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

PROMISE: Prognostic Radiomic Outcome Measurement in Acute Subdural Hematoma Evacuation Post-Craniotomy

Background/Objectives: Traumatic acute subdural hematoma (aSDH) often requires surgical intervention, such as craniotomy, to relieve mass lesions and pressure. The extent of hematoma evacuation significantly impacts patient outcomes. This study utilizes 3D Slicer software to analyse post-craniotomy hematoma volume changes and evaluate their prognostic significance in aSDH patients. Methods: Among 178 adult patients diagnosed with aSDH from January 2015 to December 2022, 64 underwent hematoma evacuation via craniotomy. Initial scans were performed within 24 h of trauma, followed by routine postoperative scans to assess residual hematoma. We conducted radiomic analysis of preoperative and postoperative volumes, surface area, Feret diameter, sphericity, flatness, and elongation. Clinical parameters, including SOFA score, APACHE score, pupillary response, comorbidities, age, anticoagulation status, and preoperative haematocrit and haemoglobin levels, were also evaluated. Results: Changes in Δ surface area significantly correlated with 30-day outcomes (p = 0.03) and showed moderate predictive accuracy (AUC = 0.65). Patients with a Δ surface area > 30,090 mm2 experienced poorer outcomes (OR = 6.66, p = 0.02). Significant features included preoperative surface area (p = 0.009), Feret diameter (p = 0.0012). In multivariate analysis, only the Feret diameter remained significant (p = 0.01). Conclusions: Postoperative Δ surface area is, among other variables, a strong predictor of 30-day outcomes, while in multivariate analysis, preoperative Feret diameter remains the only independent predictor. Radiomic analysis with 3D Slicer may enhance prognostic accuracy and inform tailored therapeutic strategies.

Changes in hematoma volume following aneurysmal subarachnoid hemorrhage and its impact on patient prognosis

Objective

This study aims to investigate the effects of preoperative intracerebral hematoma volume (HVpre), hematoma volume 6-8 days post-surgery (HVpost), and the rate of hematoma volume change (HVpre-HVpost)/HVpre on the prognosis of patients with aneurysmal subarachnoid hemorrhage (aSAH).

Materials and methods

CT imaging data from 62 aSAH patients admitted to our hospital's Neurosurgery Department between January 2022 and December 2023 were obtained, both preoperatively and 6-8 days postoperatively. The hematoma volumes were measured using 3D-Slicer. Patients' recovery at 3 months post-discharge was assessed using the Modified Rankin Scale (mRS), categorizing the patients into a good prognosis group (mRS score 1-2) and a poor prognosis group (mRS score 3-5). Multivariate logistic regression analysis was conducted to identify independent risk factors for poor prognosis. Statistical methods were employed to compare preoperative and postoperative hematoma volumes with commonly used clinical scores. The predictive value of HVpre and HVpost for poor prognosis was evaluated using ROC curves. The rate of volume change was stratified by interquartile ranges, and the impact of different change rates on prognosis was compared.

Results

Significant differences were found between good and poor prognosis groups in age, GCS score, Hunt-Hess grade, mFisher grade, BVpre, BVpost, and (HVpre-HVpost)/HVpre (p < 0.05). Logistic regression identified gender, age, BVpre, BVpost, and volume change rate as independent risk factors (p < 0.01). Increased GCS scores and higher Hunt-Hess and mFisher grades correlated with increased HVpre and HVpost. Higher hemorrhage reduction rates were linked to better outcomes. ROC curves showed HVpre and HVpost AUC values (0.831 and 0.857, respectively) were significantly higher than clinical scales. An HVpre volume over 22.25 mL and HVpost over 15.67 mL indicated a higher risk of poor prognosis, with sensitivities of 79.3 and 80.7%, and specificities of 67.1 and 69.3%.

Conclusion

HVpre, HVpost, and (HVpre-HVpost)/HVpre can serve as neuroimaging biomarkers for assessing patients after aSAH and can effectively predict clinical prognosis.

Predicting cerebral edema in patients with spontaneous intracerebral hemorrhage using machine learning

Background

The early prediction of cerebral edema changes in patients with spontaneous intracerebral hemorrhage (SICH) may facilitate earlier interventions and result in improved outcomes. This study aimed to develop and validate machine learning models to predict cerebral edema changes within 72 h, using readily available clinical parameters, and to identify relevant influencing factors.

Methods

An observational study was conducted between April 2021 and October 2023 at the Quzhou Affiliated Hospital of Wenzhou Medical University. After preprocessing the data, the study population was randomly divided into training and internal validation cohorts in a 7:3 ratio (training: N = 150; validation: N = 65). The most relevant variables were selected using Support Vector Machine Recursive Feature Elimination (SVM-RFE) and Least Absolute Shrinkage and Selection Operator (LASSO) algorithms. The predictive performance of random forest (RF), GDBT, linear regression (LR), and XGBoost models was evaluated using the area under the receiver operating characteristic curve (AUROC), precision-recall curve (AUPRC), accuracy, F1-score, precision, recall, sensitivity, and specificity. Feature importance was calculated, and the SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) methods were employed to explain the top-performing model.

