• A Hybrid CNN-SVM Approach for...

*Result*: A Hybrid CNN-SVM Approach for ECG-Based Multi-Class Differential Diagnosis of PTSD, Depression, and Panic Attack.

Title:
A Hybrid CNN-SVM Approach for ECG-Based Multi-Class Differential Diagnosis of PTSD, Depression, and Panic Attack.
Authors:
Ebrahimpour Moghaddam Tasouj P; Biomedical Device Technology, Vocational School of Health Services, Ankara Medipol University, Ankara 06050, Türkiye., Soysal G; Department of Electrical and Electronics Engineering, Ankara University, Ankara 06830, Türkiye., Eroğul O; Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara 06510, Türkiye., Yetkin S; Department of Psychiatry, Ankara Gülhane Health Application and Research Hospital, University of Health Sciences, Ankara 06010, Türkiye.
Source:
Biosensors [Biosensors (Basel)] 2026 Jan 10; Vol. 16 (1). Date of Electronic Publication: 2026 Jan 10.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: MDPI Pub Country of Publication: Switzerland NLM ID: 101609191 Publication Model: Electronic Cited Medium: Internet ISSN: 2079-6374 (Electronic) Linking ISSN: 20796374 NLM ISO Abbreviation: Biosensors (Basel) Subsets: MEDLINE
Imprint Name(s):
Original Publication: Basel, Switzerland : MDPI Pub.
MeSH Terms:
Panic Disorder*/diagnosis , Stress Disorders, Post-Traumatic*/diagnosis , Depression*/diagnosis , Electrocardiography* , Neural Networks, Computer* , Support Vector Machine*, Humans ; Diagnosis, Differential ; Male ; Female ; Adult ; Middle Aged
References:
Dialogues Clin Neurosci. 2015 Jun;17(2):141-50. (PMID: 26246789)
Am Heart J. 2013 Nov;166(5):806-14. (PMID: 24176435)
Braz J Psychiatry. 2007 May;29 Suppl 1:S3-6. (PMID: 17546345)
Front Neurorobot. 2022 Aug 31;16:873239. (PMID: 36119719)
Biosensors (Basel). 2021 Nov 14;11(11):. (PMID: 34821669)
J Clin Med. 2021 Aug 20;10(16):. (PMID: 34442005)
J Trauma Stress. 2013 Jun;26(3):299-309. (PMID: 23696449)
Heliyon. 2024 Jun 08;10(12):e32548. (PMID: 38975193)
J Clin Psychiatry. 2022 Apr 25;83(3):. (PMID: 35485933)
Front Neuroinform. 2021 Nov 25;15:777977. (PMID: 34899226)
Transl Psychiatry. 2023 Jan 27;13(1):25. (PMID: 36707505)
Biosensors (Basel). 2022 Dec 09;12(12):. (PMID: 36551120)
Psychoneuroendocrinology. 2022 Jan;135:105575. (PMID: 34741982)
Front Psychiatry. 2023 Dec 21;14:1310323. (PMID: 38179243)
Psychol Med. 2017 Oct;47(13):2260-2274. (PMID: 28385165)
Biosensors (Basel). 2023 Jan 18;13(2):. (PMID: 36831919)
Diagnostics (Basel). 2025 Jun 02;15(11):. (PMID: 40506986)
Nat Med. 2019 Jan;25(1):65-69. (PMID: 30617320)
Eur J Health Econ. 2020 Aug;21(6):917-930. (PMID: 32458163)
Front Psychiatry. 2024 May 08;15:1296569. (PMID: 38779550)
Open Cardiovasc Med J. 2011;5:164-70. (PMID: 21792377)
Biosensors (Basel). 2025 Jun 17;15(6):. (PMID: 40558474)
Front Psychiatry. 2022 Oct 19;13:953001. (PMID: 36339839)
Ann Behav Med. 2010 Feb;39(1):61-78. (PMID: 20174903)
Am J Cardiovasc Dis. 2024 Jun 15;14(3):172-179. (PMID: 39021521)
Depress Anxiety. 2023 Nov 8;2023:4453663. (PMID: 40224591)
Precis Clin Med. 2020 Aug 28;3(3):202-213. (PMID: 35694413)
Heliyon. 2024 Feb 21;10(5):e26147. (PMID: 38434292)
NPJ Digit Med. 2022 Jul 7;5(1):87. (PMID: 35798934)
J Am Acad Child Adolesc Psychiatry. 2010 Oct;49(10):980-9. (PMID: 20855043)
Comput Biol Med. 2021 Jul;134:104532. (PMID: 34102402)
Am J Physiol Regul Integr Comp Physiol. 2015 Aug 15;309(4):R315-21. (PMID: 26062635)
Biosensors (Basel). 2021 Mar 04;11(3):. (PMID: 33806367)
Contributed Indexing:
Keywords: depression; differential diagnosis; electrocardiogram; hybrid deep learning; panic attack; post-traumatic stress disorder
Entry Date(s):
Date Created: 20260127 Date Completed: 20260127 Latest Revision: 20260131
Update Code:
20260131
PubMed Central ID:
PMC12838843
DOI:
10.3390/bios16010052
PMID:
41590304
Database:
MEDLINE

*Further Information*

*Background: PTSD diagnosis is challenging. Symptoms overlap with depression and panic attacks. This causes misdiagnosis and delayed treatment. Current methods lack objective biomarkers. This study presents a hybrid AI framework. It combines CNNs and SVMs. The system detects PTSD from ECG signals.
Methods: ECG data from 79 participants were analyzed. Four groups were included. PTSD patients numbered 20. Depression patients numbered 20. Panic attack patients numbered 19. Healthy controls numbered 20. Wavelet transform created scalograms. Three CNN models were tested. AlexNet, GoogLeNet, and ResNet50 were used. Deep features were extracted. SVMs classified the features. Five-fold validation was performed. Statistical tests confirmed significance.
Results: Hybrid models performed robustly. ResNet50 + SVM and AlexNet + SVM achieved statistically equivalent results with accuracies of 97.05% and 97.26%, respectively. AUC reached 1.00 for multi-class tasks. PTSD detection was highly accurate. The system distinguished PTSD from other disorders. Hybrid models beat standalone CNNs. SVM integration improved results significantly.
Conclusions: This is the first ECG-based AI for PTSD diagnosis. The hybrid approach achieves clinical-level accuracy. PTSD is distinguished from depression and panic attacks. Objective biomarkers support psychiatric assessment. Early intervention becomes possible.*