Search Results (97)

Background/Objectives: Sellar solitary fibrous tumors (SFTs) are exceptionally rare mesenchymal neoplasms that frequently mimic non-functioning pituitary adenomas (PAs) because of overlapping clinical manifestations and nonspecific radiological findings. Consequently, preoperative diagnosis remains challenging and definitive diagnosis relies on histopathological and immunohistochemical evaluation. Methods: We report a sellar SFT initially diagnosed as a PA and analyze the diagnostic features of previously reported cases to identify recurring diagnostic pitfalls. Clinical, endocrinological, radiological, intraoperative, histopathological, and immunohistochemical findings from a patient with a sellar SFT were retrospectively reviewed. A structured literature review of previously reported sellar SFTs was performed to compare presenting symptoms, endocrine abnormalities, imaging characteristics, pathological findings, and diagnostic features. Results: A 65-year-old man presented with headache, progressive visual impairment, fatigue, and anterior hypopituitarism. Magnetic resonance imaging demonstrated a heterogeneously enhancing sellar lesion with suprasellar extension and cavernous sinus involvement, leading to an initial diagnosis of non-functioning PA. Endoscopic transsphenoidal surgery revealed an unexpectedly hypervascular and firm tumor. Histopathological examination demonstrated a spindle-cell neoplasm with a hemangiopericytoma-like vascular pattern, six mitoses per 10 high-power fields, absence of necrosis, and diffuse nuclear STAT6 positivity, establishing the diagnosis of CNS WHO grade 2 solitary fibrous tumor according to the 2021 WHO classification. Review of the literature demonstrated that most sellar SFTs share similar clinical and radiological features with PAs and are diagnosed only after surgical resection. Conclusions: Sellar SFT should be considered in the differential diagnosis of atypical sellar masses despite the absence of characteristic imaging findings. Recognition of intraoperative features, together with appropriate immunohistochemical evaluation, particularly STAT6 staining, is essential for accurate diagnosis. Current evidence remains insufficient to define the optimal postoperative management of completely resected sellar SFTs, emphasizing the importance of individualized treatment decisions and long-term surveillance. Full article

Brain tumor classification from magnetic resonance imaging (MRI) remains challenging in settings where only image-level labels are available and tumor classes exhibit overlapping visual characteristics. In this study, we consider the publicly available Brain Tumor MRI Dataset from Kaggle, a four-class dataset composed of 2D MRI slices belonging to the categories glioma, meningioma, pituitary tumor, and no tumor. Accordingly, the proposed framework is formulated as a slice-based multiclass classification approach rather than a volumetric 3D analysis pipeline. We propose a lightweight and interpretable framework that integrates handcrafted multiscale MRI descriptors, an artificial neural network (ANN), Bee Colony Optimization (BCO)-based neural architecture search, and fuzzy softmax confidence modeling. Each MRI slice is represented by a compact 9-dimensional feature vector derived from intensity, local entropy, and gradient magnitude computed globally and over non-overlapping spatial blocks. The ANN design problem is formulated as a discrete–continuous optimization task, where BCO is employed to optimize network architecture and training hyperparameters by maximizing validation macro-$F_1$. To quantify predictive reliability, the softmax outputs are interpreted as fuzzy class memberships and further analyzed using maximum membership, normalized entropy, decision margin, and ambiguity measures, enabling confidence-aware reliability assessment. These fuzzy confidence descriptors enable confidence-threshold-based selective classification and rejection of low-confidence predictions. Across repeated runs, the optimized BCO-ANN achieved a mean test accuracy of $0.781\pm 0.009$, mean macro-$F_1$ of $0.775\pm 0.010$, mean Brier score of $0.319\pm 0.012$, and mean Expected Calibration Error (ECE) of $0.0273\pm 0.0080$, compared with $0.748\pm 0.011$, $0.738\pm 0.013$, $0.352\pm 0.010$, and $0.0446\pm 0.0071$ for the baseline ANN, respectively. Under confidence-threshold-based rejection, selective macro-$F_1$ increased to $0.820\pm 0.009$ at $\tau =0.55$ and to $0.874\pm 0.020$ at $\tau =0.85$, with the expected reduction in coverage. These results indicate that the proposed framework provides a transparent and reproducible approach for optimization-aware and confidence-aware multiclass brain tumor MRI classification in a lightweight handcrafted feature setting. Full article

