Predicting Alzheimer's Disease Stage Transformations 12 Months in Advance using 3D Convolutional LSTM based on 3D Magnetic Resonance Images

Date

2024-1-25

Author

Erdemir, Elifnur

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

268
views

85
downloads

Cite This

This study evaluates the ability to predict the transition from the healthy cognitive stage to the Alzheimer's stage using a 3D LSTM model. The model was trained on a data set created from MR images taken at different times in the ADNI database and its performance was evaluated using four consecutive 3D MR images. The results reveal that while the model highlights good specificity in recognizing healthy individuals, it shows low sensitivity and F1 score in predicting the transition from MCI to Alzheimer's. While the model can identify individuals remaining in the MCI stage with high sensitivity, it shows a decrease in precision and F1 score, indicating that there are false positives in this class. All in all, he study highlights the potential of the 3D LSTM model for early diagnosis of Alzheimer's disease, shedding light on future research in this field.

Subject Keywords

Alzheimer's disease, Mild cognitive impairment, 3D Conv LSTM, ADNI Database, MR imaging

URI

https://hdl.handle.net/11511/108252

Collections

Graduate School of Informatics, Term Project