1
|
Ahn K, Cho M, Kim SW, Lee KE, Song Y, Yoo S, Jeon SY, Kim JL, Yoon DH, Kong HJ. Deep Learning of Speech Data for Early Detection of Alzheimer's Disease in the Elderly. Bioengineering (Basel) 2023; 10:1093. [PMID: 37760195 PMCID: PMC10525115 DOI: 10.3390/bioengineering10091093] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/04/2023] [Revised: 08/16/2023] [Accepted: 08/24/2023] [Indexed: 09/29/2023] Open
Abstract
BACKGROUND Alzheimer's disease (AD) is the most common form of dementia, which makes the lives of patients and their families difficult for various reasons. Therefore, early detection of AD is crucial to alleviating the symptoms through medication and treatment. OBJECTIVE Given that AD strongly induces language disorders, this study aims to detect AD rapidly by analyzing the language characteristics. MATERIALS AND METHODS The mini-mental state examination for dementia screening (MMSE-DS), which is most commonly used in South Korean public health centers, is used to obtain negative answers based on the questionnaire. Among the acquired voices, significant questionnaires and answers are selected and converted into mel-frequency cepstral coefficient (MFCC)-based spectrogram images. After accumulating the significant answers, validated data augmentation was achieved using the Densenet121 model. Five deep learning models, Inception v3, VGG19, Xception, Resnet50, and Densenet121, were used to train and confirm the results. RESULTS Considering the amount of data, the results of the five-fold cross-validation are more significant than those of the hold-out method. Densenet121 exhibits a sensitivity of 0.9550, a specificity of 0.8333, and an accuracy of 0.9000 in a five-fold cross-validation to separate AD patients from the control group. CONCLUSIONS The potential for remote health care can be increased by simplifying the AD screening process. Furthermore, by facilitating remote health care, the proposed method can enhance the accessibility of AD screening and increase the rate of early AD detection.
Collapse
Affiliation(s)
- Kichan Ahn
- Interdisciplinary Program in Medical Informatics Major, Seoul National University College of Medicine, Seoul 03080, Republic of Korea;
| | - Minwoo Cho
- Department of Transdisciplinary Medicine, Seoul National University Hospital, Seoul 03080, Republic of Korea;
- Medical Big Data Research Center, Seoul National University College of Medicine, Seoul 03080, Republic of Korea; (S.W.K.); (K.E.L.)
- Department of Medicine, Seoul National University College of Medicine, Seoul 03080, Republic of Korea
| | - Suk Wha Kim
- Medical Big Data Research Center, Seoul National University College of Medicine, Seoul 03080, Republic of Korea; (S.W.K.); (K.E.L.)
- Department of Plastic Surgery and Institute of Aesthetic Medicine, CHA Bundang Medical Center, CHA University, Seongnam 13496, Republic of Korea
| | - Kyu Eun Lee
- Medical Big Data Research Center, Seoul National University College of Medicine, Seoul 03080, Republic of Korea; (S.W.K.); (K.E.L.)
- Department of Surgery, Seoul National University Hospital and College of Medicine, Seoul 03080, Republic of Korea
| | - Yoojin Song
- Department of Psychiatry, Kangwon National University, Chuncheon 24289, Republic of Korea;
| | - Seok Yoo
- Unidocs Inc., Seoul 03080, Republic of Korea;
| | - So Yeon Jeon
- Department of Psychiatry, Chungnam National University Hospital, Daejeon 30530, Republic of Korea; (S.Y.J.); (J.L.K.)
| | - Jeong Lan Kim
- Department of Psychiatry, Chungnam National University Hospital, Daejeon 30530, Republic of Korea; (S.Y.J.); (J.L.K.)
- Department of Psychiatry, Chungnam National University College of Medicine, Daejeon 30530, Republic of Korea
| | - Dae Hyun Yoon
- Department of Psychiatry, Healthcare System Gangnam Center, Seoul National University Hospital, Seoul 03080, Republic of Korea;
| | - Hyoun-Joong Kong
- Department of Transdisciplinary Medicine, Seoul National University Hospital, Seoul 03080, Republic of Korea;
- Medical Big Data Research Center, Seoul National University College of Medicine, Seoul 03080, Republic of Korea; (S.W.K.); (K.E.L.)
