What question did this study set out to answer?

The aim is to enhance the pre-diagnosis and prediction of brain tumors using AI and deep learning segmentation methods.

February 9, 20260 citationsOpen Access

Enhancing Brain Anomaly Pre-Diagnosis with Traditional Artificial Intelligence and Popular Deep Learning models: A Comparative Analysis

Key Points

The aim is to enhance the pre-diagnosis and prediction of brain tumors using AI and deep learning segmentation methods.
Applied traditional AI (SVM) and deep learning models (ResNet50, InceptionV3)
Collected and pre-processed MRI and CT images from various sources
Conducted accuracy assessments using multiple metrics
Traditional SVM achieved 77% accuracy in diagnosis
Deep learning models (ResNet50, InceptionV3) reached approximately 97% accuracy
InceptionV3 demonstrated exceptional efficiency with data augmentation and transfer learning strategies

Abstract

Abstract- Thisstudymainlyaimspre-diagnosis and prediction of specific brain tumor by applying traditional and popular segmentation methods with deep learning models and also investigates the comparative performance between Artificial Intelligence and Deep Learning methods and models. The diagnostic methods used currently are generally subjective, timeconsuming and requires highly specialized knowledge in detail. To determine and overcome these limitations, we propose the well-developed of two deep learning segmentation methods capable of accurately and efficiently analysing the brain tumor Magnetic Resonance Imaging (MRI) and CT (Computerized Tomography) radiological imaging data. These models were Support Vector Machine (SVM) for traditional AI model and ResNet50 and InceptionV3 for popular DL models architectures and these were used for diagnosing specific important brain conditions, including ischemic stroke, low-grade glioma (LGG), and normal (tumor-free) cases. In addition, in medical area, ischemic stroke and LGG images could not be well determined and misdiagnosing procedure could occur. Because of these reasons, with using these deep learning models, the problems and limitations were overcome. The initial phase involved the meticulous collection and pre-processing of a large open source/public dataset of MRI and CT images, carefully distinguishing those from ischemic stroke and LGG patients and healthy individuals. The models underwent rigorous training using the pre-processed image dataset and were assessed using various accuracy metrics. While traditional methods utilizing Support Vector Machines (SVM) achieved an accuracy of 77%, deep learning architectures exhibited significant advancements, with ResNet50 and InceptionV3 achieving accuracies of approximately 97%. The InceptionV3 model's lightweight architecture, integrated with effective data augmentation and transfer learning strategies, demonstrated exceptional diagnostic efficiency and accuracy. These results underscore the immense potential of deep learning in revolutionizing brain tumor/lesion diagnosis

KI fragen

Bookmark

View Full Paper