RESUMEN
OBJECTIVE: Research into the effectiveness and applicability of deep learning, radiomics, and their integrated models based on Magnetic Resonance Imaging (MRI) for preoperative differentiation between Primary Central Nervous System Lymphoma (PCNSL) and Glioblastoma (GBM), along with an exploration of the interpretability of these models. MATERIALS AND METHODS: A retrospective analysis was performed on MRI images and clinical data from 261 patients across two medical centers. The data were split into a training set (n = 153, medical center 1) and an external test set (n = 108, medical center 2). Radiomic features were extracted using Pyradiomics to build the Radiomics Model. Deep learning networks, including the transformer-based MobileVIT Model and Convolutional Neural Networks (CNN) based ConvNeXt Model, were trained separately. By applying the "late fusion" theory, the radiomics model and deep learning model were fused to produce the optimal Max-Fusion Model. Additionally, Shapley Additive exPlanations (SHAP) and Grad-CAM were employed for interpretability analysis. RESULTS: In the external test set, the Radiomics Model achieved an Area under the receiver operating characteristic curve (AUC) of 0.86, the MobileVIT Model had an AUC of 0.91, the ConvNeXt Model demonstrated an AUC of 0.89, and the Max-Fusion Model showed an AUC of 0.92. The Delong test revealed a significant difference in AUC between the Max-Fusion Model and the Radiomics Model (P = 0.02). CONCLUSION: The Max-Fusion Model, combining different models, presents superior performance in distinguishing PCNSL and GBM, highlighting the effectiveness of model fusion for enhanced decision-making in medical applications. CLINICAL RELEVANCE STATEMENT: The preoperative non-invasive differentiation between PCNSL and GBM assists clinicians in selecting appropriate treatment regimens and clinical management strategies.
RESUMEN
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.
Asunto(s)
Hemorragia Cerebral , Hematoma , Tomografía Computarizada por Rayos X , Humanos , Masculino , Femenino , Hemorragia Cerebral/diagnóstico por imagen , Hemorragia Cerebral/terapia , Hematoma/diagnóstico por imagen , Hematoma/terapia , Tomografía Computarizada por Rayos X/métodos , Anciano , Persona de Mediana Edad , Estudios Retrospectivos , Pronóstico , Tratamiento Conservador/métodos , Resultado del Tratamiento , Curva ROC , RadiómicaRESUMEN
Objectives: To investigate the value of interpretable machine learning model and nomogram based on clinical factors, MRI imaging features, and radiomic features to predict Ki-67 expression in primary central nervous system lymphomas (PCNSL). Materials and methods: MRI images and clinical information of 92 PCNSL patients were retrospectively collected, which were divided into 53 cases in the training set and 39 cases in the external validation set according to different medical centers. A 3D brain tumor segmentation model was trained based on nnU-NetV2, and two prediction models, interpretable Random Forest (RF) incorporating the SHapley Additive exPlanations (SHAP) method and nomogram based on multivariate logistic regression, were proposed for the task of Ki-67 expression status prediction. Results: The mean dice Similarity Coefficient (DSC) score of the 3D segmentation model on the validation set was 0.85. On the Ki-67 expression prediction task, the AUC of the interpretable RF model on the validation set was 0.84 (95% CI:0.81, 0.86; p < 0.001), which was a 3% improvement compared to the AUC of the nomogram. The Delong test showed that the z statistic for the difference between the two models was 1.901, corresponding to a p value of 0.057. In addition, SHAP analysis showed that the Rad-Score made a significant contribution to the model decision. Conclusion: In this study, we developed a 3D brain tumor segmentation model and used an interpretable machine learning model and nomogram for preoperative prediction of Ki-67 expression status in PCNSL patients, which improved the prediction of this medical task. Clinical relevance statement: Ki-67 represents the degree of active cell proliferation and is an important prognostic parameter associated with clinical outcomes. Non-invasive and accurate prediction of Ki-67 expression level preoperatively plays an important role in targeting treatment selection and patient stratification management for PCNSL thereby improving prognosis.