Back

Causality analysis in major depressive disorder for early prediction of treatment outcomes with pharmacological and neuromodulation therapies

Bhattacharjee, M.; Vlachos, I.; Kathpalia, A.; Hlinka, J.; Brunovsky, M.; Bares, M.; Palus, M.

2025-05-31 health informatics
10.1101/2025.05.30.25328650
Show abstract

ObjectiveMajor depressive disorder affects millions of people globally, which lacks dependable biomarkers for the early identification of treatment success. This study examines two treatment modalities, pharmacological and neurostimulation. The aim is to discern alterations in brain connection patterns and direction of influence among various regions during the initial phase of the two treatment approaches. ApproachWe perform an information theory-based causality analysis on instantaneous phase time series data derived from electroen-cephalography recordings of 176 patients who underwent the aforementioned treatments. Each patient was recorded twice: prior to the commencement of treatment (visit 1) and one week after the initiation of treatment (visit 2). Main resultsTwo independent outcomes were observed. Initially, we discovered that different treatment modalities have different impacts on the brains connectivity and the direction of influence during the course of one week of treatment. The cohort administered with pharmacological agents exhibited a notable increase in both global and local information transmission in the brain within the {beta}2 (18Hz - 25.5Hz) frequency range, whereas the group subjected to stimulation exhibited a notable increase within the {delta} (1.5Hz - 3.5Hz) band. Secondly, potential respondents and nonrespondents to pharmacological therapy had distinct baseline brain causality effects at the initial session, with individuals who reacted to pharmacological therapy having considerably reduced information flow strength in the (8Hz - 12Hz) band compared to those who did not respond to treatment. SignificanceCausality analysis is conducted on the phases of the data, yielding more accurate conclusions by minimising the possible noise introduced by the signal amplitude. Our study provides significant insights into the directional influence inside the brain during depression and subsequent treatment.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

1
Frontiers in Psychiatry
based on 56 papers
Top 0.3%
12.4%
2
Journal of Affective Disorders
based on 72 papers
Top 1%
11.1%
3
Scientific Reports
based on 701 papers
Top 16%
10.1%
4
Frontiers in Neuroscience
based on 29 papers
Top 0.3%
6.3%
5
Progress in Neuro-Psychopharmacology and Biological Psychiatry
based on 10 papers
Top 0.1%
5.0%
6
NeuroImage: Clinical
based on 77 papers
Top 2%
5.0%
7
IEEE Journal of Biomedical and Health Informatics
based on 14 papers
Top 0.2%
4.7%
50% of probability mass above
8
PLOS ONE
based on 1737 papers
Top 73%
4.4%
9
Translational Psychiatry
based on 94 papers
Top 5%
2.8%
10
Chaos, Solitons & Fractals
based on 17 papers
Top 2%
2.8%
11
Neuroscience & Biobehavioral Reviews
based on 19 papers
Top 0.7%
2.8%
12
Journal of Medical Internet Research
based on 81 papers
Top 9%
1.6%
13
Frontiers in Aging Neuroscience
based on 22 papers
Top 2%
1.3%
14
Journal of Neural Engineering
based on 19 papers
Top 2%
1.3%
15
JMIR Formative Research
based on 31 papers
Top 4%
1.3%
16
Frontiers in Digital Health
based on 18 papers
Top 3%
1.3%
17
Patterns
based on 15 papers
Top 2%
1.2%
18
Computers in Biology and Medicine
based on 39 papers
Top 5%
1.2%
19
Frontiers in Physiology
based on 18 papers
Top 3%
0.8%
20
BMC Medical Research Methodology
based on 41 papers
Top 5%
0.8%
21
Frontiers in Artificial Intelligence
based on 11 papers
Top 3%
0.7%
22
Brain Stimulation
based on 27 papers
Top 3%
0.7%
23
Journal of the American Medical Informatics Association
based on 53 papers
Top 7%
0.7%