Accuracy Improvement of Detecting the Effect of Alcohol Consumption on Brain Structure in MRI Images using Deep Learning Methods

This paper proposes a deep convolution neural network model for detecting the effects of alcohol consumption in brain MRI images. The main objective of the research is to simulate and identify the structural and functional changes of the brain caused by alcohol consumption. In this research, MRI images were first pre-processed using SPM software to prepare for model training. Then, a CNN model was designed and trained to detect the effects of alcohol consumption. Experiments showed that by increasing the number of IPAKs, setting the learning rate to 0.001, and using a convolution window size of 3×3, the best performance was achieved. The experimental results indicate a precision of 95.85%, a recall rate of 95%, and an F1 criterion of 96%, which indicates the excellent performance of the model. This research can be used as an effective tool for automatically detecting the effects of alcohol consumption in MRI images in medical and clinical research and can help increase the accuracy and speed of diagnosis in clinical processes. The results could also be useful in studies related to brain disorders caused by drug and alcohol use.