Share：
Share this content in WeChat
X
[Chinese][PDF]1781
• Clinical Article •
The value of hippocampal MRI-based radiomics modelling for predicting cognitive dysfunction in patients with type 2 diabetes mellitus
YIN Lei  XU Zhigao  CAO Milan  WANG Qiang 

Cite this article as: YIN L, XU Z G, CAO M L, et al. The value of hippocampal MRI-based radiomics modelling for predicting cognitive dysfunction in patients with type 2 diabetes mellitus[J]. Chin J Magn Reson Imaging, 2025, 16(5): 80-87. DOI:10.12015/issn.1674-8034.2025.05.013.


[Abstract] Objective To develop a predictive model based on hippocampal MRI radiomics to assess whether the level of cognitive function in type 2 diabetes mellitus (T2DM) patients belongs to the category of cognitively normal (CN), mildly cognitively impaired (MCI) or dementia (Dem).Materials and Methods Clinical data and MRI imaging data of 140 T2DM patients were retrospectively collected, and they were classified into CN group, MCI group and Dem group according to the Montreal Cognitive Assessment Beijing version (MoCA-B) score, and randomly assigned to the training set (n = 98) and the test set (n = 42) according to the ratio of 7∶3 in order to validate the performance of the model. The right and left hippocampus regions of interests (ROIs) were outlined using the uAI Research Portal (uRP), the radiomics features were extracted, and the MRI radiomics features were used to construct a machine learning (ML) model using twelve classifiers, and the confusion matrix to evaluate the classification model performance. The optimal cutoffs and tuning parameters are explored in the training data, and the models are further evaluated in the experimental data. the optimal algorithm is determined by comparing the area under the curve (AUC) of each classifier.Results From the original 2313 omics features of hippocampal MRI, ten key features were selected using the K-best selection method. Subsequently, the SelectKBest algorithm was applied to identify two optimal features. When twelve classifiers were employed for training in the CN, MCI, and Dem groups, the quadratic discriminant analysis (QDA) algorithm demonstrated the best performance among the classifiers. The AUC values for each group in the training set were 0.869, 0.854, and 0.893, respectively, while the AUCs for each group in the validation set were 0.819, 0.779, and 0.811, respectively.Conclusions The MRI-based QDA model demonstrates significant potential in predicting cognitive dysfunction among patients with T2DM. When compared to various algorithms within CN group, MCI group, and Dem group, the QDA algorithm exhibits superior performance.
[Keywords] type 2 diabetes mellitus；cognitive dysfunction；magnetic resonance imaging；radiomics；hippocampus；prediction

YIN Lei1, 2   XU Zhigao1*   CAO Milan1   WANG Qiang3  

1 Department of Radiology, the Third People's Hospital of Datong, Datong 037008, China

2 The First Clinical College of Changzhi Medical College, Changzhi 046000, China

3 School of Computer and Network Engineering, Shanxi Datong University, Datong 037009, China

Corresponding author: XU Z G, E-mail: 18636281196@126.com

Conflicts of interest   None.

Received  2025-02-06
Accepted  2025-05-10
DOI: 10.12015/issn.1674-8034.2025.05.013
Cite this article as: YIN L, XU Z G, CAO M L, et al. The value of hippocampal MRI-based radiomics modelling for predicting cognitive dysfunction in patients with type 2 diabetes mellitus[J]. Chin J Magn Reson Imaging, 2025, 16(5): 80-87. DOI:10.12015/issn.1674-8034.2025.05.013.

