Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data
Abstract
:1. Introduction
- Identifying the primary socio-demographic and psychological determinants contributing to the development of depression.
- This study aims to investigate several machine learning algorithms and feature selection techniques to detect the presence of depression effectively.
- This study aims to facilitate the interpretation of high-risk groups of depressive disorders in primary care settings for primary physicians. Primary care clinicians will achieve this by applying the Depression Ensemble Stacking Generalization Model (DESGM) to identify and analyze the essential determinants associated with these high-risk groups.
2. Related Work
2.1. Risk Factors
2.2. Ensemble Learning Techniques
3. Proposed Method—Depression Ensemble Stacking Generalization Model (DESGM)
3.1. Data Source Representation
3.2. Data Pre-Processing
- Let .
- For each item f in .
- If f may indicate a symptom (e.g., trouble sleeping or sleeping too much), then add f to A.
- Let denote the projection of onto A.
3.3. Feature Selection: Iterative with Floating Elimination Algorithm (IFEA)
Algorithm 1: Iterative with Floating Elimination Algorithm (IFEA) | ||
Input: Y = {y1, y2 … yk}//Y—set of all features | ||
Output: Xd = {xi\i = 1, 2, …, d; xi Y} where d = {1, 2 … k} d is the number of desired features | ||
1. | Initialize X0 = Y | |
2. | for i = 1 → k–d: | |
// maximizes criterion function upon removal | ||
//Remove feature from the feature subset | ||
End for | ||
3. | If (Xd | |
Go to step 4 | ||
End if | ||
4. | ||
Go to step 3 | ||
End if | ||
5. | End |
3.4. Stacked Ensemble Model—DESGM
Base Learners
- Initialize w = 0.
- For each training sample , the output value is computed using the following activation function:
- Update the weights.
- The weight vector w is then updated based on the computed output and the actual class label of the ith training sample using the update rule:
3.5. Artificial Neural Network (ANN)
3.6. K-Nearest Neighbor (KNN)
3.7. Linear Support Vector Machine (LSVM)
- (a)
- Training LSVM:
- Compute the gram matrix.
- Let value:The value of parameter C controls the margin’s width.
- Determine the weight.
- Calculate the intercept.
- (b)
- Classification LSVM (For a new sample x)
- Determine the support vector’s .
- Calculate f(x).
- Examine sign (f(x)).
3.8. Linear Discriminant Analysis (LDA)
- Let be the collection of d-dimensional samples for class k, represented as .
- Let be the data matrix, stacking all the samples from all classes, with each column representing a sample, where .
- Let us specify how samples x are transformed onto a line in [(c 1)-space, for c = 2]. is a projected vector.
Meta-Learner—Logistic Regression (LR)
3.9. Depression Ensemble Stacking Generalization Model (DESGM)
Algorithm 2: Depression Ensemble Stacking Generalization Algorithm (DESGM) | |||||
Input: Healthcare Survey Data Dataset B-base learners [LSVM, LDA, ANN, KNN, Perceptron], each individual classifier is optimized for each sub-dataset, Meta-learner LR | |||||
Output: An Optimized stacking ensemble classifier H | |||||
1 | Using the questionnaire data as the base, construct the dataset . | ||||
2 | Aggregate features about seven psychological functions. | ||||
3 | Feature selection using SBFS(Ai). | ||||
for algorithm Ai in | |||||
return the sorted feature subset ) | |||||
end for | |||||
4 | Using a 5-fold cross-validation procedure, randomly divide the dataset into k subgroups of equal size. | ||||
for each sub-dataset do | |||||
For k = 1 to K do | |||||
//4.1 Classifiers of level 0 should be learned. | |||||
For t = 1 to T, do | |||||
Construct a classifier using | |||||
end for | |||||
//4.2 Design a training set at the level-1 | |||||
For do | |||||
= Create a new instance | |||||
end for | |||||
end for | |||||
end | |||||
5 | Concatenate the SB produced by all base classifiers. | ||||
6 | //Learn a classifier at level 2. | ||||
Construct a new classifier h’ using the obtained SB//LR data. | |||||
7 | Rebuild all of the level-1 classifier. | ||||
For t = 1 to 5, do | |||||
Develop the classification model using | |||||
end for | |||||
return |
4. Results and Discussions
4.1. Performance Metrics
4.2. Results of Feature Selection Method
4.3. Results of Ensemble Model
4.4. Results of DESGM
4.5. Role of Feature Importance—Random Forest (RF)
4.6. Discussion
5. Conclusions and Future Work
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Institute of Health Metrics and Evaluation. Global Health Data Exchange (GHDx). Available online: https://vizhub.healthdata.org/gbd-results/ (accessed on 4 March 2023).
