell count. All analyses were 2-tailed, and the significance level was determined by P <0.05. RESULTS: A total of 226 patients were included in the final analysis. The optimal cut-off values for PHE volume increase to predict poor outcomes were determined as 5.5 mL. For clinical applicability, we identified a value of 5.5 mL as the optimal threshold for early PHE growth. In the multivariate logistic regression analyses, we finally found that baseline hematoma surface area (p < 0.001), expansion-prone hematoma (p < 0.001), and PLR (p = 0.033) could independently predict PHE growth. The comprehensive prediction model demonstrated good performance in predicting PHE growth, with an area under the curve of 0.841, sensitivity of 0.807, and specificity of 0.732. CONCLUSION: In this study, we found that baseline hematoma surface area, expansion-prone hematoma, and PLR were independently associated with PHE growth. Additionally, a risk nomogram model was established to predict the PHE growth in patients with ICH.

Clinical Reasoning: A 50-Year-Old Man With Intracerebral Hemorrhage and Tortuous Retinal Arterioles.

A 50-year-old man presented with headache. Examination showed left sided ataxic hemiparesis and elevated blood pressure. Brain imaging revealed an acute intracerebral hemorrhage in the right lentiform nucleus, deep and periventricular white matter hyperintensities, and predominantly deep cerebral microbleeds. Fundus examination showed important arteriolar tortuosity involving several blood vessels. In this young patient, we explain the diagnostic approach to intracerebral hemorrhage, the causes of cerebral small vessel disease, and the interpretation of biomolecular tests.

Characteristics, predictors and outcomes of early postoperative cerebral infarction on computed tomography in spontaneous intracerebral hemorrhage.

Early postoperative cerebral infarction (ePCI) is a serious complication of spontaneous intracerebral hemorrhage (SICH). Yet, no study has specifically focused on ePCI among SICH patients. Our study aims to investigate the characteristics, predictors, and outcomes of ePCI observed on computed tomography (CT) within 72 h after surgery in patients with supratentorial SICH. Data from a single-center SICH study conducted from May 2015 to September 2022 were retrospectively analyzed. We described the characteristics of ePCI. Predictors were identified through logistic regression analysis, and the impact of ePCI on six-month mortality was examined using a Cox regression model. Subgroup analyses and the "E-value" approach assessed the robustness of the association between ePCI and mortality. A retrospective analysis of 637 out of 3938 SICH patients found that 71 cases (11.1%) developed ePCI. The majority of ePCI cases occurred on the bleeding side (40/71, 56.3%) and affected the middle cerebral artery (MCA) territory (45/71, 63.4%). Multivariable analysis showed that the Glasgow Coma Scale (GCS) score (odds ratio (OR), 0.62; 95% CI, 0.48-0.8; p < 0.001), bleeding volume (per 100 ml) (OR, 1.17; 95% CI, 1.03-1.32; p = 0.016), hematoma volume (per 10 ml) (OR, 1.14; 95%CI, 1.02-1.28; p = 0.023) and bilateral brain hernia (OR, 6.48; 95%CI, 1.71-24.48; p = 0.006) independently predicted ePCI occurrence. ePCI was significantly associated with increased mortality (adjusted hazard ratio (HR), 3.6; 95% CI, 2.2-5.88; p < 0.001). Subgroup analysis and E-value analysis (3.82-6.66) confirmed the stability of the association. ePCI is a common complication of SICH and can be predicted by low GCS score, significant bleeding, large hematoma volume, and brain hernia. Given its significant increase in mortality, ePCI should be explored in future studies.

Synergistic insights: the integrated role of CT/CTP and clinical parameters in hemorrhagic transformation prediction.