Posterior pituitary and hypothalamic neuronal tumors are uncommon sellar and suprasellar neoplasms that can mimic pituitary neuroendocrine tumors clinically and radiologically. The 5th edition World Health Organization classifications (Endocrine and Neuroendocrine Tumors) reinforce a lineage-based framework that separates anterior pituitary tumors from posterior pituitary and hypothalamic neuronal lineages, which is particularly important in hormone-negative lesions and limited tissue samples. This narrative review provides a practical, pathology-centered approach to classification by integrating key anatomic and radiologic clues with histomorphology and targeted immunohistochemistry. We highlight the value and limitations of thyroid transcription factor 1, outline a stepwise workflow incorporating anterior pituitary transcription factors and neuronal differentiation markers, and discuss when vasopressin immunostaining is informative. We also summarize selected molecular insights and clinical management considerations relevant to surgical planning and follow-up. Full article

The diagnosis and treatment planning for brain tumors remain a complex task in medical imaging, largely due to the intricate structure of such abnormalities. This study introduces an interpretable stacked deep learning framework consisting of three sequential stages: (i) tumor segmentation, (ii) feature extraction, and (iii) tumor classification. The segmentation stage introduces a three-parameter lambda distribution (TPLD), a symmetric special case of generalized lambda distribution (GLD), used as a statistical intensity prior that is fused into the gating signal of an Attention U-Net for enhancing boundary delineation. The segmented outputs are processed using InceptionV3 for deep feature extraction and followed by a convolutional neural network (CNN) classifier. We evaluated the proposed model on the BRISC 2025 dataset, consisting of T1 weighted brain MRI images with pixel wise segmentation masks, which is validated by medical experts. The dataset consists of 3933 training images and 860 test images with ground truth masks, containing the four classes: meningioma, glioma, no tumor, and pituitary tumor. We utilized a region-of-interest–based training strategy to reduce the computational complexity and minimize overfitting. The data split followed the official image-level partition distributed with BRISC 2025; because patient identifiers are not released with the dataset, patient-level separation could not be independently verified, and this is acknowledged as a limitation. To ensure methodological transparency and clinical robustness, we systematically report the learning dynamics across 20, 60, and 100 training epochs at multiple decision thresholds (0.50, 0.60, 0.70), providing evidence of stable model convergence without overfitting. We also introduce a composite loss function by integrating cross-entropy, focal losses, and Dice to further boost performance. Experimental results demonstrate 97.8% classification accuracy on the test set, 92.4% Dice coefficient, and 85.9% IoU at the optimal threshold of 0.60. An ablation study further confirms the contribution of each loss component, supporting reproducibility and transparency in model evaluation. These findings confirm the practical utility and reliability of the proposed framework in the context of brain tumor segmentation and clinical diagnosis. Full article

Background: The tumor immune microenvironment, particularly the role of cytotoxic CD8+ T lymphocytes, is crucial in cancer progression but remains poorly understood in pituitary neuroendocrine tumors (PitNETs). The significance of CD8+ cell infiltration varies across PitNET subtypes, suggesting a complex interplay with tumor cell lineage. This study aimed to characterize the distribution of CD8+ tumor-infiltrating lymphocytes across different PitNET subtypes defined by the current WHO classification and to explore their association with clinicopathological features. Methods: We conducted a retrospective study on 40 surgically resected PitNETs. All cases were classified based on immunohistochemical expression of pituitary hormones and lineage-specific transcription factors (PIT-1, TPIT, SF-1). CD8+ lymphocyte density was quantified using immunohistochemistry and calculated as cells/mm². Exploratory statistical analysis was performed based on non-parametric tests to compare CD8+ cell density across tumor subtypes and with parameters like tumor size, invasiveness (Knosp grade), and proliferation index (Ki-67). Findings are to be treated as observational trends. Results: The highest density of CD8+ lymphocytes was observed in plurihormonal PIT-1-positive tumors [17.61 cells/mm² (IQR: 17.61–60.36)], followed by somatotroph [13.2 (6.6–15.72)] and mammosomatotroph [13.83 (0–21.38)] tumors. A difference in CD8+ density was found between PIT-1-positive and PIT-1-negative tumors (n1 = 34, n2 = 6, U = 49.5, p_exact = 0.050, r = 0.33); the medium effect size indicates a possible lineage-related trend. Another difference was observed between SF-1-positive and SF-1-negative tumors (p = 0.025), with SF-1 lineage tumors showing the lowest infiltration. No correlations were found between CD8+ density and tumor size, Knosp grade, or Ki-67 index. Conclusions: The distribution of intratumoral CD8+ T lymphocytes in PitNETs is highly heterogeneous and appears to be strongly dictated by the transcription factor-defined tumor lineage rather than by traditional clinicopathological markers of aggressiveness. PIT-1 lineage tumors harbor a more active immune microenvironment, while SF-1 lineage tumors are relatively ‘immune-poor’. These findings highlight the immunological diversity of PitNETs and support further investigation of the tumor immune landscape. Collaborative multi-institutional studies are required to validate these trends. Full article