- Chakrabarty, T.; Sarfati, D.; Lam, R.W. Chapter 24—Neurobiological Aspects of Functional Recovery in Major Depressive Disorder. In Neurobiology of Depression; Academic Press: Cambridge, MA, USA, 2019; pp. 277–284. [Google Scholar] [CrossRef]
- Murthy, R.S. National Mental Health Survey of India 2015–2016. Indian J. Psychiatry 2017, 59, 21–26. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Kessler, R.C.; Aguilar-Gaxiola, S.; Alonso, J.; Chatterji, S.; Lee, S.; Ormel, J.; Üstün, T.B.; Wang, P.S. The global burden of mental disorders: An update from the WHO World Mental Health (WMH) surveys. Epidemiol. Psichiatr. Soc. 2009, 18, 23–33. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Sagar, R.S.; Mohan, D.; Kumar, V.; Khandelwal, S.K.; Nair, P.G. Physical illnesses among elderly psychiatric out-patients with depression. Indian J. Psychiatry 1992, 34, 41–45. [Google Scholar] [PubMed] [PubMed Central]
- Nguyen, D.-K.; Chan, C.-L.; Li, A.-H.A.; Phan, D.-V. Deep Stacked Generalization Ensemble Learning models in early diagnosis of Depression illness from wearable devices data. In Proceedings of the 5th International Conference on Medical and Health Informatics (ICMHI ‘21), Association for Computing Machinery, New York, NY, USA, 7–12 May 2021. [Google Scholar] [CrossRef]
- Bohr, A.; Memarzadeh, K. The rise of artificial intelligence in healthcare applications. In Artificial Intelligence in Healthcare; Academic Press: Cambridge, MA, USA, 2020; pp. 25–60. [Google Scholar] [CrossRef] [PubMed Central]
- Aleem, S.; Huda, N.U.; Amin, R.; Khalid, S.; Alshamrani, S.S.; Alshehri, A. Machine Learning Algorithms for Depression: Diagnosis, Insights, and Research Directions. Electronics 2022, 11, 1111. [Google Scholar] [CrossRef]
- Tao, X.; Chi, O.; Delaney, P.J.; Li, L.; Huang, J. Detecting depression using an ensemble classifier based on Quality of Life scales. Brain Inf. 2021, 8, 2198–4026. [Google Scholar] [CrossRef]
- Merikangas, A.K.; Mendola, P.; Pastor, P.N.; Reuben, C.A.; Cleary, S.D. The association between major depressive disorder and obesity in US adolescents: Results from the 2001–2004 National Health and Nutrition Examination Survey. J. Behav. Med. 2011, 35, 149–154. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Kroenke, K.; Spitzer, R.L.; Williams, J.B.; Löwe, B. The Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptom Scales: A systematic review. Gen. Hosp. Psychiatry 2010, 32, 345–359. [Google Scholar] [CrossRef] [PubMed]
- Gerych, W.; Agu, E.; Rundensteiner, E. Classifying depression in imbalanced datasets using an autoencoder-based anomaly detection approach. In Proceedings of the 2019 IEEE 13th International Conference on Semantic Computing (ICSC), Newport Beach, CA, USA, 30 January–1 February 2019; pp. 124–127. [Google Scholar]
- Oh, J.; Yun, K.; Maoz, U.; Kim, T.S.; Chae, J.H. Identifying depression in the National Health and Nutrition Examination Survey data using a deep learning algorithm. J. Affect. Disord. 2019, 257, 623–631. [Google Scholar] [CrossRef] [PubMed]
- Dipnall, J.F.; Pasco, J.A.; Berk, M.; Williams, L.J.; Dodd, S.; Jacka, F.N.; Meyer, D. Why so GLUMM? Detecting depression clusters through graphing lifestyle-environs using machine-learning methods (GLUMM). Eur. Psychiatry 2017, 39, 40–50. [Google Scholar] [CrossRef] [PubMed]