BACKGROUND: Acute ischemic stroke presents significant challenges in healthcare, notably due to the risk and poor prognosis associated with hemorrhagic transformation (HT). Currently, there is a notable gap in the early clinical stage for a valid and reliable predictive model for HT. METHODS: This single-center retrospective study analyzed data from 224 patients with acute ischemic stroke due to large vessel occlusion. We collected comprehensive clinical data, CT, and CTP parameters. A predictive model for HT was developed, incorporating clinical indicators alongside imaging data, and its efficacy was evaluated using decision curve analysis and calibration curves. In addition, we have also built a free browser-based online calculator based on this model for HT prediction. RESULTS: The study identified atrial fibrillation and hypertension as significant risk factors for HT. Patients with HT showed more extensive initial ischemic damage and a smaller ischemic penumbra. Our novel predictive model, integrating clinical indicators with CT and CTP parameters, demonstrated superior predictive value compared to models based solely on clinical indicators. CONCLUSIONS: The research highlighted the intricate interplay of clinical and imaging parameters in HT post-thrombectomy. It established a multifaceted predictive model, enhancing the understanding and management of acute ischemic stroke. Future studies should focus on validating this model in broader cohorts, further investigating the causal relationships, and exploring the nuanced effects of these parameters on patient outcomes post-stroke.

Precranial artery calcification burden: a potential indicator of the clinical outcome of reperfusion in patients with acute large artery occlusion.

OBJECTIVE: With mechanical thrombectomy (MT), we investigated the prognostic importance of aortic arch calcification (AoAC) and carotid sinus calcification (CaSC) for symptomatic intracerebral hemorrhage (sICH) and poor outcome in acute large artery occlusion (LAO). METHODS: In this retrospective observational study, we calculated pre-cranial artery calcification burden (PACB) scores (burden score of AoAC and CaSC) using the AoAC grading scale score plus Woodcock visual score. The outcome measure was sICH per the European Cooperative Acute Stroke Study III definition. A 3-month modified Rankin scale score 3-6 was designated as poor outcome. RESULTS: Compared with patients who had PACB <3, those with PACB ≥3 showed substantially higher risks of sICH (odds ratio [OR] = 2.567, 95% confidence interval [CI] = 1.187-5.550) and poor outcome (OR = 4.777, 95% CI = 1.659-13.756). According to receiver operating characteristic (ROC) curves, adding PACB to the regression model enhanced the predictive value for poor outcome (area under the ROC curve [AUC]: 0.718 vs. 0.519, Z = 2.340) and in patients receiving MT (AUC: 0.714 vs. 0.584, Z = 2.021), independently. CONCLUSIONS: Factors related to PACB were consistent with common risk factors of systemic atherosclerosis. Low PACB scores indicated better prognosis. In patients with LAO following MT, PACB was useful in predicting sICH and poor clinical outcome.

Associations of Microbleeds and Their Topography With Imaging and CSF Biomarkers of Alzheimer Pathology in Individuals With Down Syndrome.

BACKGROUND AND OBJECTIVES: Cerebral hemorrhages are an exclusion criterion and potential adverse effect of antiamyloid agents. It is, therefore, critical to characterize the natural history of cerebral microbleeds in populations genetically predisposed to Alzheimer disease (AD), such as Down syndrome (DS). We aimed to assess microbleed emergence in adults with DS across the AD spectrum, defining their topography and associations with clinical variables, cognitive outcomes, and fluid and neuroimaging biomarkers. METHODS: This cross-sectional study included participants aged 18 years or older from the Down-Alzheimer Barcelona Neuroimaging Initiative and Sant Pau Initiative on Neurodegeneration with T1-weighted and susceptibility-weighted images. Participants underwent comprehensive assessments, including apolipoprotein E (APOE) genotyping; fluid and plasma determinations of beta-amyloid, tau, and neurofilament light; cognitive outcomes (Cambridge Cognitive Examination and modified Cued Recall Test); and vascular risk factors (hypertension, diabetes mellitus, and dyslipidemia). We manually segmented microbleeds and characterized their topography. Associations between microbleed severity and AD biomarkers were explored using between-group comparisons (none vs 1 vs 2+) and multivariate linear models. RESULTS: We included 276 individuals with DS and 158 healthy euploid controls (mean age = 47.8 years, 50.92% female). Individuals with DS were more likely to have microbleeds than controls (20% vs 8.9%, p < 0.001), with more severe presentation (12% with 2+ vs 1.9%). Microbleeds increased with age (12% 20-30 years vs 60% > 60 years) and AD clinical stage (12.42% asymptomatic, 27.9% prodromal, 35.09% dementia) were more common in APOEε4 carriers (26% vs 18.3% noncarriers, p = 0.008), but not associated with vascular risk factors (p > 0.05). Microbleeds were predominantly posterior (cerebellum 33.66%; occipital 14.85%; temporal 21.29%) in participants with DS. Associations with microbleed severity were found for neuroimaging and fluid AD biomarkers, but only hippocampal volumes (standardized ß = -0.18 [-0.31, -0.06], p < 0.005) and CSF p-tau-181 concentrations (ß = 0.26 [0.12, 0.41], p < 0.005) survived regression controlling for age and disease stage, respectively. Microbleeds had limited effect on cognitive outcomes. DISCUSSION: In participants with DS, microbleeds present with a posterior, lobar predominance, are associated with disease severity, but do not affect cognitive performance. These results suggest an interplay between AD pathology and vascular lesions, implicating microbleeds as a risk factor limiting the use of antiamyloid agents in this population.

Predicting intracerebral hemorrhage after endovascular therapy for anterior circulation strokes using CT-ASPECT, CTP-ASPECT and DWI-ASPECT: Protocol for a systematic review.

INTRODUCTION: Over the last decade, there have been significant advances in treatments for anterior ischemic stroke, most notably endovascular thrombectomy (EVT). Despite the success of EVT on overall outcomes, intracerebral hemorrhage (ICH) is an important post-procedure complication, often associated with mortality and disability. Hence, predicting the risk of ICH can inform EVT decision making. The ASPECT score is used globally to predict patients' prognosis post-reperfusion therapy. Our objective is to perform a systematic review to collect and synthesize data on the association between ASPECT scores on CT, CTP and DWI-MRI (CT-ASPECT, CTP-ASPECT, and DWI-ASPECT) and the risk of symptomatic ICH after EVT for anterior circulation strokes. METHODS AND ANALYSIS: We will conduct a broad search of various electronic databases (MEDLINE, EMBASE, CINAHL, PsycINFO, Web of Science, and the Cochrane Database of Systematic Reviews) to identify studies published after January 1st, 2012 (commonly accepted as the beginning of the modern EVT era based on availability of stent-retrievers). Two independent reviewers will screen and include studies evaluating associations between symptomatic ICH after thrombectomy and baseline CT-ASPECT, CTP-ASPECT and DWI-ASPECT scores. Data will be extracted to quantify the risk of sICH after EVT based on the ASPECT scoring. TRIAL REGISTRATION: PROSPERO registration number: CRD42023459860.

Prognostic value of CT scan-based radiomics in intracerebral hemorrhage patients: A systematic review and meta-analysis.