Background/Objectives: Brain tumor analysis using magnetic resonance imaging (MRI) remains a challenging task due to tumor heterogeneity, complex anatomical structures, and reliance on expert interpretation. Although deep learning approaches have shown promising results in medical image analysis, many existing studies focus on either tumor classification or segmentation independently, limiting their applicability in comprehensive automated brain tumor analysis workflows. This study proposes an integrated dual-task deep learning framework for automated brain tumor classification and segmentation using MRI scans. The framework aims to provide complementary diagnostic support by combining tumor-type prediction and tumor boundary delineation within an integrated workflow. Methods: The proposed framework utilizes EfficientNet-based convolutional neural networks for multi-class brain tumor classification and U-Net++ architectures with EfficientNet encoders for tumor segmentation. Experiments were conducted using the BRISC2025 dataset, consisting primarily of 6000 T1-weighted 2D MRI slices collected from axial, coronal, and sagittal planes. Standard preprocessing, augmentation, transfer learning, and selective fine-tuning strategies were applied. Multiple architectures were systematically evaluated using evaluation metrics. Results: EfficientNet-B1 achieved a classification accuracy of 99.70% with near-perfect precision, recall, and F1-scores across glioma, meningioma, pituitary tumor, and no-tumor classes. For segmentation, U-Net++ with an EfficientNet-B1 encoder achieved a Dice score of 0.9055, an IoU score of 0.8442, and an HD95 value of 12.21 pixels on the held-out test set. The proposed framework demonstrated robust performance in detecting small and low-contrast tumor regions while maintaining strong generalization performance across diverse MRI samples. Conclusions: The proposed integrated framework demonstrated strong performance in both brain tumor classification and segmentation tasks, effectively detecting small and low-contrast tumor regions while maintaining good generalization across diverse MRI samples. These findings suggest that the framework may serve as a reliable decision-support tool for automated brain tumor analysis in clinical practice. Full article

Automated brain tumor classification from MRI scans requires optimized CNN architectures deployable on embedded FPGA platforms. This paper presents an integrated approach combining the Multi-Chaotic Enhanced HEOA (MC-HEOA) for automatic CNN architecture discovery with deployment validation on a Xilinx Zynq-7000 FPGA. A CEC2023 benchmark across 10 test functions evaluates 6 chaotic maps and selects the Tent map as the optimal diversity generator. The NAS search space spans a massive combinatorial space of $1.31 \times {10}^{16}$ configurations encoding architectural choices (layers, convolutions, channels, pooling) under a strict constraint of fewer than one million parameters for FPGA compatibility. The optimal discovered architecture, trained and evaluated using single-channel grayscale input (224 × 224 × 1)—the natural representation for intrinsically monochromatic MRI data— achieves 91.33% test accuracy and 92.44% validation accuracy with 724,200 parameters on the 4-class Brain Tumor MRI dataset (glioma, meningioma, pituitary, no tumor). HLS synthesis on the Zynq-7000 (xc7z020clg484-1) validates embedded deployment feasibility, with DSP utilization of 16%, LUT utilization of 57%, FF utilization of 28%, and an inference latency of 374 ms at 100 MHz. This study demonstrates the effectiveness of MC-HEOA for discovering compact, high-performing CNN architectures compatible with FPGA deployment, opening new perspectives for real-time embedded medical diagnosis. Full article