- Sharma, A.; Verbeke, W.J.M.I. Improving Diagnosis of Depression with XGBOOST Machine Learning Model and a Large Biomarkers Dutch Dataset (n = 11,081). Front. Big Data 2020, 3, 15. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Wang, L.; Chignell, M.; Jiang, H.; Lokuge, S.; Mason, G.; Fotinos, K.; Katzman, M. Prioritization of Multi-level Risk Factors, and Predicting Changes in Depression Ratings after Treatment Using Multi-Task Learning. In Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Houston, TX, USA, 9–12 December 2021; pp. 3239–3244. [Google Scholar] [CrossRef]
- Wang, L.; Zhu, D.; Towner, E.; Dong, M. Obesity risk factors ranking using multi-task learning. In Proceedings of the 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), Las Vegas, NV, USA, 4–7 March 2018; pp. 385–388. [Google Scholar]
- Jayawickreme, N.; Atefi, E.; Jayawickreme, E.; Qin, J.; Gandomi, A.H. Association Rule Learning Is an Easy and Efficient Method for Identifying Profiles of Traumas and Stressors that Predict Psychopathology in Disaster Survivors: The Example of Sri Lanka. Int. J. Environ. Res. Public. Health 2020, 17, 2850. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Jia, Z.; Li, X.; Yuan, X.; Zhang, B.; Liu, Y.; Zhao, J.; Li, S. Depression is associated with diabetes status of family members: NHANES (1999–2016). J. Affect. Disord. 2019, 249, 121–126. [Google Scholar] [CrossRef] [PubMed]
- Byeon, H. Developing a Predictive Model for Depressive Disorders Using Stacking Ensemble and Naive Bayesian Nomogram: Using Samples Representing South Korea. Front. Psychiatry 2022, 12, 773290. [Google Scholar] [CrossRef] [PubMed]
- Habib, M.; Wang, Z.; Qiu, S.; Zhao, H.; Murthy, A.S. Machine Learning Based Healthcare System for Investigating the Association Between Depression and Quality of Life. IEEE J. Biomed. Health Inform. 2022, 26, 2008–2019. [Google Scholar] [CrossRef] [PubMed]
- Thanathamathee, P. Boosting with feature selection technique for screening and predicting adolescents depression. In Proceedings of the 2014 Fourth International Conference on Digital Information and Communication Technology and its Applications (DICTAP), Bangkok, Thailand, 6–8 May 2014; pp. 23–27. [Google Scholar] [CrossRef]
- Lee, S.-J.; Moon, H.-J.; Kim, D.-J.; Yoon, Y. Genetic algorithm-based feature selection for depression scale prediction. In Proceedings of the GECCO ‘19: Genetic and Evolutionary Computation Conference, Prague, Czech Republic, 13–17 July 2019; pp. 65–66. [Google Scholar] [CrossRef]
- Santana, R. Genetic Algorithms for Feature Selection in the Children and Adolescents Depression Context. In Proceedings of the 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA), Boca Raton, FL, USA, 16–19 December 2019; pp. 1470–1475. [Google Scholar] [CrossRef]
- Yu, B.; Zhang, X.; Wang, C.; Sun, M.; Jin, L.; Liu, X. Trends in depression among Adults in the United States, NHANES 2005–2016. J. Affect. Disord. 2019, 263, 609–620, ISSN 0165-0327. [Google Scholar] [CrossRef]
- Thieme, A.; Belgrave, D.; Doherty, G. Machine learning in mental health: A systematic review of the hci literature to support the development of effective and implementable ML systems. ACM Trans. Comp. Hum. Interact (TOCHI) 2020, 27, 1–53. [Google Scholar] [CrossRef]
