A Hybrid-Based Ensemble Learning Model and DNN for Automated Diagnosis of Brain Tumor from MRI Images

Jehangir, Basra; Nayak, Soumya Ranjan

doi:10.1007/978-981-16-7076-3_27

Cited by 3 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With regard to mammography image analysis and various anticipated AI-based models, such as machine learning (ML), deep learning (DL), transfer learning (TL), and extreme learning machine (ELM), this study attempts to give a technical reviewsummary of current research progress. The paper outlines the features, benefits, and drawbacks of these models as well as how they have been used to produce more precise breast cancer diagnosis with less B.Jehangir et al [4] As the tenth most prevalent type of tumour affecting people of all ages, brain tumours are a major cause of death globally. While brain tumours are largely curable when discovered early, it is not always possible to do a biopsy, which is usually done prior to definitive brain surgery.…”

Section: Literaturereviewmentioning

confidence: 99%

A Survey on Lung Cancer Detection Using Ensemble Neural Networks

2024

International Journal of Science and Engineering Applications

View full text Add to dashboard Cite

Lung cancer stands as the primary cause of cancer-related deaths globally, marked by a notable fatality rate owing to delayed diagnosis. Timely detection proves pivotal for efficacious treatment and enhanced patient prognosis. Within this study, we introduce an innovative methodology for identifying lung cancer, employing a fusion of 3D Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Radial Basis Function (RBF). This proposed method harnesses 3D CT scans to extract characteristics from lung nodules and categorize them as either malignant or benign. Initially, the 3D CNN is employed for feature extraction, capturing the spatial correlations and nodule patterns. Subsequently, the CNN's output is inputted into the SVM and RBF classifiers for further classification. Trained on the extracted features, the SVM and RBF classifiers undertake the final categorization of lung nodules. Experiments using a publicly available dataset of lung CT scans were carried out in order to evaluate the effectiveness of the suggested strategy. Our results demonstrate that the suggested strategy outperforms other modern approaches in achieving high accuracy in the identification of lung cancer. In particular, the suggested approach yields 96.2% accuracy, 96.6% sensitivity, 95.8% specificity, and 0.988 AUC. Combining 3DCNN, SVM, and RBF may increase lung cancer detection's accuracy and effectiveness, leading to earlier diagnosis and better patient outcomes. Additionally, there is a chance that this suggested approach will find use in other medical imaging fields, like the detection of prostate and breast cancer. To summarize, the proposed methodology signifies a noteworthy stride towards enhancing the precision and effectiveness of lung cancer detection, thereby promising a substantial impact on patient care and outcomes.

show abstract

Section: Literaturereviewmentioning

confidence: 99%