.png){kind=icon}
.png){kind=icon}
Return
An Innovative Algorithm-Assisted Neuroimaging Technique for Calculating Brain Age
.png){kind=icon}
PRE
AI
V
DOI：10.17576/jsm-2026-5503-18.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Brain scans and machine learning algorithms can now be used to determine a person's age. In this assessment, we discuss a brief summary of the multiple medicinal purposes of brain-age estimation in neuropsychiatry and general populations. This verified technique has created new opportunities for resolving clinical concerns in neurology. For the purpose of developing a framework for brain-age projection, we first give an overview of common neuroimaging modalities, feature extraction techniques, and machine learning models. In this study, we proposed a novel wild horse optimized multi-tiered convolutional neural network (WHO-MCNN) strategy for predicting brain age. We employed magnetic resonance imaging (MRI) to collect brain neuroimage data for this study. To retain edges and reduce noise in images, pre-processed data was exposed to a bilateral filter. The histogram of oriented gradients (HOG) was used to extract the features from the data to record shape and texture information that is valuable for object recognition. The proposed method is further compared to other machine learning algorithms. The results show the proposed method achieved better performance in MAE, RMSE, and R2, such as 2.982, 3.925, and 0.537 for brain age prediction. Through early identification and treatment of age-related neurological diseases, this approach facilitates a greater understanding of brain aging processes. Finally, we offer some recommendations for future study approaches and talk about the real-world issues and difficulties that have been discussed in the literature.
Keywords:
Brain age
magnetic resonance imaging (MRI)
neuroimage
wild horse optimized multi-tiered convolutional neural network (WHO-MCNN)
Journal
IF:
0.8
Papers:
113
Citations:
2.9K
Organization
Cited Papers