Integrating Psychosocial Factors into Artificial Intelligence Models for Predicting Addiction Treatment Outcomes: A Systematic Review.

Depression is a major mental disorder with long-term health consequences, being one of the leading sources of reduced quality of life and lost productivity worldwide []. According to current estimates, up to 5% of adults suffer from depressive disorders []. Typical symptoms of depression include low mood, anhedonia, sleep and appetite disturbances, fatigue, and impaired concentration [, ]. Due to the absence of validated biological markers, a diagnosis is established through clinical assessment (history and mental status examination). Whilst response to pharmacotherapy is variable. Risk factors for the development of depression include both genetic and environmental influences, such as a family history of the disorder, adverse childhood experiences, and stressful life events [, ]. Given the diverse causes of depression, several hypotheses have been proposed to explain its pathophysiology, including the widely accepted neurotransmitter and receptor hypothesis, among others []. The most prominent is the monoamine hypothesis, which suggests that depression results from the depletion of brain neurotransmitters such as serotonin and noradrenaline []. This hypothesis guided pharmacotherapy, with selective serotonin reuptake inhibitors (SSRIs) or noradrenaline reuptake inhibitors(NRIs) as the mainstay of treatment. Such drugs act to increase extracellular monoamine levels [, , ]. However, monoamine-based treatments have several limitations, including a delayed onset [] and unwanted side effects [], prompting research into non-monoaminergic mechanisms. In particular, targeting glutamatergic and GABAergic neurotransmission has been prioritized. Recent evidence indicates that disturbances in glutamatergic and GABAergic neurotransmission play a key role in the pathophysiology of depressive disorders [–]. Within the glutamatergic system, N-methyl-D-aspartate (NMDA) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors are central nodes linking glutamate signaling to rapid changes in synaptic efficacy and plasticity []. AMPA receptors are mainly responsible for fast excitatory transmission in the cerebral cortex and hippocampus [], whereas NMDA receptors act as voltage-dependent coincidence detectors and are crucial for neuroplasticity []. In contrast to NMDA receptors, which predominantly initiate plasticity-related signaling pathways, AMPA receptors directly determine synaptic strength by regulating postsynaptic function through receptor trafficking, thereby shaping functional neural connectivity [, ].