Answers are grounded in this paper and cite their sources.
Hybrid model improves brain tumor detection accuracy in MRI images, suggesting advanced algorithms enhance diagnosis.
Brain tumours (BT) are considered one of the most aggressive and common diseases, leading to a short life expectancy. Therefore, timely and prompt treatment planning is the primary phase to progress the patient's quality of life. Usually, numerous image methods are employed to assess the cancer area in a brain. However, magnetic resonance imaging (MRI) is generally utilized owing to its higher image quality and trust in no ionizing radioactivity. The BT detection task by MRI was typically expensive and time-consuming for the specialists. So, computer-aided diagnosis (CAD)-based machine learning (ML) and deep learning (DL) are mainly progressed to perceive BTs early in less time without human involvement. This study proposes an Enhanced Detection of Brain Tumor Using Hybrid Model with Sparrow Search Algorithm (EDBT-HDLMSSA) approach. The proposed EDBT-HDLMSSA approach relies on improving the brain cancer recognition and classification model in biomedical images using advanced techniques. Initially, the image pre-processing applies a Sobel filter (SF) to eliminate the noise and skull removal using U-Net segmentation. Furthermore, the ShuffleNetv2.3 method is employed for feature extraction. The hybrid autoencoder and long short-term memory (AE-LSTM) method is implemented for the BT classification process. Finally, the hyperparameter selection of the AE-LSTM method is accomplished by utilizing the sparrow search algorithm (SSA) model. A wide range of experiments using the EDBT-HDLMSSA technique are achieved under the BT MRI dataset. The performance validation of the EDBT-HDLMSSA technique portrayed a superior accuracy value of 98.57% over existing models.
No takes yet. Share an insight, caveat, or question.
Chanemougavel et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: