Carla Crivoi’s research while affiliated with University of Bucharest and other places

Background: The Glasgow Coma Scale (GCS) is a cornerstone in neurological assessment, providing critical insights into consciousness levels in patients with traumatic brain injuries and other neurological conditions. Despite its clinical importance, traditional methods for predicting GCS scores often fail to capture the complex, multi-dimensional nature of patient data. This study aims to address this gap by leveraging machine learning (ML) techniques to develop accurate, interpretable models for GCS prediction, enhancing decision making in critical care. Methods: A comprehensive dataset of 759 patients, encompassing 25 features spanning pre-, intra-, and post-operative stages, was used to develop predictive models. The dataset included key variables such as cognitive impairments, Hunt and Hess scores, and aneurysm dimensions. Six ML algorithms, including random forest (RF), XGBoost, and artificial neural networks (ANN), were trained and rigorously evaluated. Data preprocessing involved numerical encoding, standardization, and stratified splitting into training and validation subsets. Model performance was assessed using accuracy and receiver operating characteristic area under the curve (ROC AUC) metrics. Results: The RF model achieved the highest accuracy (86.4%) and mean ROC AUC (0.9592 ± 0.0386, standard deviation), highlighting its robustness and reliability in handling heterogeneous clinical datasets. XGBoost and SVM models also demonstrated strong performance (ROC AUC = 0.9502 and 0.9462, respectively). Key predictors identified included the Hunt and Hess score, aneurysm dimensions, and post-operative factors such as prolonged intubation. Ensemble methods outperformed simpler models, such as K-nearest neighbors (KNN), which struggled with high-dimensional data. Conclusions: This study demonstrates the transformative potential of ML in GCS prediction, offering accurate and interpretable tools that go beyond traditional methods. By integrating advanced algorithms with clinically relevant features, this work provides a dynamic, data-driven framework for critical care decision making. The findings lay the groundwork for future advancements, including multi-modal data integration and broader validation, positioning ML as a vital tool in personalized neurological care.

Introduction Glioblastoma (GBM), the most aggressive primary brain tumor, poses a significant challenge in predicting patient survival due to its heterogeneity and resistance to treatment. Accurate survival prediction is essential for optimizing treatment strategies and improving clinical outcomes. Methods This study utilized metadata from 135 GBM patients, including demographic, clinical, and molecular variables such as age, Karnofsky Performance Status (KPS), MGMT promoter methylation, and EGFR amplification. Six machine learning models—XGBoost, Random Forests, Support Vector Machines, Artificial Neural Networks, Extra Trees Regressor, and K- Nearest Neighbors—were employed to classify patients into predefined survival categories. Data preprocessing included label encoding for categorical variables and MinMax scaling for numerical features. Model performance was assessed using ROC-AUC and accuracy metrics, with hyperparameters optimized through grid search. Results XGBoost demonstrated the highest predictive accuracy, achieving a mean ROC-AUC of 0.90 and an accuracy of 0.78. Ensemble models outperformed simpler classifiers, emphasizing the predictive value of metadata. The models identified key prognostic markers, including MGMT promoter methylation and KPS, as significant contributors to survival prediction. Conclusions The application of machine learning to GBM metadata offers a robust approach to predicting patient survival. The study highlights the potential of ML models to enhance clinical decision-making and contribute to personalized treatment strategies, with a focus on accuracy, reliability, and interpretability.

The convergence of Artificial Intelligence (AI) and neuroscience is redefining our understanding of the brain, unlocking new possibilities in research, diagnosis, and therapy. This review explores how AI’s cutting-edge algorithms—ranging from deep learning to neuromorphic computing—are revolutionizing neuroscience by enabling the analysis of complex neural datasets, from neuroimaging and electrophysiology to genomic profiling. These advancements are transforming the early detection of neurological disorders, enhancing brain–computer interfaces, and driving personalized medicine, paving the way for more precise and adaptive treatments. Beyond applications, neuroscience itself has inspired AI innovations, with neural architectures and brain-like processes shaping advances in learning algorithms and explainable models. This bidirectional exchange has fueled breakthroughs such as dynamic connectivity mapping, real-time neural decoding, and closed-loop brain–computer systems that adaptively respond to neural states. However, challenges persist, including issues of data integration, ethical considerations, and the “black-box” nature of many AI systems, underscoring the need for transparent, equitable, and interdisciplinary approaches. By synthesizing the latest breakthroughs and identifying future opportunities, this review charts a path forward for the integration of AI and neuroscience. From harnessing multimodal data to enabling cognitive augmentation, the fusion of these fields is not just transforming brain science, it is reimagining human potential. This partnership promises a future where the mysteries of the brain are unlocked, offering unprecedented advancements in healthcare, technology, and beyond.

