article · Scientific Reports
Glioma classification and grading from histopathological whole slide images are vital for prognosis and treatment planning. A hybrid deep learning system combines YOLOv5 and ResNet50 architectures to localise tumours and extract complex features while stabilising training dynamics against gradient explosion. Tumour grades are subsequently predicted using an extreme gradient boosting classifier designed to handle high-dimensional, non-linear relationships in pathology images. The framework aims to assist clinicians by visualising tumour features and heterogeneity. Evaluated on The Cancer Genome Atlas dataset, the system achieved 97.2 per cent accuracy, 97.8 per cent precision, 98.6 per cent sensitivity, and a 97 per cent Dice similarity coefficient across four glioma grades. It demonstrates particular effectiveness in differentiating between lower-grade glioma subtypes II and III, exceeding the performance of standard approaches.
Accurate grading of brain tumours directly affects clinical prognosis and treatment choices for patients. Interpreting complex pathology slides manually can be challenging, but automated computational tools can support medical practitioners by accurately highlighting tumour regions and distinguishing difficult cancer subtypes, such as differentiating grade II from grade III gliomas, leading to more reliable diagnostic assessments.
This framework could enable computer-aided diagnostic tools for pathologists and clinical teams analysing digital histology slides. By pinpointing tumour boundaries and classifying glioma grades automatically, the software could integrate into digital pathology platforms. Currently at the applied and tested research stage, the model has demonstrated strong performance on retrospective benchmark data but requires further validation in clinical workflows.
AI-generated from the published abstract. Always read the original work before citing.
Gliomas are primary brain tumors caused by glial cells. These cancers' classification and grading are crucial for prognosis and treatment planning. Deep learning (DL) can potentially improve the digital pathology investigation of brain tumors. In this paper, we developed a technique for visualizing a predictive tumor grading model on histopathology pictures to help guide doctors by emphasizing characteristics and heterogeneity in forecasts. The proposed technique is a hybrid model based on YOLOv5 and ResNet50. The function of YOLOv5 is to localize and classify the tumor in large histopathological whole slide images (WSIs). The suggested technique incorporates ResNet into the feature extraction of the YOLOv5 framework, and the detection results show that our hybrid network is effective for identifying brain tumors from histopathological images. Next, we estimate the glioma grades using the extreme gradient boosting classifier. The high-dimensional characteristics and nonlinear interactions present in histopathology images are well-handled by this classifier. DL techniques have been used in previous computer-aided diagnosis systems for brain tumor diagnosis. However, by combining the YOLOv5 and ResNet50 architectures into a hybrid model specifically designed for accurate tumor localization and predictive grading within histopathological WSIs, our study presents a new approach that advances the field. By utilizing the advantages of both models, this creative integration goes beyond traditional techniques to produce improved tumor localization accuracy and thorough feature extraction. Additionally, our method ensures stable training dynamics and strong model performance by integrating ResNet50 into the YOLOv5 framework, addressing concerns about gradient explosion. The proposed technique is tested using the cancer genome atlas dataset. During the experiments, our model outperforms the other standard ways on the same dataset. Our results indicate that the proposed hybrid model substantially impacts tumor subtype discrimination between low-grade glioma (LGG) II and LGG III. With 97.2% of accuracy, 97.8% of precision, 98.6% of sensitivity, and the Dice similarity coefficient of 97%, the proposed model performs well in classifying four grades. These results outperform current approaches for identifying LGG from high-grade glioma and provide competitive performance in classifying four categories of glioma in the literature.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1038/s41598-024-54864-6
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
