+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Machine-learning-based classification between post-traumatic stress disorder and major depressive disorder using P300 features

      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.



          The development of optimal classification criteria for specific mental disorders which share similar symptoms is an important issue for precise diagnosis. We investigated whether P300 features in both sensor-level and source-level could be effectively used to classify post-traumatic stress disorder (PTSD) and major depressive disorder (MDD).


          EEG signals were recorded from fifty-one PTSD patients, 67 MDD patients, and 39 healthy controls (HCs) while performing an auditory oddball task. Amplitude and latency of P300 were evaluated, and the current source analysis of P300 components was conducted using sLORETA. Finally, we classified two groups using machine-learning methods with both sensor- and source-level features. Moreover, we checked the comorbidity effects using the same approaches (PTSD-mono diagnosis (PTSDm, n = 28) and PTSD-comorbid diagnosis (PTSDc, n = 23)).


          PTSD showed significantly reduced P300 amplitudes and prolonged latency compared to HCs and MDD. Moreover, PTSD showed significantly reduced source activities, and the source activities were significantly correlated with symptoms of depression and anxiety. Also, the best classification accuracy at each pair was as follows: 80.00% (PTSD-HCs), 67.92% (MDD-HCs), 70.34% (PTSD-MDD), 82.09% (PTSDm-HCs), 71.58% (PTSDm-MDD), 82.56% (PTSDc-HCs), and 76.67% (PTSDc- MDD).


          Since abnormal P300 reflects pathophysiological characteristics of PTSD, PTSD patients were well-discriminated from MDD and HCs when using P300 features. Thus, altered P300 characteristics in both sensor- and source-level may be useful biomarkers to diagnosis PTSD.


          • Post-traumatic stress disorder (PTSD) showed altered P300 characteristics in both sensor- and source-level compared to major depressive disorder (MDD) and healthy controls (HCs).
          • PTSD-comorbid diagnosis (PTSDc) group showed significantly altered P300 characteristics in both sensor- and source-level compared to MDD, whereas PTSD-mono diagnosis (PTSDm) showed a significant difference only in P300 latency (sensor-level).
          • The acceptable precision and recall values were obtained when discriminating PTSD from MDD and HCs using P300 features in both sensor- and source-level with machine-learning algorism, demonstrating the versatility and suitability of our approach.

          Related collections

          Most cited references 58

          • Record: found
          • Abstract: found
          • Article: not found

          Updating P300: an integrative theory of P3a and P3b.

           John Polich (2007)
          The empirical and theoretical development of the P300 event-related brain potential (ERP) is reviewed by considering factors that contribute to its amplitude, latency, and general characteristics. The neuropsychological origins of the P3a and P3b subcomponents are detailed, and how target/standard discrimination difficulty modulates scalp topography is discussed. The neural loci of P3a and P3b generation are outlined, and a cognitive model is proffered: P3a originates from stimulus-driven frontal attention mechanisms during task processing, whereas P3b originates from temporal-parietal activity associated with attention and appears related to subsequent memory processing. Neurotransmitter actions associating P3a to frontal/dopaminergic and P3b to parietal/norepinephrine pathways are highlighted. Neuroinhibition is suggested as an overarching theoretical mechanism for P300, which is elicited when stimulus detection engages memory operations.
            • Record: found
            • Abstract: found
            • Article: not found

            Event-related EEG/MEG synchronization and desynchronization: basic principles.

            An internally or externally paced event results not only in the generation of an event-related potential (ERP) but also in a change in the ongoing EEG/MEG in form of an event-related desynchronization (ERD) or event-related synchronization (ERS). The ERP on the one side and the ERD/ERS on the other side are different responses of neuronal structures in the brain. While the former is phase-locked, the latter is not phase-locked to the event. The most important difference between both phenomena is that the ERD/ERS is highly frequency band-specific, whereby either the same or different locations on the scalp can display ERD and ERS simultaneously. Quantification of ERD/ERS in time and space is demonstrated on data from a number of movement experiments.
              • Record: found
              • Abstract: found
              • Article: not found

              Identifying true brain interaction from EEG data using the imaginary part of coherency.

              The main obstacle in interpreting EEG/MEG data in terms of brain connectivity is the fact that because of volume conduction, the activity of a single brain source can be observed in many channels. Here, we present an approach which is insensitive to false connectivity arising from volume conduction. We show that the (complex) coherency of non-interacting sources is necessarily real and, hence, the imaginary part of coherency provides an excellent candidate to study brain interactions. Although the usual magnitude and phase of coherency contain the same information as the real and imaginary parts, we argue that the Cartesian representation is far superior for studying brain interactions. The method is demonstrated for EEG measurements of voluntary finger movement. We found: (a) from 5 s before to movement onset a relatively weak interaction around 20 Hz between left and right motor areas where the contralateral side leads the ipsilateral side; and (b) approximately 2-4 s after movement, a stronger interaction also at 20 Hz in the opposite direction. It is possible to reliably detect brain interaction during movement from EEG data. The method allows unambiguous detection of brain interaction from rhythmic EEG/MEG data.

                Author and article information

                Neuroimage Clin
                Neuroimage Clin
                NeuroImage : Clinical
                05 September 2019
                05 September 2019
                : 24
                [a ]Department of Biomedical Sciences, University of Missouri, Kansas City, USA
                [b ]Clinical Emotion and Cognition Research Laboratory, Goyang, Republic of Korea
                [c ]Department of Psychology, Chung-Ang University, Seoul, Republic of Korea
                [d ]Department of Biomedical Engineering, Hanyang University, Seoul, Republic of Korea
                [e ]Psychiatry Department, Ilsan Paik Hospital, Inje University, Goyang, Republic of Korea
                Author notes
                [* ]Corresponding author at: 170, Juhwa-ro, Ilsanseo-gu, Goyang-si, Gyeonggi-do 10380, Republic of Korea. lshpss@
                S2213-1582(19)30351-1 102001
                © 2019 Published by Elsevier Inc.

                This is an open access article under the CC BY-NC-ND license (

                Regular Article


                Comment on this article