OBJECTIVES: We conducted a systematic review and meta-analysis of current publications on the potential role of non-contrast-enhanced computed tomography (NCCT) radiomics as a prognostic indicator in patients with intracerebral hemorrhage (ICH). METHODS: We systematically searched PubMed, EMBASE, and the Web of Science from inception until January 8, 2024. Studies with NCCT-based radiomics features for predicting the prognostic outcomes of ICH patients were included. We calculated the pooled sensitivity, specificity, diagnostic odds ratio (DOR), and area under curve (AUC) values. The radiomics quality score (RQS), METhodological RadiomICs Score (METRICS), and the quality assessment of diagnostic accuracy studies (QUADAS-2) were used for quality assessment. RESULTS: Twenty-two studies were included. The pooled sensitivity, specificity, DOR, and AUC of radiomics models were 0.73, 0.78, 10.03, and 0.83, respectively, while on the combined radiomics models with other non-radiomics features were 0.80, 0.80, 16.28, and 0.86. Subgroup analysis showed that studies with the following covariates have a higher accuracy: single center, modified Rankin Scale (mRS) criteria for the ICH outcomes assessment, following patients for evaluation of ICH outcomes for more than a month, automatic segmentation, capturing the radiomics feature from the only intra-hematomal region, using PyRadiomic tool for features extraction, and using non-logistic regression for modeling. The quality of literature using QUADAS-2 and METRICS tools was good and was under-average using the RQS tool. No publication bias was detected. CONCLUSIONS: Radiomics features showed moderate to high accuracy for predicting ICH prognostic outcomes. Although the QUADAS-2 and METRICS assessments indicated good quality, the radiomics pipeline quality was under-average. CLINICAL RELEVANCE: NCCT-based radiomics features can provide information about the prognostic outcomes of ICH patients after patient admission. This study exploits the value of current evidence on NCCT-based radiomics methodology in the prediction of ICH prognosis.

Determination of hemorrhagic transformation risk in acute ischemic cerebrovascular disease: The relationship between ADC values and GRE hemo sequence microhemorrhage.

Background and purpose:

To determine the risk of hemorrhagic transformation in patients with acute ischemic cerebrovascular disease, we investigated the relationship between Apparent Diffusion Coefficient Magnetic Resonance Imaging values measured within the infarct area and microbleeds observed on Gradient Echo Sequence Magnetic Resonance Imaging.

A total of 172 patients who were hospitalized to the Neurology Clinic of the Istanbul Provincial Directorate of Health Erenkoy Mental and Nervous Diseases Training and Research Hospital between June 2019 and March 2020 were included in this cross-sectional study. The patients were classified according to their demographic and clinical characteristics, by age, gender, hypertension, diabetes mellitus, smoking, and alcohol use. In the Cranial Magnetic Resonance Imaging taken in the application for the diagnosis of acute ischemic cerebrovascular disease, the infarction area in Apparent Diffusion Coefficient sequences with the Region of Interest value and the relationship between microhemorrhage observed in Gradient Echo hemo-sequence was evaluated in Magnetic Resonance Imaging applied between 3-7 days. While calculating spherical Region of Interest values, 5 Region of Interest values were obtained for lesions larger than 1.5x1.5 cm, and their arithmetic mean was obtained, and single spherical Region of Interest value was taken for smaller ones. Apparent Diffusion Coefficient Magnetic Resonance Imaging Region of Interest mean values were divided into 2 groups as <500x10–6 mm2 /s and >500x10–6 mm2/s. 

Patients in the group with Apparent Diffusion Coefficient Region of Interest mean values below 500x10-6 mm2/s, had a significantly higher probability for microhemorrhage observes in Gradient Echo Sequence Magnetic Resonance Imaging (p: 0.001) and also more likely to experience microhemorrhage in other areas, which was statistically significant (p: 0.001).The probability of another micro-bleeding observed in patients with microhemorrhage Gradient Echo Sequence Magnetic Resonance Imaging was also statistically significant (p: 0,001). The risk of microbleeding in areas other than ischemia was also found to be significantly higher in patients with microbleeding in the ischemia area in Gradient Echo Sequence Magnetic Resonance Imaging.

In our study, a statistically significant relationship was found between the microhemorrhage in the infarct area and the Apparent Diffusion Coefficient Region of Interest values. When the literature was reviewed, no such study was found to determine the risk of bleeding.

Improved Precision and Safety of Supra-Tentorial Intracranial Hematoma Puncture Using C-Arm CT Four-Dimensional Navigation.

