Regression is a technique for relieving past stress and achieving inner peace. It is a way to reach your inner deep peace and heal past traumas. Anyone can do this!
The term "regression" refers to the process of returning to an earlier state, and for regression therapy this means going back to the source of the psychological problem. Regression therapy uses a range of hypnotherapy and psychotherapeutic techniques, focusing on regression and working with transpersonal experiences to transform them.
Linear regression
Regression is a technique used to model and analyze the relationships between variables and to see how those variables work together to produce a particular outcome. Linear regression refers to a type of regression model that consists of interrelated variables. Let's start with something simple. Paired (simple) linear regression is a model that allows you to model the relationship between the values of one input independent variable and one output dependent variable using a linear model, such as a straight line.
A more common model is multiple linear regression, which involves establishing a linear relationship between many input independent variables and one output dependent variable. Such a model remains linear because the output is a linear combination of the input variables. We can build a multiple linear regression model as follows:
Y = a_1*X_1 + a_2*X_2 + a_3*X_3 ……. a_n*X_n + b
Where a_n are the coefficients, X_n are the variables and b is the bias. As we can see, this function does not contain nonlinear coefficients and, thus, is only suitable for modeling linear separable data. It's very simple: we weight the value of each variable X_n using the weighting factor a_n. These weights a_n as well as the bias b are calculated using stochastic gradient descent. Look at the graph below as an illustration!
A few important points about linear regression:
- It is easy to model and is especially useful when creating not very complex dependencies or when there is a small amount of data.
- The designations are intuitive.
- Sensitive to emissions.
Can fibroids resolve?
Sometimes the patient is already undergoing examinations for hospitalization for the removal of a fibroid that has grown to a large size, the last preoperative ultrasound is performed, and an unexpected verdict is made about the absence of pathology.
The reasons for this condition have been reflected in a number of Italian studies. Summarizing the statistics, gynecologists of the city of Padua came to a relationship between the formation of myomatous nodes and their disappearance. The most common cases of fibroid resorption were identified:
- the period of appearance in adolescence and complete withdrawal after the onset of active sexual life - 5% of the total number of cases. At this point, one should take into account the relatively late onset of sexual activity for Mediterranean women, the lower level of which begins at 21 years of age;
- a neoplasm that arose as a result of the termination of one pregnancy and its disappearance in the next one – 30%. This depends on the climate of the zone in question, since the higher the external temperature, the more intense the effect hormones have on the body;
- self-resorption of the tumor after childbirth occurred in 50% of cases;
- fibroids that arose as a result of suddenly interrupted lactation disappeared when it resumed – 14% of cases;
- resorption as a result of the onset of menopause - 40% of cases;
- disappearance of a tumor-like node due to the disappearance of a third-party factor, such as the influence of climate or a stress component – 18%
- There are also sad statistics when the disease disappeared as a result of the appearance of oncology of another endocrine gland - 0.8% of cases.
The relationship between the disappearance of nodes and the factors that once provoked their formation is obvious. Undoubtedly, 99.9% are hormonal changes when fibroids disappear when the effect on female reproductive function changes.
Polynomial regression
Polynomial regression can be used to create such a model that is suitable for non-linearly separable data. In this method, a curved line is drawn depending on the points of the plane. In polynomial regression, some of the independent variables have degrees greater than 1. For example, you'd get something like this:
Y = a_1*X_1 + (a_2)²*X_2 + (a_3)⁴*X_3 ……. a_n*X_n + b
Some variables have degrees, others do not. It is also possible to select a specific power for each variable, but this requires some knowledge of how the inputs relate to the outputs. Compare linear and polynomial regression below.
A few important points about polynomial regression:
- Models non-linearly separated data (which linear regression cannot). It is more flexible and can model complex relationships.
- Full control over the modeling of object variables (selection of degree).
- The model must be created carefully. It is necessary to have some knowledge about data in order to select the most suitable degree.