Background/Objectives: Automated brain tumor classification from MRI is particularly challenging when restricted to single post-contrast axial T1-weighted slices without volumetric or clinical context. Methods: We present a four-class (glioma, meningioma, pituitary tumor, no tumor) slice-level classification framework that combines a fine-tuned Swin-Tiny Transformer with inverse-frequency class-weighted learning and a prototype SMT-based symbolic auditing layer for post hoc logical consistency checks. All architectures were trained and evaluated under identical preprocessing, augmentation, optimization, and evaluation protocols. Results: On an internal clinical dataset from Bandırma Onyedi Eylül University Hospital (n = 8040 slices), Swin-Tiny achieved 97.42% slice-level accuracy (macro-F1 97.42%, macro-AUC 0.994), exceeding matched convolutional baselines by approximately eight percentage points. Five-fold stratified cross-validation confirmed stability (mean accuracy 97.40% ± 0.28%). Zero-shot evaluation on the independent BRISC-2025 dataset (n = 6000 slices) yielded 94.82% accuracy and macro-AUC 0.97, indicating maintained performance under acquisition-related distribution shift. Per-class metrics were consistently high across tumor types, with residual errors dominated by glioma–meningioma confusion, reflecting known radiologic overlap on single contrast-enhanced T1 slices. The symbolic auditing layer flagged 1.2–2.9% of predictions as constraint-violating; most such cases were borderline but correctly classified, suggesting sensitivity of heuristic thresholds rather than systematic model failure. Conclusions: These findings support the value of hierarchical shifted-window attention for integrating local texture and broader spatial context in slice-level MRI classification. While patient-wise, multimodal, and prospective validation remain necessary for clinical deployment, this study provides a controlled empirical benchmark and a prototype mechanism for post hoc logical auditing in neuro-oncologic imaging. Full article

Magnetic resonance imaging (MRI) is a critical clinical tool that requires precise and reliable interpretation for effective brain tumor diagnosis and timely treatment planning. Deep learning methods have advanced automated tumor classification greatly in the last few years, but many of the current methods are still challenged by a lack of interpretability, a lack of testing on region-focused data, and a lack of model robustness testing. Such limitations reduce clinical trust and limit the practice of automated diagnostic systems. To address these challenges, this study proposes an interpretable deep learning model for classifying brain tumors using the PMRAM dataset, which is a Bangladeshi brain MRI collection containing four categories: glioma, meningioma, pituitary tumor, and normal brain.. The proposed pipeline combines image preprocessing and feature enhancement methods, and then it trains a series of squeeze-and-excitation (SE)-enhanced convolutional neural networks such as VGG19, DenseNet201, MobileNetV3-Large, InceptionV3, and EfficientNetB3. The SE-enhanced EfficientNetB3 performed best, with 98.70% accuracy, 98.77% precision, 98.70% recall, and 98.70% F1-score. Cross-validation also demonstrated stable performance, with a mean accuracy of 96.89%. The model also exhibited efficient inference with low GPU memory consumption, enabling predictions in about 2–4 s per MRI image. Grad-CAM++ and saliency maps were used to improve the transparency of the results, and it was found that the network was concentrated on the clinically significant parts of the tumor, which affected the model predictions. Further robustness analysis and cross-dataset testing are additional evidence of the generalization possibility of the model. An online application was also implemented to allow real-time prediction and visual explanation of brain tumors. Overall, the proposed framework offers a precise, interpretable, and promising solution to automated brain tumor classification using MRI images. Full article

In this comprehensive review, we explore the evolving landscape of research and clinical practices in sellar lesions, emphasizing recent advancements in histopathology and molecular biology. Distinct lesions can arise from the sellar area, predominantly comprising different tumor types, inflammatory conditions, or systemic conditions. The recent CNS5 World Health Organization classification integrates genetic modifications into histopathological characteristics, enhancing the ability to predict the biological behavior and malignant potential of these lesions. Furthermore, the molecular alterations discovered in these tumors may act as valuable diagnostic and prognostic indicators, facilitating a tailored approach, especially for those demonstrating aggressive characteristics resistant to conventional treatments. The scope of the present review is to provide a comprehensive insight into the current understanding of sellar lesions with regard to emerging prognostic factors, like molecular alterations, to advances in clinical strategies, and to identification of potential new therapeutic targets within the oncology field. Full article