- Wang, Y.; Lopez, J.M.S.; Bolge, S.C.; Zhu, V.J.; Stang, P.E. Depression among people with type 2 diabetes mellitus, US National Health and Nutrition Examination Survey (NHANES), 2005–2012. BMC Psychiatry 2016, 16, 88. [Google Scholar] [CrossRef] [PubMed]
- Burns, L.; Teesson, M. Alcohol use disorders comorbid with anxiety, depression, and drug use disorders. Findings from the Australian National Survey of Mental Health and Well Being. Drug Alcohol. Depend. 2002, 68, 299–307. [Google Scholar] [CrossRef] [PubMed]
- Bromet, E.; Andrade, L.H.; Hwang, I.; Sampson, N.A.; Alonso, J.; de Girolamo, G.; de Graaf, R.; Demyttenaere, K.; Hu, C.; Iwata, N.; et al. Cross-national epidemiology of DSM-IV major depressive episode. BMC Med. 2011, 9, 90. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Ferrari, A.J.; Charlson, F.J.; Norman, R.E.; Patten, S.B.; Freedman, G.; Murray, C.J.; Vos, T.; Whiteford, H.A. Burden of depressive disorders by country, sex, age, and year: Findings from the global burden of disease study 2010. PLoS Med. 2013, 10, e1001547. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Vincent, P.M.D.R.; Mahendran, N.; Nebhen, J.; Deepa, N.; Srinivasan, K.; Hu, Y.-C. Performance Assessment of Certain Machine Learning Models for Predicting the Major Depressive Disorder among IT Professionals during Pandemic times. Comput. Intell. Neurosci. 2021, 2021, 9950332. [Google Scholar] [CrossRef]
- Dipnall, J.F.; Pasco, J.A.; Berk, M.; Williams, L.J.; Dodd, S.; Jacka, F.N.; Meyer, D. Fusing Data Mining, Machine Learning and Traditional Statistics to Detect Biomarkers Associated with Depression. PLoS ONE 2016, 11, e0148195. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Hassan, A.U.; Hussain, J.; Hussain, M.; Sadiq, M.; Lee, S. Sentiment analysis of social networking sites (SNS) data using machine learning approach for the measurement of depression. In Proceedings of the 2017 International Conference on Information and Communication Technology Convergence (ICTC), Jeju, Republic of Korea, 18–20 October 2017; pp. 138–140. [Google Scholar] [CrossRef]
- Yang, H.; Bath, P.A. Automatic Prediction of Depression in Older Age. In Proceedings of the third International Conference on Medical and Health Informatics 2019, ICMHI 2019, Xiamen, China, 17–19 May 2019; pp. 36–44. [Google Scholar] [CrossRef]
- Tanha, J.; Abdi, Y.; Samadi, N.; Razzaghi, N.; Asadpour, M. Boosting methods for multi-class imbalanced data classification: An experimental review. J. Big Data 2020, 7, 70. [Google Scholar] [CrossRef]
- Hsieh, W.-H.; Shih, D.-H.; Shih, P.-Y.; Lin, S.-B. An Ensemble Classifier with Case-Based Reasoning System for Identifying Internet Addiction. Int. J. Environ. Res. Public Health 2019, 16, 1233. [Google Scholar] [CrossRef]
- Kumar, U.K.; Nikhil, M.B.S.; Sumangali, K. Prediction of breast cancer using voting classifier technique. In Proceedings of the 2017 IEEE International Conference on Smart Technologies and Management for Computing, Communication, Controls, Energy and Materials (ICSTM), Chennai, India, 2–4 August 2017; pp. 108–114. [Google Scholar] [CrossRef]
- Centers for Disease Prevention and Control National Health and Nutrition Examination Survey Overview. National Center for Health Statistics. 2017. Available online: https://wwwn.cdc.gov/nchs/nhanes/search/datapage.aspx?Component=Questionnaire&CycleBeginYear=2015 (accessed on 15 March 2023).