Results

A total of 84 (39.1%) patients developed cerebral edema changes. In the validation cohort, GDBT outperformed LR and RF, achieving an AUC of 0.654 (95% CI: 0.611-0.699) compared to LR of 0.578 (95% CI, 0.535-0.623, DeLong: p = 0.197) and RF of 0.624 (95% CI, 0.588-0.687, DeLong: p = 0.236). XGBoost also demonstrated similar performance with an AUC of 0.660 (95% CI, 0.611-0.711, DeLong: p = 0.963). However, in the training set, GDBT still outperformed XGBoost, with an AUC of 0.603 ± 0.100 compared to XGBoost of 0.575 ± 0.096. SHAP analysis revealed that serum sodium, HDL, subarachnoid hemorrhage volume, sex, and left basal ganglia hemorrhage volume were the top five most important features for predicting cerebral edema changes in the GDBT model.

Conclusion

The GDBT model demonstrated the best performance in predicting 72-h changes in cerebral edema. It has the potential to assist clinicians in identifying high-risk patients and guiding clinical decision-making.

A correlation study between prostate necrosis rate calculated by 3D Slicer software and clinical efficacy of prostatic artery embolization, along with an analysis of predictors of clinical success after prostatic artery embolization

PURPOSE

To analyze the correlation between the prostate necrosis rate at 1-month after prostatic artery embolization (PAE) and the clinical efficacy at 1-year after PAE, and to explore potential predictors of clinical success after PAE for the treatment of lower urinary tract symptoms secondary to benign prostatic hyperplasia (BPH).

METHODS

The prostate magnetic resonance imaging data at 1-month after PAE were imported into 3D Slicer software for calculating the prostate necrosis rate and thus analyzing the relationship between the prostate necrosis rate at 1-month after PAE and the efficacy score ratio at 1-year after PAE. The 151 patients with PAE technical success were divided into a clinical success group (n = 126) and a clinical failure group (n = 25). Independent predictors of clinical success after PAE were analyzed by multifactorial logistic regression, and the predictive performance of each factor was evaluated by applying the receiver operating characteristic curve and the area under the curve (AUC).

RESULTS

There was a linear negative correlation between the prostate necrosis rate at 1-month after PAE and the efficacy score ratio at 1-year after surgery (P < 0.001). In the clinical success group, both the initial prostate volume (PV) and the prostate necrosis rate at 1-month after PAE were significantly higher than in the clinical failure group (P < 0.001), and acute urinary retention (AUR) and adenomatous-dominant BPH were also associated with clinical success (P < 0.05). Multifactorial logistic regression analysis revealed that larger initial PV, a higher prostate necrosis rate at 1-month after surgery, and AUR were independent predictors of clinical success after PAE. The AUC values for these three indicators and their combination were 0.720, 0.928, 0.599, and 0.951, respectively, in which the prostate necrosis rate at 1-month after PAE demonstrating a high predictive value.

CONCLUSION

The higher the prostate necrosis rate at 1-month after PAE, the better the clinical efficacy at 1-year after PAE is likely to be, and the prostate necrosis rate at 1-month after PAE is expected to become a predictor of clinical success after PAE.

Percutaneous Endoscopic Transforaminal Discectomy for the Treatment of Lumbar Disc Herniation with Different Migration Levels: A Retrospective Study

Objective

To investigate the surgical method and efficacy of percutaneous endoscopic transforaminal discectomy (PETD) for the treatment of lumbar disc herniation (LDH) with different migration levels by introducing the strategy of foramenoplasty with the "distal nucleus pulposus as the core".

Methods

Clinical data of LDH patients who underwent single-segment PETD surgery were retrospectively analyzed. Three groups were categorized according to the degree of nucleus pulposus migration in the sagittal position: no migration group, mild migration group, and high migration group. Different sites of foramenoplasty were used for LDH with different degrees of migration. All patients were followed up for at least 12 months. The clinical and follow-up data of the three groups were compared.

Results

A total of 102 patients were included, of which 46 (45.1%) were in the no migration group, 36 (35.3%) in the mild migration group, and 20 (19.6%) in the high migration group. Encouraging treatment results were obtained in all three groups.

Conclusion

PETD is effective in the treatment of LDH with different degrees of migration, and the foramenoplasty concept of "distal nucleus pulposus as the core" can effectively guide the molding site of foramenoplasty and facilitate the accurate placement of the working trocar.

Exploring the Application of the Artificial-Intelligence-Integrated Platform 3D Slicer in Medical Imaging Education

Artificial Intelligence (AI) has revolutionized medical imaging procedures, specifically with regard to image segmentation, reconstruction, interpretation, and research. 3D Slicer, an open-source medical image analysis platform, has become a valuable tool in medical imaging education due to its integration of various AI applications. Through its open-source architecture, students can gain practical experience with diverse medical images and the latest AI technology, reinforcing their understanding of anatomy and imaging technology while fostering independent learning and clinical reasoning skills. The implementation of this platform improves instruction quality and nurtures skilled professionals who can meet the demands of clinical practice, research institutions, and technology innovation enterprises. AI algorithms' application in medical image processing have facilitated their translation from the lab to practical clinical applications and education.