Search

A recent study conducted at the Central University of Haryana in India has proposed a privacy-preserving collaborative model for the classification of brain tumors using magnetic resonance imaging (MRI) images. The study utilized deep learning and transfer learning algorithms to analyze an open-source dataset of MRI images, which included four types of tumors and no tumor. The researchers found that federated deep learning models, involving multiple clients, outperformed conventional pretrained models in terms of accuracy. The proposed framework offers a solution for early diagnosis of brain tumors while ensuring data privacy for edge devices with limited resources. [Extracted from the article]

A new report from researchers at the Department of Management in Dehradun, India, discusses the use of computer-aided detection (CAD) in classifying brain diseases, with a focus on Alzheimer's Disease (AD) and brain tumors. The study highlights the importance of expertise in interpreting MRI, PET, and CT scans for accurate diagnosis. The researchers propose pre-processing methods that improve the classification performance of MRI images and reduce training time for learning algorithms. The study concludes that their proposed algorithms outperformed existing methods in terms of accuracy and sensitivity. For more information, the full article can be accessed through the provided link. [Extracted from the article]

A study conducted by researchers at VIT-AP University in Andhra Pradesh, India, has developed a deep learning algorithm called RU-Net2+ for accurate brain tumor segmentation and survival rate prediction. The algorithm was applied to a dataset of brain tumor images and achieved impressive results, surpassing current benchmarks in classification accuracy, tumor segmentation precision, and survival rate prediction. The framework shows promise for automating brain tumor diagnosis and enhancing patient care, providing valuable insights for medical professionals making treatment decisions. The study was published in IEEE Access. [Extracted from the article]