1
|
Chen M, Xia X, Kang Z, Li Z, Dai J, Wu J, Chen C, Qiu Y, Liu T, Liu Y, Zhang Z, Shen Q, Tao S, Deng Z, Lin Y, Wei Q. Distinguishing schizophrenia and bipolar disorder through a Multiclass Classification model based on multimodal neuroimaging data. J Psychiatr Res 2024; 172:119-128. [PMID: 38377667 DOI: 10.1016/j.jpsychires.2024.02.024] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/15/2023] [Revised: 02/05/2024] [Accepted: 02/07/2024] [Indexed: 02/22/2024]
Abstract
This study aimed to identify neural biomarkers for schizophrenia (SZ) and bipolar disorder (BP) by analyzing multimodal neuroimaging. Utilizing data from structural magnetic resonance imaging (sMRI), diffusion tensor imaging (DTI), and resting-state functional magnetic resonance imaging (rs-fMRI), multiclass classification models were created for SZ, BP, and healthy controls (HC). A total of 113 participants (BP: 31, SZ: 39, and HC: 43) were recruited under strict enrollment control, from which 272, 200, and 1875 features were extracted from sMRI, DTI, and rs-fMRI data, respectively. A support vector machine (SVM) with recursive feature elimination (RFE) was employed to build the models using a one-against-one approach and leave-one-out cross-validation, achieving a classification accuracy of 70.8%. The most discriminative features were primarily from rs-fMRI, along with significant findings in sMRI and DTI. Key biomarkers identified included the increased thickness of the left cuneus cortex and decreased regional functional connectivity strength (rFCS) in the left supramarginal gyrus as shared indicators for BP and SZ. Additionally, decreased fractional anisotropy in the left superior fronto-occipital fasciculus was suggested as specific to BP, while decreased rFCS in the left inferior parietal area might serve as a specific biomarker for SZ. These findings underscore the potential of multimodal neuroimaging in distinguishing between BP and SZ and contribute to the understanding of their neural underpinnings.
Collapse
Affiliation(s)
- Ming Chen
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China; Guangdong Mental Health Institute, Guangdong ProvincialPeople's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Xiaowei Xia
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Zhuang Kang
- Department of Radiology, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Zhinan Li
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Jiamin Dai
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Junyan Wu
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China; Department of Psychiatry, The Affiliated Brain Hospital of Guangzhou Medical University, Guangzhou, China
| | - Cai Chen
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Yong Qiu
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China; Department of Psychiatry, Mindfront Caring Medical, Guangzhou, China
| | - Tong Liu
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China; Department of Psychiatry, The First Affiliated Hospital of Xi'an Jiaotong University, Shaanxi, China
| | - Yanxi Liu
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Ziyi Zhang
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China; Department of Medical Division, Sun Yat-sen University Cancer Center, Guangzhou, China
| | - Qingni Shen
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Sichu Tao
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Zixin Deng
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
| | - Ying Lin
- Department of Psychology, Sun Yat-sen University, Guangzhou, China.
| | - Qinling Wei
- Department of Psychiatry, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China.
| |
Collapse
|
2
|
Physiological State Evaluation in Working Environment Using Expert System and Random Forest Machine Learning Algorithm. Healthcare (Basel) 2023; 11:healthcare11020220. [PMID: 36673588 PMCID: PMC9859340 DOI: 10.3390/healthcare11020220] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/24/2022] [Revised: 01/06/2023] [Accepted: 01/09/2023] [Indexed: 01/13/2023] Open
Abstract
Healthy lifestyle is one of the most important factors in the prevention of premature deaths, chronic diseases, productivity loss, obesity, and other economic and social aspects. The workplace plays an important role in promoting the physical activity and wellbeing of employees. Previous studies are mostly focused on individual interviews, various questionnaires that are a conceptual information about individual health state and might change according to question formulation, specialist competence, and other aspects. In this paper the work ability was mostly related to the employee's physiological state, which consists of three separate systems: cardiovascular, muscular, and neural. Each state consists of several exercises or tests that need to be performed one after another. The proposed data transformation uses fuzzy logic and different membership functions with three or five thresholds, according to the analyzed physiological feature. The transformed datasets are then classified into three stages that correspond to good, moderate, and poor health condition using machine learning techniques. A three-part Random Forest method was applied, where each part corresponds to a separate system. The obtained testing accuracies were 93%, 87%, and 73% for cardiovascular, muscular, and neural human body systems, respectively. The results indicate that the proposed work ability evaluation process may become a good tool for the prevention of possible accidents at work, chronic fatigue, or other health problems.
Collapse
|
3
|
ElOuassif B, Idri A, Hosni M, Abran A. Classification techniques in breast cancer diagnosis: A systematic literature review. COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING: IMAGING & VISUALIZATION 2021. [DOI: 10.1080/21681163.2020.1811159] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 10/22/2022]
Affiliation(s)
- Bouchra ElOuassif
- Department of Web and Mobile Engineering, Software Project Management Research Team, ENSIAS, Mohammed V University, Rabat, Morocco
| | - Ali Idri
- Department of Web and Mobile Engineering, Software Project Management Research Team, ENSIAS, Mohammed V University, Rabat, Morocco
| | - Mohamed Hosni
- Department of Web and Mobile Engineering, Software Project Management Research Team, ENSIAS, Mohammed V University, Rabat, Morocco
| | - Alain Abran
- Department of Software Engineering and Information Technology, Ecole De Technologie Supérieure, –university of Québec, Montreal, Canada
| |
Collapse
|
4
|
Mao N, Yin P, Wang Q, Liu M, Dong J, Zhang X, Xie H, Hong N. Added Value of Radiomics on Mammography for Breast Cancer Diagnosis: A Feasibility Study. J Am Coll Radiol 2019; 16:485-491. [DOI: 10.1016/j.jacr.2018.09.041] [Citation(s) in RCA: 28] [Impact Index Per Article: 5.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/25/2018] [Accepted: 09/14/2018] [Indexed: 01/22/2023]
|
5
|
DNA Methylation as a Noninvasive Epigenetic Biomarker for the Detection of Cancer. DISEASE MARKERS 2017; 2017:3726595. [PMID: 29038612 PMCID: PMC5605861 DOI: 10.1155/2017/3726595] [Citation(s) in RCA: 78] [Impact Index Per Article: 11.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Subscribe] [Scholar Register] [Received: 04/28/2017] [Revised: 07/10/2017] [Accepted: 08/07/2017] [Indexed: 12/30/2022]
Abstract
In light of the high incidence and mortality rates of cancer, early and accurate diagnosis is an important priority for assigning optimal treatment for each individual with suspected illness. Biomarkers are crucial in the screening of patients with a high risk of developing cancer, diagnosing patients with suspicious tumours at the earliest possible stage, establishing an accurate prognosis, and predicting and monitoring the response to specific therapies. Epigenetic alterations are innovative biomarkers for cancer, due to their stability, frequency, and noninvasive accessibility in bodily fluids. Epigenetic modifications are also reversible and potentially useful as therapeutic targets. Despite this, there is still a lack of accurate biomarkers for the conclusive diagnosis of most cancer types; thus, there is a strong need for continued investigation to expand this area of research. In this review, we summarise current knowledge on methylated DNA and its implications in cancer to explore its potential as an epigenetic biomarker to be translated for clinical application. We propose that the identification of biomarkers with higher accuracy and more effective detection methods will enable improved clinical management of patients and the intervention at early-stage disease.
Collapse
|