- Department of Medicine, Seoul National University College of Medicine, Seoul 03080, Republic of Korea
| |
Collapse
|
2
|
Prins S, de Kam ML, Teunissen CE, Groeneveld GJ. Inflammatory plasma biomarkers in subjects with preclinical Alzheimer's disease. Alzheimers Res Ther 2022; 14:106. [PMID: 35922871 PMCID: PMC9347121 DOI: 10.1186/s13195-022-01051-2] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/23/2021] [Accepted: 07/21/2022] [Indexed: 12/30/2022]
Abstract
Background This study investigated plasma biomarkers for neuroinflammation associated with Alzheimer’s disease (AD) in subjects with preclinical AD compared to healthy elderly. How these biomarkers behave in patients with AD, compared to healthy elderly is well known, but determining these in subjects with preclinical AD is not and will add information related to the onset of AD. When found to be different in preclinical AD, these inflammatory biomarkers may be used to select preclinical AD subjects who are most likely to develop AD, to participate in clinical trials with new disease-modifying drugs. Methods Healthy elderly (n= 50; age 71.9; MMSE >24) and subjects with preclinical AD (n=50; age 73.4; MMSE >24) defined by CSF Aβ1-42 levels < 1000 pg/mL were included. Four neuroinflammatory biomarkers were determined in plasma, GFAP, YKL-40, MCP-1, and eotaxin-1. Differences in biomarker outcomes were compared using ANCOVA. Subject characteristics age, gender, and APOE ε4 status were reported per group and were covariates in the ANCOVA. Least square means were calculated for all 4 inflammatory biomarkers using both the Aβ+/Aβ− cutoff and Ptau/Aβ1-42 ratio. Results The mean (standard deviation, SD) age of the subjects (n=100) was 72.6 (4.6) years old with 62 male and 38 female subjects. Mean (SD) overall MMSE score was 28.7 (0.49) and 32 subjects were APOE ε4 carriers. The number of subjects in the different APOE ε4 status categories differed significantly between the Aβ+ and Aβ− groups. Plasma GFAP concentration was significantly higher in the Aβ+ group compared to the Aβ− group with significant covariates age and sex, variables that also correlated significantly with GFAP. Conclusion GFAP was significantly higher in subjects with preclinical AD compared to healthy elderly which agrees with previous studies. When defining preclinical AD based on the Ptau181/Aβ1-42 ratio, YKL-40 was also significantly different between groups. This could indicate that GFAP and YKL-40 are more sensitive markers of the inflammatory process in response to the Aβ misfolding and aggregation that is ongoing as indicated by the lowered Aβ1-42 levels in the CSF. Characterizing subjects with preclinical AD using neuroinflammatory biomarkers is important for subject selection in new disease-modifying clinical trials. Trial registration ISRCTN.org identifier: ISRCTN79036545 (retrospectively registered).
Collapse
Affiliation(s)
- Samantha Prins
- Centre for Human Drug Research, Leiden, the Netherlands.,Leiden University Medical Center, Leiden, the Netherlands
| | | | - Charlotte E Teunissen
- Neurochemistry Laboratory, Department of Clinical Chemistry, Amsterdam Neuroscience, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands
| | - Geert Jan Groeneveld
- Centre for Human Drug Research, Leiden, the Netherlands. .,Leiden University Medical Center, Leiden, the Netherlands.
| |
Collapse
|
3
|
Neuroimmune contributions to Alzheimer's disease: a focus on human data. Mol Psychiatry 2022; 27:3164-3181. [PMID: 35668160 PMCID: PMC9168642 DOI: 10.1038/s41380-022-01637-0] [Citation(s) in RCA: 15] [Impact Index Per Article: 7.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/15/2021] [Revised: 05/05/2022] [Accepted: 05/18/2022] [Indexed: 12/12/2022]
Abstract
The past decade has seen the convergence of a series of new insights that arose from genetic and systems analyses of Alzheimer's disease (AD) with a wealth of epidemiological data from a variety of fields; this resulted in renewed interest in immune responses as important, potentially causal components of AD. Here, we focus primarily on a review of human data which has recently yielded a set of robust, reproducible results that exist in a much larger universe of conflicting reports stemming from small studies with important limitations in their study design. Thus, we are at an important crossroads in efforts to first understand at which step of the long, multiphasic course of AD a given immune response may play a causal role and then modulate this response to slow or block the pathophysiology of AD. We have a wealth of new experimental tools, analysis methods, and capacity to sample human participants at large scale longitudinally; these resources, when coupled to a foundation of reproducible results and novel study designs, will enable us to monitor human immune function in the CNS at the level of complexity that is required while simultaneously capturing the state of the peripheral immune system. This integration of peripheral and central perturbations in immune responses results in pathologic responses in the central nervous system parenchyma where specialized cellular microenvironments composed of multiple cell subtypes respond to these immune perturbations as well as to environmental exposures, comorbidities and the impact of the advancing life course. Here, we offer an overview that seeks to illustrate the large number of interconnecting factors that ultimately yield the neuroimmune component of AD.
Collapse
|