Brain Tumor Segmentation and Classification from Sensor-Based Portable Microwave Brain Imaging System Using Lightweight Deep Learning Models

Abstract

Automated brain tumor segmentation from reconstructed microwave (RMW) brain images and image classification is essential for the investigation and monitoring of the progression of brain disease. The manual detection, classification, and segmentation of tumors are extremely time-consuming but crucial tasks due to the tumor’s pattern. In this paper, we propose a new lightweight segmentation model called MicrowaveSegNet (MSegNet), which segments the brain tumor, and a new classifier called the BrainImageNet (BINet) model to classify the RMW images. Initially, three hundred (300) RMW brain image samples were obtained from our sensors-based microwave brain imaging (SMBI) system to create an original dataset. Then, image preprocessing and augmentation techniques were applied to make 6000 training images per fold for a 5-fold cross-validation. Later, the MSegNet and BINet were compared to state-of-the-art segmentation and classification models to verify their performance. The MSegNet has achieved an Intersection-over-Union (IoU) and Dice score of 86.92% and 93.10%, respectively, for tumor segmentation. The BINet has achieved an accuracy, precision, recall, F1-score, and specificity of 89.33%, 88.74%, 88.67%, 88.61%, and 94.33%, respectively, for three-class classification using raw RMW images, whereas it achieved 98.33%, 98.35%, 98.33%, 98.33%, and 99.17%, respectively, for segmented RMW images. Therefore, the proposed cascaded model can be used in the SMBI system.

Advancing Healthcare: Synergizing Biosensors and Machine Learning for Early Cancer Diagnosis

“…In another study, Hossain et al [ 97 ] employed a sensor-based portable microwave brain imaging (SMBI) system to obtain the reconstructed microwave (RMW) brain images. The proposed method consists of a segmentation model called MicrowaveSegNet (MsegNet) and a classifier called BrainImageNet (BINet).…”

A Lightweight Attention based MobileNetv2 Model for Brain Tumor Segmentation and Severity of Tumor Classification using Support Vector Machine

“…Both the BRATS 2018 with Figshare datasets are used to measure performance, with the BRATS dataset achieving the highest levels of accuracy (0.9210), sensitivity (0.9313), and specificity (0.9284). The BrainImageNet (BINet) model is offered as a classifier for RMW pictures, and the MicrowaveSegNet (MSegNet) model is proposed as a lightweight segmentation model that segments the brain tumor [31]. Authors sensors-based microwaves brain scan (SMBI) system first collected 300 RMW brain image samples to use as the basis for a new dataset.…”

Microwave Imaging System Based on Signal Analysis in a Planar Environment for Detection of Abdominal Aortic Aneurysms

“…This is a non-ionizing technology that utilizes a very low emission power—more than 100 times lower than the power emitted by current mobile phones and roughly 100 times lower than the limits imposed by the SAR (Specific Absorption Rate) regulations, which all the electromagnetic emissions in biological contexts must comply with [ 15 , 16 , 17 ]—thus ensuring the full safety of these systems and the consequent possibility of using them repeatedly [ 18 ]. Currently, we can find several MWI devices and prototypes for different medical applications, including breast cancer detection [ 19 , 20 , 21 ], dynamic image building for cardiovascular systems [ 22 ], brain tumor classification [ 23 ], fast image building after cerebrovascular accident [ 24 , 25 ], or brain-shift detection during brain tumor surgery [ 18 ].…”