Background: Difficult benign intracranial tumors (including meningiomas, schwannomas, neurofibromatosis-related tumors, and pituitary neuroendocrine tumors) have substantial morbidity in patients. Due to their limited treatment options, there is a need for individualized treatment beyond histological and surgical approaches. Objective: To summarize how novel treatment innovations have been implemented for these tumors, meningiomas and schwannomas are prioritized, followed by NF-associated neoplasms, and then pituitary neuroendocrine tumors in comparison to low-grade gliomas. Methods: We summarize the current knowledge relating to targeted therapies for gliomas, meningiomas, schwannomas, neurofibromatosis (NF) tumors, and pituitary neuroendocrine tumors to investigate an individual’s treatment options for difficult benign brain tumors. This review synthesizes evidence on tumor genomics and molecular markers, supported by methylation-based classification, immunohistochemistry, and functional assays, emphasizing current clinical applications. Evidence Synthesis: The recent data show that DNA methylation-based models can predict post-surgical outcomes and radiotherapy responses, enabling risk stratification and radiotherapy benefit prediction. Early signals support target-directed treatment, including cMET blockade that radiosensitizes NF2 schwannoma models, brigatinib-associated tumor shrinkage in NF2-deficient models, and PitNET organoid data. Conclusions: We support clinical decision-making that utilizes molecular profiling with functional testing to guide targeted treatment. We also identify evidence gaps such as biomarker-defined prospective trials that are needed for broader clinical implementation. Full article

Objective: To compare tumor growth rate between patients with pituitary microadenomas who had mild to moderate prolactin elevation and symptoms leading to initiation of cabergoline therapy, and asymptomatic microadenomas without prolactin elevation managed with observation. Materials and Methods: In this retrospective cohort study, 139 patients diagnosed with pituitary microadenoma between 2019 and 2024 and with at least 12 months of clinical and radiological follow-up were included. Patients who received cabergoline therapy due to symptoms were classified as the dopamine agonist-positive [DA(+)] group, while those who did not receive treatment were classified as the dopamine agonist-negative [DA(−)] group. Tumor growth rate was calculated as the annual change (mm/year) in maximum tumor diameter on serial magnetic resonance imaging. Between-group comparisons were performed using the Mann–Whitney U test. A mixed-effects linear model was constructed to evaluate the interaction between time and treatment. Results: Of the 139 patients included in the study, 42 were in the DA(+) group and 97 were in the DA(−) group. There were no significant differences between the groups in terms of baseline age, follow-up duration, or tumor size (p > 0.05). The mean tumor growth rate was 0.67 ± 0.80 mm/year in the DA(−) group and 0.36 ± 0.38 mm/year in the DA(+) group (p = 0.0208). In the mixed-effects model analysis, the time × treatment interaction was statistically significant (β = −0.021 mm/month; p = 0.009). Patients receiving cabergoline showed a marked reduction in prolactin levels and improvement in symptoms in 78% of cases. Importantly, no tumor shrinkage was observed in either group; the primary observed effect was a reduction in growth velocity rather than true tumor regression. No serious treatment-related adverse effects were observed. Conclusions: In patients with pituitary microadenomas, cabergoline therapy was associated with a reduced tumor growth rate over time, while no true tumor regression was observed. These findings suggest that cabergoline exposure may influence longitudinal tumor growth dynamics in clinically ambiguous cases encountered in routine practice, without implying definitive tumor subtype classification. Full article