[1]
JIA L F, DU Y F, CHU L, et al. Prevalence, risk factors, and management of dementia and mild cognitive impairment in adults aged 60 years or older in China: a cross-sectional study[J/OL]. Lancet Public Health, 2020, 5(12): e661-e671 [2025-02-06]. https://www.thelancet.com/journals/lanpub/article/PIIS2468-2667(20)30185-7/fulltext. DOI: 10.1016/S2468-2667(20)30185-7.
[2]
BIESSELS G J, DESPA F. Cognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications[J]. Nat Rev Endocrinol, 2018, 14(10): 591-604. DOI: 10.1038/s41574-018-0048-7.
[3]
LONG X X, YUAN M Q, FANG Y. Different perspectives on the subtypes of mild cognitive impairment and their influencing factors[J]. Chinese Journal of Geriatrics, 2024, 43(1): 13-17. DOI: 10.3760/cma.j.issn.0254-9026.2024.01.003.
[4]
LEI H, HU R, LUO G H, et al. Altered Structural and Functional MRI Connectivity in Type 2 Diabetes Mellitus Related Cognitive Impairment: A Review[J/OL]. Front Hum Neurosci, 2022, 15: 755017 [2025-02-06]. https://pubmed.ncbi.nlm.nih.gov/35069149/. DOI: 10.3389/fnhum.2021.755017.
[5]
BAO H, LIU Y M, ZHANG M G, et al. Increased β-site APP cleaving enzyme 1-mediated insulin receptor cleavage in type 2 diabetes mellitus with cognitive impairment[J]. Alzheimers Dement, 2021, 17(7): 1097-1108. DOI: 10.1002/alz.12276.
[6]
Chinese Society of Endocrinology, Chinese Adult Type 2 Diabetes Hypertension Treatment Target Research Working Group. Chinese Expert Consensus on Prevention and Treatment of Cognitive Dysfunction in Patients with Type 2 Diabetes Mellitus[J]. Chin J Endocrinol Metab, 2022, 38(6): 453-464. DOI: 10.3760/cma.j.cn311282-20220518-00320.
[7]
XUE C Y, GAO T, MAO E, et al. Hippocampus Insulin Receptors Regulate Episodic and Spatial Memory Through Excitatory/Inhibitory Balance[J/OL]. ASN Neuro, 2023, 15: 17590914231206657 [2025-02-06]. https://doi.org/10.1177/17590914231206657. DOI: 10.1177/17590914231206657.
[8]
Akhtar A, Sah S P. Insulin signaling pathway and related molecules: Role in neurodegeneration and Alzheimer's disease[J/OL]. Neurochem Int, 2020, 135: 104707 [2025-02-06]. https://doi.org/10.1016/j.neuint.2020.104707. DOI: 10.1016/j.neuint.2020.104707.
[9]
JANGRA V, TOPLE J. Can Alzheimer's Disease Be Secondary to Type-2 Diabetes Mellitus?[J/OL]. Cureus, 2022, 14(11): e31273 [2025-02-06]. https://doi.org/10.7759/cureus.31273. DOI: 10.7759/cureus.31273.
[10]
WU J J, FANG Y Q, TAN X, et al. Detecting type 2 diabetes mellitus cognitive impairment using whole-brain functional connectivity[J/OL]. Sci Rep, 2023, 13(1): 3940 [2025-02-06]. https://doi.org/10.1038/s41598-023-28163-5. DOI: 10.1038/s41598-023-28163-5.
[11]
PAN P, ZHANG D, LI J, et al. The enlarged perivascular spaces in the hippocampus is associated with memory function in patients with type 2 diabetes mellitus[J/OL]. Sci Rep, 2025, 15(1): 3644 [2025-02-06]. https://doi.org/10.1038/s41598-025-87841-8. DOI: 10.1038/s41598-025-87841-8.
[12]
WANG L Y, FENG Q, GE X H, et al. Textural features reflecting local activity of the hippocampus improve the diagnosis of Alzheimer's disease and amnestic mild cognitive impairment: A radiomics study based on functional magnetic resonance imaging[J/OL]. Front Neurosci, 2022, 16: 970245 [2025-02-06]. https://doi.org/10.3389/fnins.2022.970245. DOI: 10.3389/fnins.2022.970245.
[13]
MAYERHOFER M E, MATERKA A, LANGS G, et al. Introduction to Radiomics[J]. J Nucl Med, 2020, 61(4): 488-495. DOI: 10.2967/jnumed.118.222893.