- Byeon, H. Exploring Factors for Predicting Anxiety Disorders of the Elderly Living Alone in South Korea Using Interpretable Machine Learning: A Population-Based Study. Int. J. Environ. Res. Public Health 2021, 18, 7625. [Google Scholar] [CrossRef] [PubMed]
- Yusta, S.C. Different metaheuristic strategies to solve the feature selection problem. Pattern Recognit. Lett. 2009, 30, 525–534. [Google Scholar] [CrossRef]
- Zian, S.; Kareem, S.A.; Varathan, K.D. An Empirical Evaluation of Stacked Ensembles with Different Meta-Learners in Imbalanced Classification. IEEE Access 2021, 9, 87434–87452. [Google Scholar] [CrossRef]
- Pasco, J.A.; Williams, L.J.; Jacka, F.N.; Ng, F.; Henry, M.J.; Nicholson, G.C.; Kotowicz, M.A.; Berk, M. Tobacco Smoking as a Risk Factor for Major Depressive Disorder: Population-based Study. Br. J. Psychiatry 2008, 193, 322–326. [Google Scholar] [CrossRef]
- Riemann, D.; Krone, L.B.; Wulff, K.; Nissen, C. Sleep, insomnia, and depression. Neuropsychopharmacology 2020, 45, 74–89. [Google Scholar] [CrossRef]
- Yan, C.; Shum, D.; Deng, C.P. Responses to academic stress mediate the association between sleep difficulties and depressive/anxiety symptoms in Chinese adolescents. J. Affect. Disord. 2020, 263, 89–98. [Google Scholar] [CrossRef]
- Cahuas, A.; He, Z.; Zhang, Z.; Chen, W. Relationship of physical activity and sleep with depression in college students. J. Am. Coll. Health 2020, 68, 557–564. [Google Scholar] [CrossRef] [PubMed]
- Francis, H.M.; Stevenson, R.J.; Chambers, J.R.; Gupta, D.; Newey, B.; Lim, C.K. A brief diet intervention can reduce symptoms of depression in young adults—A randomised controlled trial. PLoS ONE 2019, 14, e0222768. [Google Scholar] [CrossRef] [PubMed]
- Lai, J.S.; Hiles, S.; Bisquera, A.; Hure, A.J.; McEvoy, M.; Attia, J. A systematic review and meta-analysis of dietary patterns and depression in community-dwelling adults. Am. J. Clin. Nutr. 2014, 99, 181–197. [Google Scholar] [CrossRef] [PubMed]
Dataset | No. of Features | Feature Selection |
---|---|---|
NHANES (1999–2014) [14] | 157 | Reduce multi-collinearity, Remove self-reported depression questions |
KLoSA [23], UFMG [24] | 50, 55 | Genetic algorithm |
NHANES (2005–16) [25] | 20 | Features taken from factors based on related work and the PHQ-9 depression questionnaire |
NHANES (2005–2012) [26] | 51 | Features selected based on Type 2 diabetes mellitus and those factors correlated with depression |
Thasala Hospital [22] | 12 | Max-Relevance Min-Redundancy (MRMR) |
Class Label | No. of Participants |
---|---|
Depressed | 415 |
Not depressed | 4719 |
Feature Selection | No. of Features | DESGM Accuracy |
---|---|---|
IFEA | 110 | 0.9251 |
GA | 0.8547 | |
MRMR | 0.8214 | |
IFEA | 95 | 0.9136 |
GA | 0.8765 | |
MRMR | 0.8355 | |
IFEA | 80 | 0.9495 |
GA | 0.9125 | |
MRMR | 0.8674 | |
IFEA | 70 | 0.9554 |
GA | 0.9210 | |
MRMR | 0.8956 | |
IFEA | 62 | 0.9687 |
GA | 0.9286 | |
MRMR | 0.9015 | |
IFEA | 55 | 0.9512 |
GA | 0.9124 | |
MRMR | 0.8967 |
IFEA Algorithm | |
---|---|
Feature Subset | No of Features |
Socio-Life Style | 9 |
General | 10 |
Pain | 13 |
Physical Health Status | 7 |
Physical Activity | 10 |
Food Behaviour | 4 |
Depression Screener | 9 |
Overall | 62 |
Published Research Paper | Year | Techniques Used | Accuracy |
---|---|---|---|
Jihoon Oh et al. [13] | 2019 | Deep learning algorithm | 92.00% |
Amita Sharma and Willem J [15] | 2020 | Extreme gradient boosting | 90.00% |
Xiaohui Tao et al. [9] | 2021 | Ensemble model | 95.40% |
Masood Habib et al. [21] | 2022 | Posterior probability multi-class Support vector machine | 91.16% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Selvaraj, A.; Mohandoss, L. Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data. Appl. Sci. 2024, 14, 7366. https://doi.org/10.3390/app14167366
Selvaraj A, Mohandoss L. Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data. Applied Sciences. 2024; 14(16):7366. https://doi.org/10.3390/app14167366
Chicago/Turabian StyleSelvaraj, Annapoorani, and Lakshmi Mohandoss. 2024. "Enhancing Depression Detection: A Stacked Ensemble Model with Feature Selection and RF Feature Importance Analysis Using NHANES Data" Applied Sciences 14, no. 16: 7366. https://doi.org/10.3390/app14167366