Thyrotroph pituitary neuroendocrine tumors (PitNETs) are rare functional pituitary tumors characterized by autonomous secretion of thyroid-stimulating hormone (TSH), leading to central hyperthyroidism. Under the 2022 World Health Organization classification, these tumors are defined as PIT1-lineage PitNETs, reflecting lineage-specific differentiation and improving pathological accuracy. Clinically, thyrotroph PitNETs often present as macroadenomas with invasive growth, making complete surgical resection challenging and necessitating multimodal treatment strategies. From a molecular oncology perspective, thyrotroph PitNETs lack recurrent driver mutations and instead exhibit heterogeneous alterations involving dysregulated cell-cycle control, impaired thyroid hormone-mediated negative feedback, and aberrant growth factor signaling. Immunohistochemically, tumor cells express PIT1 and TSH and show strong membranous expression of somatostatin receptor subtype 2, providing a biological rationale for somatostatin receptor ligand -based therapy. Somatostatin receptor ligands play a central role in the management of thyrotroph PitNETs as preoperative, adjuvant, or primary treatment and achieve effective hormonal control and tumor stabilization or shrinkage in many patients. Accurate differentiation between thyrotroph PitNETs and resistance to thyroid hormone β is essential, as these entities share biochemical features but require fundamentally different management. Advances in lineage-based tumor classification, receptor profiling, and molecular pathology have refined diagnostic strategies and enabled a more personalized, tumor-oriented therapeutic approach. This review highlights current insights into the tumor biology and treatment of thyrotroph PitNETs and discusses future perspectives for receptor-targeted and molecularly informed therapies. Full article

Pituitary neuroendocrine tumors (PitNETs) constitute a significant proportion of primary intracranial neoplasms and were historically differentiated based on clinical hormone excess syndromes and tinctorial properties. The 5th edition of the WHO classification introduces a paradigm shift towards the lineage-based taxonomy based on the cell-specific expression of transcription factors (TFs). This overview focuses on the biological justifications and diagnostic value of the core TFs of Pituitary-Specific Positive Transcription Factor 1 (PIT1), T-Box Pituitary Transcription Factor (TPIT), and Steroidogenic Factor 1 (SF1), which signify the somatotroph, lactotroph, thyrotroph, corticotroph, and gonadotroph lineages, respectively. By focusing on TF expressions instead of hormone immunoreactivity, pathologists can better subtype clinically non-functioning tumors, effectively relegating the previously overutilized null cell category to about 1% of cases. The TF-based classification is also essential in discriminating high-risk histotypes of silent corticotroph tumors, sparsely granulated somatotrophs, and immature PIT1-lineage PitNETs, which are linked to a higher invasiveness and recurrence. We suggest a practical, stepwise immunohistochemical diagnostic algorithm with the integration of ancillary markers (e.g., GATA3 and ERα) to refine lineage assignment. New molecular correlates such as GNAS and USP8 mutations also add to this framework and guide the use of individualized treatment involving somatostatin analogs or dopamine agonists. And lastly, we discuss the ongoing issues of diagnosis of triple-negative and multilineage tumors and the growing importance of DNA methylation profiling and artificial intelligence in standardized reporting and improving precision management. Full article

The accurate detection and classification of brain tumors from magnetic resonance imaging (MRI) are critical for diagnosis and treatment planning. While deep learning has shown remarkable success in this domain, many state-of-the-art models rely on complex, end-to-end convolutional neural networks (CNNs) that require extensive computational resources and large, annotated datasets for training. This work proposes a novel and efficient methodology that, for the first time, leverages self-supervised DINO vision transformer backbones (DINO v1, DINOv2, and DINOv3) on a large corpus of natural images as powerful feature extractors for brain tumor analysis. We utilize the rich, general-purpose features from DINO-family backbones without fine-tuning the core model. These extracted features are then fed into a simpler, task-specific classifier (such as a support vector machine or a multi-layer perceptron) for the final detection and multi-class classification (e.g., glioma, meningioma, and pituitary tumor). Our methodology is evaluated on two benchmark medical imaging datasets with various classifying granularities. The results demonstrate that the proposed method achieves competitive and, in some cases, superior classification accuracy compared to representative fine-tuned convolutional neural networks and attention-based architectures, while significantly reducing the number of trainable parameters and training time. In particular, the best configuration achieves up to 98.17% accuracy and an F1-score of 98.18% on the 15-class dataset and 99.08% accuracy and an F1-score of 99.02% on the 4-class dataset. This study confirms the exceptional transfer learning capabilities of self-supervised vision transformers like DINO in the medical imaging domain, establishing it as a highly effective and efficient backbone for robust brain tumor detection and classification systems. Full article