[14]
MAJUMDER S, KATZ S, KONTOS D, et al. State of the art: radiomics and radiomics-related artificial intelligence on the road to clinical translation[J/OL]. BJR Open, 2023, 6(1): tzad004 [2025-02-06]. https://doi.org/10.1093/bjro/tzad004. DOI: 10.1093/bjro/tzad004.
[15]
AREVALO-RODRIGUEZ I, SMAILAGIC N, ROQUÉ-FIGULS M, et al. Mini-Mental State Examination (MMSE) for the early detection of dementia in people with mild cognitive impairment (MCI)[J/OL]. Cochrane Database Syst Rev, 2021, 7(7): CD010783 [2025-02-06]. https://doi.org/10.1002/14651858.CD010783.pub3. DOI: 10.1002/14651858.CD010783.pub3.
[16]
MILLER J B, KAUWE J S K. Predicting Clinical Dementia Rating Using Blood RNA Levels[J/OL]. Genes (Basel), 2020, 11(6): 706 [2025-02-06]. https://doi.org/10.3390/genes11060706. DOI: 10.3390/genes11060706.
[17]
SON, C, PARK, J H. Ecological Effects of VR-Based Cognitive Training on ADL and IADL in MCI and AD patients: A Systematic Review and Meta-Analysis[J/OL]. Int J Environ Res Public Health, 2022, 19(23): 15875 [2025-02-06]. https://doi.org/10.3390/ijerph192315875. DOI: 10.3390/ijerph192315875.
[18]
PEROSA V, ZANON ZOTIN M C, SCHOEMAKER D, et al. Association Between Hippocampal Volumes and Cognition in Cerebral Amyloid Angiopathy[J/OL]. Neurology, 2024, 102(2): e207854 [2025-02-06]. https://doi.org/10.1212/WNL.0000000000207854. DOI: 10.1212/WNL.0000000000207854.
[19]
HANSEEUW B J, JACOBS H I L, SCHULTZ A P, et al. Association of Pathologic and Volumetric Biomarker Changes With Cognitive Decline in Clinically Normal Adults[J/OL]. Neurology, 2023, 101(24): e2533-e2544 [2025-02-06] https://doi.org/10.1212/WNL.0000000000207962. DOI: 10.1212/WNL.0000000000207962.
[20]
SHIMODA M, KANEKO K, NAKAGAWA T, et al. Relationship Between Fasting Blood Glucose Levels in Middle Age and Cognitive Function in Later Life: The Aichi Workers' Cohort Study[J]. J Epidemiol, 2023, 33(2): 76-81. DOI: 10.2188/jea.JE20210128.
[21]
LI J, QI J L, PAN W L, et al. Volume changes in the hippocampus and internal olfactory cortex and their relationship with MMSE scores in Alzheimer's disease patients with different degrees of cognitive impairment[J]. Chinese Journal of Anatomy and Clinics, 2023, 28(3): 159-164. DOI: 10.3760/cma.j.cn101202-20220726-00223.
[22]
GUO P F, LI X W, LI S Y. Magnetic resonance imaging study of structural changes in the anterior and posterior hippocampal subregions in schizophrenia[J]. Chin J Magn Reson Imaging, 2018, 9(6): 433-438. DOI: 10.12015/issn.1674-8034.2018.06.007.
[23]
CAI L N, LI X L, PAN Y, et al. Research progress of MRI radiomics in mild cognitive impairment[J]. Chin J Magn Reson Imaging, 2022, 13(6): 131-134. DOI: 10.12015/issn.1674-8034.2022.06.027.
[24]
MEI L L, ZHANG M M, YANG H K, et al. Research progress on neuroimaging biomarkers of cognitive impairment in patients with type 2 diabetes[J]. Chin J Magn Reson Imaging, 2023, 14(9): 108-113. DOI: 10.12015/issn.1674-8034.2023.09.020.
[25]
UYSAL G, OZTURK M. Hippocampal atrophy based Alzheimer's disease diagnosis via machine learning methods[J/OL]. J Neurosci Methods, 2020, 337: 108669 [2025-02-06]. https://doi.org/10.1016/j.jneumeth.2020.108669. DOI: 10.1016/j.jneumeth.2020.108669.
[26]