- If the degree is chosen incorrectly, this model may be oversaturated.
Can fibroids resolve?
Sometimes the patient is already undergoing examinations for hospitalization for the removal of a fibroid that has grown to a large size, the last preoperative ultrasound is performed, and an unexpected verdict is made about the absence of pathology.
The reasons for this condition have been reflected in a number of Italian studies. Summarizing the statistics, gynecologists of the city of Padua came to a relationship between the formation of myomatous nodes and their disappearance. The most common cases of fibroid resorption were identified:
- the period of appearance in adolescence and complete withdrawal after the onset of active sexual life - 5% of the total number of cases. At this point, one should take into account the relatively late onset of sexual activity for Mediterranean women, the lower level of which begins at 21 years of age;
- a neoplasm that arose as a result of the termination of one pregnancy and its disappearance in the next one – 30%. This depends on the climate of the zone in question, since the higher the external temperature, the more intense the effect hormones have on the body;
- self-resorption of the tumor after childbirth occurred in 50% of cases;
- fibroids that arose as a result of suddenly interrupted lactation disappeared when it resumed – 14% of cases;
- resorption as a result of the onset of menopause - 40% of cases;
- disappearance of a tumor-like node due to the disappearance of a third-party factor, such as the influence of climate or a stress component – 18%
- There are also sad statistics when the disease disappeared as a result of the appearance of oncology of another endocrine gland - 0.8% of cases.
Ridge regression
In case of high collinearity of variables, standard linear and polynomial regression become ineffective. Collinearity is a relationship between independent variables that is close to linear. The presence of high collinearity can be determined in several ways:
- The regression coefficient is not important, even though, theoretically, the variable should have a high correlation with Y.
- When you add or remove a variable from the X matrix, the regression coefficient changes greatly.
- The X matrix variables have high pairwise correlations (look at the correlation matrix).
You can first look at the standard linear regression optimization function to better understand how ridge regression can help:
min || Xw - y ||²
Where X is a matrix of variables, w is weights, y is reliable data. Ridge regression is an adjustment measure to reduce collinearity among predictor variables in a regression model. Collinearity is a phenomenon in which one variable in a multiple regression model can be predicted linearly from the other properties with a significant degree of accuracy. Thus, due to the high correlation of the variables, the final regression model is reduced to the minimum limits of the approximate value, that is, it has high variance.
Ridge regression adds a small quadratic bias factor to reduce variance:
min || Xw - y ||² + z|| w ||²
This bias factor releases the coefficients of the variables from strict constraints, introducing a small bias into the model but significantly reducing the variance.
A few important points about ridge regression:
- The assumptions of this regression are the same as those of the least squares method, except for the fact that a normal distribution is not assumed in ridge regression.
- This reduces the significance of the coefficients, leaving them non-zero, suggesting no feature selection.
Causes of regression of myomatous nodes
Scientists have identified several theories regarding the development of the disease. One of them is hormonal imbalance. In this case, the amount of estrogen, which leads to rapid and pronounced growth of the node, significantly exceeds the level of progesterone. Education is affected by damage to the uterus due to inflammation, curettage, and surgery. At the site of damage, growth hormone is released, which changes hormonal receptors, increasing their consumption of the hormone estrogen. Thus, women with a preserved menstrual cycle are susceptible to the formation of fibroids.
Reducing the size of uterine fibroids and its regression can be achieved by reducing the level of estrogen and increasing gestagens in the blood. There are the following reasons for tumor regression:
- The onset of menopause in a woman.
- Correction of hormonal balance with an increase in the amount of gestagens.
- Pregnancy and lactation period.
- Stopping the nutrition of fibroids.
Hormonal treatment can not only slow down the growth of the tumor, but also reduce its size. Stopping the medication leads to the return of fibroids to their original size. During menopause, a gradual regression of the node is noted, up to complete disappearance.