Background: This study presents a comprehensive analysis of 135 cases of vestibular schwannoma (VS) treated between 2006 and 2022 at the National Institute of Neurology and Neurovascular Diseases in Bucharest, Romania. The investigation focuses on the clinical presentation, treatment outcomes, and demographic trends of VS patients, highlighting region-specific insights that fill critical gaps in Eastern European data. Methods: Patients were treated with either open surgery (93.3%) or gamma knife radiosurgery (6.6%). The study identifies predominant symptoms, including hearing impairment, facial palsy, and balance disorders, with variations observed across age and gender subgroups. Comorbidities such as hypertension and obesity were prevalent, and they influenced perioperative risks. Results: Post-treatment outcomes showed a significant correlation between clinical symptoms and treatment modalities, with a majority achieving favorable results. The findings emphasize the need for tailored approaches in VS management and underscore the importance of region-specific factors in influencing clinical outcomes. Conclusions: This study contributes to refining treatment strategies and improving healthcare delivery for VS patients in Romania and beyond.

Background: Intracranial AVMs are a highly heterogeneous group of lesions that, while not very common, can pose significant risks. The therapeutic management of AVMs is complicated by ambiguous guidelines, particularly regarding which Spetzler–Martin grades should dictate specific treatment options. This study analyzed the clinical presentations and treatment approaches of 128 brain AVM cases managed between 2014 and 2022 at the National Institute of Neurology and Neurovascular Diseases in Bucharest, Romania. Methods: A retrospective analysis was conducted on patient demographics, clinical symptoms, Spetzler–Martin categorization, nidus localization, therapeutic management, and outcomes. Statistical analysis was performed using Python 3.10. Results: In our cohort of patients, the median age was 45 years, with a slight male predominance (67 males, 61 females). At admission, 51.5% presented with elevated blood pressure. The majority of patients had a Spetzler–Martin score of 2 (37.5%), followed by scores of 3 (31.3%) and 1 (20.3%). Treatment strategies included microsurgical resection in 32% of cases, conservative management in 31.2%, Gamma Knife radiosurgery in 22.6%, and endovascular embolization in 13.3%. Notably, open surgery was predominantly chosen for Grade II AVMs. The functional outcomes were favorable, with 69.5% achieving a good recovery score on the Glasgow Outcome Scale. Only four in-hospital deaths occurred, all in patients who underwent open surgery, and no deaths were recorded during the two-year follow-up. Conclusions: AVMs within the same Spetzler–Martin grade display considerable complexity, necessitating personalized treatment strategies. Our findings highlight the limitations of open surgery for Grade I cases but affirm its effectiveness for Grade II AVMs.

In this retrospective unicentric study, we analyzed the medical records of 11 patients who were surgically treated for CNS lymphoma, both primary and secondary, between 2009 and 2024. Given the rarity of CNS lymphomas and their diverse signs and symptoms based on tumoral location, our aim was to describe key aspects, such as clinical presentations and surgical management. A possible relationship between obesity and CNS lymphoma progression was investigated through an analysis of previous study findings. The literature suggests a wide spectrum of manifestations, from nausea and headaches to loss of equilibrium and speech impairment. A predominance of unsystematized balance disorders and epileptic seizures were affirmed. Notably, as emerged from our study, aphasia was a particularly interesting neurological symptom due to its rarity in the clinical features of CNSL. Other significant factors, such as tumor localization and perioperative phases, were thoroughly investigated, with the latter highlighted by an illustrative case report. Additionally, a literature review was included, comprising nine recent retrospective studies on the efficacy of surgical resection for patients diagnosed with PCNSL.