ANTAL B, MCMAHON L P, SULTAN S F, et al. Type 2 diabetes mellitus accelerates brain aging and cognitive decline: Complementary findings from UK Biobank and meta-analyses[J/OL]. Elife, 2022, 11: e73138 [2025-02-06]. https://doi.org/10.7554/eLife.73138. DOI: 10.7554/eLife.73138.
[27]
CHAKRABORTY A, HEGDE S, PRAHARAJ S K, et al. Age Related Prevalence of Mild Cognitive Impairment in Type 2 Diabetes Mellitus Patients in the Indian Population and Association of Serum Lipids With Cognitive Dysfunction[J/OL]. Front Endocrinol (Lausanne), 2021, 12: 798652 [2025-02-06]. https://doi.org/10.3389/fendo.2021.798652. DOI: 10.3389/fendo.2021.798652.
[28]
LI W, YUE L, SUN L, et al. Elevated Fasting Plasma Glucose Is Associated With an Increased Risk of MCI: A Community-Based Cross-Sectional Study[J/OL]. Front Endocrinol (Lausanne), 2021, 12: 739257 [2025-02-06]. https://doi.org/10.3389/fendo.2021.739257. DOI: 10.3389/fendo.2021.739257.
[29]
LIU Z, ZAID M, HISAMATSU T, et al. Elevated Fasting Blood Glucose Levels Are Associated With Lower Cognitive Function, With a Threshold in Non-Diabetic Individuals: A Population-Based Study[J]. J Epidemiol, 2020, 30(3): 121-127. DOI: 10.2188/jea.JE20180193.
[30]
DAI J, CUI Y. Correlation analysis of glycated hemoglobin variability and cognitive dysfunction in elderly patients with type 2 diabetes mellitus[J]. Practical Geriatrics, 2024, 38(9): 931-935. DOI: 10.3969/j.issn.1003-9198.2024.09.016.
[31]
KADAR T, ARBEL I, SILBERMANN M, et al. Morphological hippocampal changes during normal aging and their relation to cognitive deterioration[J]. J Neural Transm Suppl, 1994, 44: 133-143. DOI: 10.1007/978-3-7091-9350-1_10.
[32]
LI M R, LI Y F, LIU Y J, et al. Altered Hippocampal Subfields Volumes Is Associated With Memory Function in Type 2 Diabetes Mellitus[J/OL]. Front Neurol, 2021, 12: 756500 [2025-02-06]. https://doi.org/10.3389/fneur.2021.756500. DOI: 10.3389/fneur.2021.756500.
[33]
ZHAO L, WANG D Q, ZHU Y, et al. Imaging characteristics of hippocampal morphology and volume changes in patients with Alzheimer's disease[J]. Chin J Behav Med Brain Sci, 2010, 19(3): 200-202. DOI: 10.3760/cma.j.issn.1674-6554.2010.03.003.
[34]
KASSRAIAN P, BIGLER S K, SUAREZ D M G, et al. The hippocampal CA2 region discriminates social threat from social safety[J]. Nat Neurosci, 2024, 27(10): 1710-1720. DOI: 10.1038/s41593-024-01771-8.
[35]
GAO Y N, ZHOU S B, HUANG Y J, et al. A two-stage estimation method for ultra-high-dimensional sparse quadratic discriminant analysis[J]. Statistics and Decision Making, 2022, 38(6): 9-14. DOI: 10.13546/j.cnki.tjyjc.2022.06.002.
[36]
WU R Y, HAO N. Quadratic discriminant analysis by projection[J/OL]. J Multivar Anal, 2022, 190: 104987 [2025-02-06]. https://doi.org/10.1016/j.jmva.2022.104987. DOI: 10.1016/j.jmva.2022.104987.
[37]
XIA H W, LUAN X Q, BAO Z K, et al. A multi-cohort study of the hippocampal radiomics model and its associated biological changes in Alzheimer's Disease[J/OL]. Transl Psychiatry, 2024, 14(1): 111 [2025-02-06]. https://doi.org/10.1038/s41398-024-02836-9. DOI: 10.1038/s41398-024-02836-9.

PREV Study on the correlation between brain structure-function coupling and cognitive function in end-stage renal disease patients using multimodal magnetic resonance imaging
NEXT Association of MRI indexes of brain glymphatic function with sleep status in insomnia patients disorders and the effects of repetitive transcranial magnetic stimulation treatment on them
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn