An MRI brain tumour detection using logistic regression-based machine learning model

The MRI brain diagnosis process is much complex with earlier procedures, the shape and location of brain tumour approximation are not practicable with existing models. Brain-related anomalies are sometimes caused death, according to WHO every year around 10 billion people are exaggerated with brain-related diseases furthermore gradients to death. Therefore, a modern brain disease or tumour detection application design is compulsory. In this work, automatic brain abnormality detection can be performed through the logistic regression machine learning technique. The MRI brain images are collected for training and testing, many researchers have built several detection methods for brain tumours; however, they are outdated and limited in operation. To crossover the limitations of the earlier models there is machine intelligence system has been needed. Real-time MRI images are acquired for testing to assess various orientations and age groups. The disease classification has been processed through logistic regression and threshold segmentation . The ADNI-1 and ADNI-2 datasets are used to train the TSLR model, as well as real-time MRI brain samples, are chosen for testing. Finally, the performance measures like accuracy 97%, precision 97.9% and recall 97% had been attained, which are compared with existing models and noticed the outperforms the methodology.