Mental health, unlike physical health, presents unique challenges in its definition and diagnosis. What constitutes “good” mental health? When does stress become a mental health issue? At what point is a mental health problem clinically relevant? And can we reliably diagnose such conditions with our current tools?
These are not new questions. Since the dawn of modern psychology in the late 19th century, experts have wrestled with the complexities of mental health. Today, if you sought a diagnosis, you would likely meet a trained practitioner who combines their expertise with tools like psychometric tests and diagnostic criteria from manuals such as the Diagnostic and Statistical Manual of Mental Disorders (DSM) or the International Classification of Diseases (ICD). Yet, despite their widespread use, these tools are far from perfect. Diagnoses can vary depending on who makes them, when, and where. A single set of symptoms might lead to different diagnoses depending on the practitioner’s perspective or local standards. This means mental health conditions can be overdiagnosed, underdiagnosed, or even misdiagnosed, creating significant barriers to treatment and prolonging the suffering of the people affected.
Inconsistent diagnosis also undermines trust in healthcare systems, preventing people from seeking the help they need. Given that at least half the population will face a mental health issue at some point, the need for empirical, replicable, and objective measures is critical.
The Search for a Mental Health Biomarker
For decades, researchers have searched for a biomarker—an objective biological measure—to diagnose mental health conditions. Unlike physical health, where blood tests and scans can reveal underlying issues, mental health has resisted such quantifiable measures. There is no blood test for anxiety, nor, contrary to popular belief, is there a hormone that signals depression.
Fortunately, advances in neuroscience are offering new hope. Researchers have identified consistent patterns in brain activity that are associated with specific mental health conditions. Unlike traditional diagnostic methods, these patterns are not subject to the same cultural or social biases, as they remain consistent across populations and settings, allowing for greater replicability and reliability.
Neuroimaging: A Promising Tool
Despite these advances, neuroimaging—using techniques like MRI or CT scans to visualize brain activity—remains largely inaccessible to the average person. The machines are costly and complex, requiring highly trained professionals to operate. These machines are primarily reserved for life-threatening conditions, such as traumatic brain injuries, rather than for mental health issues like depression.
However, one form of neuroimaging holds particular promise: EEG (electroencephalography). First developed over a century ago, EEG measures the electrical activity of the brain through electrodes placed on the scalp. While EEG has long been used in clinical settings, its real potential is only now being realised thanks to artificial intelligence (AI). AI can sift through EEG data, identifying patterns linked to various mental health and neurological conditions.
In the last 15 years and especially the last 5, AI techniques have been used to identify the following psychiatric and neurological conditions from EEG data:
- Depression 1,2,3
- Anxiety disorders 2,3,4, including Obsessive-Compulsive Disorder (OCD) 3 and Post-Traumatic Stress Disorder (PTSD) 3,5
- Bipolar disorder 6
- Chronic insomnia 7
- Substance abuse disorder 3,8
- Schizophrenia 3,4
- Conditions tied to the Autism Spectrum (ASD) 9
- Alzheimer’s Disease 10
- Parkinson’s Disease 11
- Migraine prediction 12,13
- Epilepsy 14
- Brain strokes 15
A Future Where EEG Is Accessible to All
The ability of EEG to identify such a wide range of conditions is impressive, but its use has historically been limited by the high cost of EEG equipment. Thankfully, this is changing. EEG headsets are now increasingly available. However, these devices are either too expensive for the consumer or, in the case of consumer-grade devices, they often lack the signal quality or appropriate sensor placement necessary for this type of application.
At Dendron, we are pushing the boundaries further. We are developing an affordable EEG headset that allows the average consumer to monitor their brain activity from the comfort of their home—all without sacrificing the signal quality or ergonomics found in more expensive clinical and lab models. Our headset is designed to identify EEG patterns associated with various mental health conditions, providing users with valuable feedback to help improve their well-being. While not a replacement for clinical diagnosis, our goal is to make EEG a standard tool for healthcare professionals and individuals alike. And we’re excited to make it available to you soon.
By making EEG more accessible, we hope to empower people to take charge of their mental well-being and provide clinicians with an additional tool for understanding and treating mental health conditions.
References
[1] M. Ahmadlou, H. Adeli, and A. Adeli, “Fractality analysis of frontal brain in major depressive disorder,” International Journal of Psychophysiology, vol. 85, no. 2, pp. 206–211, 2012.
[2] L. R. Trambaiolli and C. E. Biazoli, “Resting-state global EEG connectivity predicts depression and anxiety severity,” in 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, pp. 3707–3710, 2020.
[3] S. M. Park, B. Jeong, D. Y. Oh, C. H. Choi, H. Y. Jung, J. Y. Lee, and D. Lee, “Identification of major psychiatric disorders from resting-state electroencephalography using a machine learning approach,” Frontiers in Psychiatry, vol. 12, 707581, 2021.
[4] A. Al-Ezzi, A. A. Al-Shargabi, F. Al-Shargie, and A. T. Zahary, “Complexity analysis of EEG in patients with social anxiety disorder using fuzzy entropy and machine learning techniques,” IEEE Access, vol. 10, pp. 39 926–39 938, 2022.
[5] A. Othmani, B. Brahem, Y. Haddou, and M. Khan, “Machine learning-based approaches for post-traumatic stress disorder diagnosis using video and EEG sensors: A review,” IEEE Sensors Journal. DOI: 10.1109/JSEN.2023.3312172, 2023.
[6] J. W. Kam, A. R. Bolbecker, B. F. O’Donnell, W. P. Hetrick, and C. A. Brenner, “Resting state EEG power and coherence abnormalities in bipolar disorder and schizophrenia,” Journal of psychiatric research, vol. 47, no. 12, pp. 1893–1901, 2013.
[7] M. Corsi-Cabrera, O. A. Rojas-Ramos, and Y. del Río-Portilla, “Waking EEG signs of non-restoring sleep in primary insomnia patients,” Clinical Neurophysiology, vol. 127, no. 3, pp. 1813–1821, 2016.
[8] Y. Liu, Y. Chen, G. Fraga-González, V. Szpak, J. Laverman, R. W. Wiers, and K. Richard Ridderinkhof, “Resting-state EEG, substance use and abstinence after chronic use: A systematic review,” Clinical EEG and Neuroscience, vol. 53, no. 4, pp. 344–366, 2022.
[9] M. McVoy, S. Lytle, E. Fulchiero, M. E. Aebi, O. Adeleye, and M. Sajatovic, “A systematic review of quantitative EEG as a possible biomarker in child psychiatric disorders,” Psychiatry research, vol. 279, pp. 331–344, 2019.
[10] A. Horvath, A. Szucs, G. Csukly, A. Sakovics, G. Stefanics, and A. Kamondi, “EEG and ERP biomarkers of Alzheimer’s disease: a critical review.” Frontiers in bioscience (Landmark edition), vol. 23, pp. 183–220, 2018.
[11] H. Railo, I. Suuronen, V. Kaasinen, M. Murtojärvi, T. Pahikkala, and A. Airola, “Resting state EEG as a biomarker of Parkinson’s disease: Influence of measurement conditions,” BioRxiv, pp. 2020–05, 2020.
[12] Z. Cao, K.-L. Lai, C.-T. Lin, C.-H. Chuang, C.-C. Chou, and S.-J. Wang, “Exploring resting-state EEG complexity before migraine attacks,” Cephalalgia, vol. 38, no. 7, pp. 1296–1306, 2018.
[13] Y. Li, G. Chen, J. Lv, L. Hou, Z. Dong, R. Wang, M. Su, and S. Yu, “Abnormalities in resting-state EEG microstates are a vulnerability marker of migraine,” The journal of headache and pain, vol. 23, no. 1, p. 45, 2022.
[14] S. Badani, S. Saha, A. Kumar, S. Chatterjee, and R. Bose, “Detection of epilepsy based on discrete wavelet transform and teager-kaiser energy operator,” in 2017 IEEE Calcutta Conference (CALCON). IEEE, 2017, pp. 164–167.
[15] M. Gottlibe, O. Rosen, B. Weller, A. Mahagney, N. Omar, A. Khuri, I. Srugo, and J. Genizi, “Stroke identification using a portable EEG device–a pilot study,” Neurophysiologie Clinique, vol. 50, no. 1, pp. 21–25, 2020.