BACKGROUND Spontaneous intracerebral hemorrhage has a high fatality rate within the initial month after onset. This study determined the safety and therapeutic efficacy of minimally invasive puncture for supra-tentorial intracranial hematoma under C-arm computed tomography (CT) 4-dimensional navigation. MATERIAL AND METHODS We retrospectively analyzed 64 patients with supra-tentorial cerebral hemorrhage from June 2020 to May 2023; 31 patients were assigned to the study group (C-arm CT navigation puncture) and 33 patients were in the control group (conventional CT-guided puncture). The analysis focused on assessment of puncture error, postoperative complication rate, and the Glasgow Outcome Scale (GOS) and National Institute of Health Stroke Scale (NIHSS) scores 30 and 90 days after surgery. RESULTS C-arm CT navigation puncture had improved precision, with significantly reduced transverse (3.17±1.75 mm) and longitudinal (1.83±1.21 mm) deviations, compared with the control group (7.88±1.74 mm and 5.50±1.84 mm, respectively; P<0.05). The overall postoperative complication rate was significantly lower in the study group than in the control group (12.90% vs 36.36%, P<0.05). The mean GOS score was higher in the study group than in the control group 30 and 90 days postoperatively (3.42±0.96 and 3.97±0.95 vs 2.94±0.79 and 3.46±0.90, respectively; P<0.05), while the mean NIHSS score was lower in the study group than in the control group 30 and 90 days postoperatively (10.58±6.52 and 5.97±4.55 vs 14.42±8.13 and 9.55±8.31, respectively; P<0.05). CONCLUSIONS Supra-tentorial intracranial hematoma puncture under C-arm CT 4-dimensional navigation is accurate, safe, and beneficial.

An interpretable artificial intelligence model based on CT for prognosis of intracerebral hemorrhage: a multicenter study.

OBJECTIVES: To develop and validate a novel interpretable artificial intelligence (AI) model that integrates radiomic features, deep learning features, and imaging features at multiple semantic levels to predict the prognosis of intracerebral hemorrhage (ICH) patients at 6 months post-onset. MATERIALS AND METHODS: Retrospectively enrolled 222 patients with ICH for Non-contrast Computed Tomography (NCCT) images and clinical data, who were divided into a training cohort (n = 186, medical center 1) and an external testing cohort (n = 36, medical center 2). Following image preprocessing, the entire hematoma region was segmented by two radiologists as the volume of interest (VOI). Pyradiomics algorithm library was utilized to extract 1762 radiomics features, while a deep convolutional neural network (EfficientnetV2-L) was employed to extract 1000 deep learning features. Additionally, radiologists evaluated imaging features. Based on the three different modalities of features mentioned above, the Random Forest (RF) model was trained, resulting in three models (Radiomics Model, Radiomics-Clinical Model, and DL-Radiomics-Clinical Model). The performance and clinical utility of the models were assessed using the Area Under the Receiver Operating Characteristic Curve (AUC), calibration curve, and Decision Curve Analysis (DCA), with AUC compared using the DeLong test. Furthermore, this study employs three methods, Shapley Additive Explanations (SHAP), Grad-CAM, and Guided Grad-CAM, to conduct a multidimensional interpretability analysis of model decisions. RESULTS: The Radiomics-Clinical Model and DL-Radiomics-Clinical Model exhibited relatively good predictive performance, with an AUC of 0.86 [95% Confidence Intervals (CI): 0.71, 0.95; P < 0.01] and 0.89 (95% CI: 0.74, 0.97; P < 0.01), respectively, in the external testing cohort. CONCLUSION: The multimodal explainable AI model proposed in this study can accurately predict the prognosis of ICH. Interpretability methods such as SHAP, Grad-CAM, and Guided Grad-Cam partially address the interpretability limitations of AI models. Integrating multimodal imaging features can effectively improve the performance of the model. CLINICAL RELEVANCE STATEMENT: Predicting the prognosis of patients with ICH is a key objective in emergency care. Accurate and efficient prognostic tools can effectively prevent, manage, and monitor adverse events in ICH patients, maximizing treatment outcomes.