MATERIALS AND METHODS: 18F-FDG PET/CT images of 14 healthy control (HC) subjects (MoCA score > 26 (mean+SD~ 26.93+0.92) with no clinical evidence of cognitive deficits or neurological disease) and 16 AD patients (MoCA ≤22 (mean+SD~18.6+9.28)) were pre-processed in SPM12 while using our developed Malaysian healthy control brain template. The AD patients were assessed for disease severity using ADAS-Cog neuropsychological test. KNE96 template was used for registration-induced deformation in comparison with the ICBM templates. All deformation fields were corrected using the Malaysian healthy control template. The images were then nonlinearly modified by DARTEL to segment grey matter (GM), white matter (WM) and cerebrospinal fluid (CSF) to produce group-specific templates. Age, intracranial volume, MoCA score, and ADASCog score were used as variables in two sample t test between groups. The inference of our brain analysis was based on a corrected threshold of p<0.001 using Z-score threshold of 2.0, with a positive value above it as hypometabolic. The relationship between regional atrophy in GM and WM atrophy were analysed by comparing the means of cortical thinning between normal control and three AD stages in 15 clusters of ROI based on Z-score less than 2.0 as atrophied.

RESULTS: One-way ANOVA indicated that the means were equal for TIV, F(2,11) = 1.310, p=0.309, GMV, F(2,11) = 0.923, p=0.426, WMV, F(2,11) = 0.158, p=0.856 and CSF, F(2,11) = 1.495 p=0.266. Pearson correlations of GM, WM and CSF volume between HC and AD groups indicated the presence of brain atrophy in GM (p=-0.610, p<0.0001), WM (p=-0.178, p=0.034) and TIV (p=-0.374, p=0.042) but showed increased CSF volume (p=0.602, p<0.0001). Voxels analysis of the 18FFDG PET template revealed that GM atrophy differs significantly between healthy control and AD (p<0.0001). Zscore comparisons in the region of GM & WM were shown to distinguish AD patients from healthy controls at the prefrontal cortex and parahippocampal gyrus. The atrophy rate within each ROI is significantly different between groups (c2=35.9021, df=3, p<0.0001), Wilcoxon method test showed statistically significant differences were observed between Moderate vs. Mild AD (p<0.0001), Moderate AD vs. healthy control (p=0.0005), Mild AD vs. HC (p=0.0372) and Severe AD vs. Moderate AD (p<0.0001). The highest atrophy rate within each ROI between the median values ranked as follows severe AD vs. HC (p<0.0001) > mild AD vs. HC (p=0.0091) > severe AD vs. moderate AD (p=0.0143).

MATERIALS AND METHODS: A literature search was carried out to gather eligible studies from the following widely sourced electronic databases such as Scopus, PubMed and Google Scholar using the combination of the following keywords: AD, MRS, brain metabolites, deep learning (DL), machine learning (ML) and artificial intelligence (AI); having the aim of taking the readers through the advancements in the usage of MRS analysis and related AI applications for the detection of AD.

RESULTS: We elaborate on the MRS data acquisition, processing, analysis, and interpretation techniques. Recommendation is made for MRS parameters that can obtain the best quality spectrum for fingerprinting the brain metabolomics composition in AD. Furthermore, we summarise ML and DL techniques that have been utilised to estimate the uncertainty in the machine-predicted metabolite content, as well as streamline the process of displaying results of metabolites derangement that occurs as part of ageing.

CASE REPORT: The patient presented with upper respiratory tract infection, fever, seizures and abdominal pain. An initial diagnosis of encephalitis was made. In view of the unexplained abdominal pain with other clinical findings such as posterior reversible encephalopathy syndrome by CT brain, temporary blindness as well as hyponatraemia, acute intermittent porphyria was suspected. Urine delta aminolaevulinic acid (δ-ALA) and porphobilinogen were elevated confirming the diagnosis of AIP. Genetic studies were done for this patient. The patient had a complete resolution of her symptoms with carbohydrate loading and high caloric diet.