Evaluation of the role of stress in patients with breast cancer and depression by paykel's life event and adaptive neuro-fuzzy approach.

Abstract Introduction The aim of this study was to identify and analyze the stress factors and the level of stress a year preceding the onset of breast cancer and depression in the studied female patients. Methods The research in this work was mostly prospectively (clinical and analytical). During the research, Scale of Life Events‐Paykel was applied. Stressful life events differed significantly between the groups. In the studied group, the following events were significantly more prevalent: partner infidelity (χ 2 = 12.663; p < .001), failure at work (χ 2 = 44.429; p < .001), and spontaneous abortions or stillbirths (χ 2 = 13.818; p < .001). Results According to the results of this study, stressful life events differed significantly between the observed groups. These stressful life events had a significant impact on the increase of risk for breast cancer, as well as on depressivity. Afterward, adaptive neuro‐fuzzy inference system was used for prediction of the Paykel's Life Event according to Fisher's exact test. Conclusion The obtained results could be of practical usage for improving stress behavior of the patients with breast cancer and depression.


| INTRODUC TI ON
Stress is a characteristic problem of modern human society, and as such, it should be highly prioritized in the sphere of medical research (Picardi & Gaetano, 2014). Inability to cope successfully with stressful life events leads to maladaptive responses and susceptibility to stress-associated pathologies.
Our reaction to an acute stress is almost invariably tumultuous.
If stressful situations come one after another, our organism does not have enough time to recover adequately (Biondi & Picardi, 1999).
Chronic stress is the result of poor adaptive response to a sequence of unpleasant and upsetting long-term life situations which tend to exhaust human defense mechanisms. Individuals with insufficient (deficient) defense mechanisms in stressful situations tend to react with 2.1 | ANFIS methodology ANFIS network has five layers as shown in Figure 1. The main core of the ANFIS network is fuzzy inference system. Layer 1 receives the inputs and converts them in the fuzzy value by membership functions.
Second layer multiplies the fuzzy signals from the first layer and provides the firing strength of rule. The third layer is the rule layers where all signals from the second layer are normalized. The fourth layer provides the inference of rules, and all signals are converted in crisp values. The final layers summarized the all signals and provided the output crisp value. Among those with breast cancer, significantly more prevalent were spouse infidelity (χ 2 = 12.663; p < .001), failure at work (χ 2 = 44.429; p < .001), and spontaneous abortions or stillbirths (χ 2 = 13.818; p < .001).

| RE SULTS
In the control group of women, the following life events were significantly more prevalent: death of a close family member (χ 2 = 36.959; p < .001), serious financial difficulties (χ 2 = 57.994; p < .001), hospitalization of a family member (χ 2 = 17.389; p < .001), death of a close friend (χ 2 = 6.872; p < .001), frequent marital disputes (χ 2 = 8.461; p = .004), disputes with an extended family member or a colleague from work (χ 2 = 10.648; p = .001), less serious (mild) organic disease (χ 2 = 4.175; p = .041), and marriage of a child with parental consent (χ 2 = 4.175; p = .041).   The results demonstrated that women who frequently or continually experienced stress had a risk for breast cancer similar to that in women who never experienced any stress or experienced it only occasionally.

F I G U R E 1 ANFIS layers
The analysis showed that the risk for breast cancer was slightly higher in women who had lost their mothers early in life, but not if they had lost their fathers. The Center for Chronic Disease Biology studied the issue and showed that stress increased cortisol binding by the glucocorticoid receptor (GR).
According to the results of this study, stressful life events were different between the groups. In the studied group of women, significantly more prevalent were spouse infidelity (χ 2 = 12.663; There are great individual differences in response even to the same stressors. The authors in one paper (Vlajković, 1992) justly emphasized that stress was a highly personalized process, that is the process highly dependent on personality traits.
Numerous investigations have documented that the status of our psyche has an impact on our health. Cancer occurs only when a human defense system is overpowered and cannot cope with various threats. In a cancer treatment, a significant accent is put on human will, wish, and decision to undergo treatment (Lazarus, 1991).
Our immune system is suppressed by various psychic factors, out of which the most important are long-lasting grief, feeling of failure in life, anger, anxiety, and stress. Chronic exposure to conflict situations, especially when there is a conflict between our needs, wishes and abilities, creates chronic frustration. When it can no longer be dealt with using the usual defense mechanisms (rationalization, projections, sublimation), a physical disease occurs. Cancer most commonly affects depressive individuals, too rational, afraid of their emotions, with poor fantasies, psychologically restrained and passive (Polansky, 2003).
Numerous studies have shown that stressogenic life events have an impact on the course of depression as well, the quality of remission, and frequency of relapses (Haberstick et al., 2016;Polansky & Javaherian, 2015).
However, the etiological relevance of these facts has not yet been sufficiently elucidated (Paykel, Prusoff, & Uhlenhuth, 1971). For instance, although most depressive patients experience stressful life events before the episodes of depression, only a small portion of individuals exposed to such stressors become actually depressive.
F I G U R E 2 ANFIS relationship between Fisher's exact test (input 1) and Paykel's Life Event (output)