Kamagata et al. 2022) explored the difference in the temporal progression patterns of brain atrophy between Corticobasal degeneration (CBD) and progressive supranuclear palsy with Richardson's syndrome (PSP-RS) patients using anatomical MRI data coupled with a novel unsupervised machine learning technique called SuStaIn. The proposed SuSta technique was applied to the brain features of all subjects to discover the disease subtypes and trajectories of CBS and PSP-RS. Using this technique, the authors discovered different temporal progression patterns of brain atrophy between CBS and PSP-RS groups which was able to distinguish two groups with an accuracy of 87.5%, indicating that SuStaIn can be viewed as a reliable method for separating PSP-RS patients from CBS patients and predicting treatment options among them.In another article, entitled "Application of High-Frequency Conductivity Map Using MRI to Evaluate It in the Brain of Alzheimer's Disease Patients" the authors aimed to explore the differences among three participant groups of cognitively normal (CN) elderly people, amnestic mild cognitive impairment (MCI) and AD in terms of a high-frequency conductivity (HFC) mapping technique extracted from a 3T MRI system. They also investigated the association between HFC values and cognitive decline as well as the utility of the HFC technique for classifying AD patients from CN (Park, Jung et al. 2022). They reported a higher HFC value in the AD group in comparison to the CN and MCI groups. Besides, they stated a positive association between HFC values and age as well as between HFC values and disease severity in AD. Using the HFC value in the insula, the authors achieved to an AUC of 0.902 for distinguishing AD patients from CN which was higher than gray matter volume (GMV) data in the hippocampus (AUC = 0.880). The authors pointed out that this classification between AD patients from CN can be improved by concatenating the insular HFC value with hippocampal GMV to an AUC of AUC = 0.928.In a paper titled "Identifying Depression in Parkinson's Disease by Using Combined Diffusion Tensor Imaging and Support Vector Machine" (Yang, Pan et al. 2022) used diffusion tensor imaging (DTI) with a support vector machine algorithm to identify depressed PD patients. To this end, they extracted fractional anisotropy (FA) and mean diffusivity (MD) among depressed PD patients, non-depressed PD, and healthy control subjects. In the analysis stage, they stated an increased MD signal in the left superior longitudinal fasciculus-temporal and right cingulum (cingulate gyrus) as well as decreased FA signal in the right cingulum (cingulate gyrus), bilateral inferior longitudinal fasciculus, bilateral corticospinal tract, left cingulum hippocampus, and bilateral superior longitudinal fasciculus among depressed PD patients compared to non-depressed PD patients. The Support Vector Machine algorithm showed an accuracy of 0.73 for distinguishing depressed PD patients from non-depressed PD patients with DTI data. They concluded that depression in PD is linked to changes in the white matter's microstructure and machine learning models with DTI data could be valuable for detecting depression in PD.In the last paper in this special issue entitled "Automated Evaluation of Conventional Clock-Drawing Test Using Deep Neural Network: Potential as a Mass Screening Tool to Detect Individuals With Cognitive Decline", (Sato, Niimi et al. 2022) aimed to develop an automated cognitive decline assessment based on Clock-Drawing Test (CDT) data and deep neural network model. This study was conducted on a large set of CDT images (N= 40,000) from the National Health and Aging Trends Study (NHATS) database. The authors reported a good balance accuracy of 90% for identifying elderly people with cognitive decline from those without, indicating that the proposed method can be used as a robust screening tool for identifying dementia or signs of impaired executive function among that group of subjects.In conclusion, these five accepted papers included in this Research Topic provided unique perspectives on addressing clinical questions using advanced artificial intelligence methods in the area of neurodegeneration
All authors listed invited authors to participate in the collection and wrote the editorial letter.
This work was supported by the Intramural Research Grant (3-10) for Neurological and Psychiatric Disorders of the National Center of Neurology and Psychiatry (N.M).
As editors of this Research Topic, it was our great pleasure to review a wide range of fascinating articles and reviews them. We sincerely appreciate the contributions of all the authors who submitted their studies for consideration for this Research Topic and the reviewers who assessed the articles.