Myoma in menopause
The development of menopause in a woman begins after 45 years, although it depends on many factors, one of which is heredity. During this period, the ovaries stop producing the hormone estrogen, which is the main cause of the appearance of the tumor. During this period, not only the tissue of the uterus undergoes changes, but also other organs sensitive to estrogen.
Myoma regresses during menopause gradually, over 3 to 5 years, until the onset of the postmenopausal period.
At the same time, blood circulation in the reproductive organ decreases, and, accordingly, in education. During the first three years, strict control over the size of the node is necessary to ensure its regression. The patient should regularly visit the antenatal clinic and undergo ultrasound diagnostics. In case of complaints or bleeding, you should consult a doctor: this may indicate an imbalance of hormones and progression of the disease.
Regression of fibroids and intrauterine device
A popular method of treating fibroids has become the hormonal intrauterine device. It is based on the release of a hormone - gestagen (levonorgestrel), which has an anti-estrogenic effect and leads to regression of the node. The drug is used as hormone replacement therapy and is an analogue of oral contraceptives.
The spiral is immersed in the uterine cavity, and even a small dose acts directly on the fibroids, leading to regression. At the same time, the level of the drug in the blood is significantly lower than when taking oral medications and side effects are less likely to be observed. The intrauterine device acts directly on the tumor, blocking the sensitivity of estrogen receptors. The work of the ovaries is not disrupted, the appearance and development of follicles continues.
Indications for the use of the progesterone IUD are:
- The presence of several fibroids up to 2 cm in size.
- Heavy, prolonged menstruation.
- The woman has other gynecological diseases: endometriosis, cystic formations.
The installation of the spiral is carried out by a doctor under sterile conditions. This type of therapy has a number of disadvantages and contraindications:
- Individual sensitivity.
- Malignant tumors.
- Infectious diseases of the genital organs.
- Liver diseases.
- Pregnancy.
- Polycystic ovary syndrome.
The use of an intrauterine device can lead to chronic inflammation, as it is a foreign body.
Aseptic conditions during installation lead to infection. In some cases, patients report spontaneous loss of the IUD.
Regression with hormone therapy
In order to achieve lasting regression, hormone therapy is used. Treatment with hormones to reduce tumor size involves reducing the level of estrogen in relation to gestagens. Conservative treatment with hormonal drugs is indicated:
- If it is not possible to perform surgical removal.
- If the nodes are in the submucosal area.
- If the woman does not experience intense bleeding.
- With malignancy (malignancy) of the tumor.
To transfer fibroids to the regression stage, not only sex hormones are used, but also a number of other means. Hormone therapy drugs:
- Oral contraceptives with low and high doses.
- Progestogens, including intrauterine device.
- Gonadotropin-releasing hormone agonists.
- Progesterone receptor modulators: Progesterone stimulants (progestins and progesterone).
- Progesterone receptor blockers (mifepristone).
- Mixed-action drugs.
Therapy is contraindicated in case of liver diseases, thrombosis, oncological processes, as well as in case of a negative reaction of the body to therapy.
Hormonal drugs are used in injection and tablet forms. The duration of use is limited to 6 months; after discontinuation, myomatous nodes return to their original size and continue to increase. Complete recovery occurs only after the tumor is removed.
Regression during arterial embolization
Good results in therapy are achieved using the method of embolization of the uterine arteries located close to the tumor. This type of treatment allows you to slow down not only its growth, but also achieve a state of regression. Many patients, wanting to preserve the reproductive organ, refuse hormonal or surgical treatment, and the embolization method becomes a good alternative. This method has the following advantages:
- Minimal invasiveness.
- Local anesthesia.
- The effectiveness of treatment, with a decrease in symptoms of the disease.
- Preservation of reproductive function.
Preparation for arterial embolization includes a full list of tests and studies, as before any surgical operation.
The technique involves introducing microparticles (emboli) into the arteries through which the uterus is supplied with blood, causing their occlusion (blockage). The blood supply to the reproductive organ continues through the ovarian arteries. Myoma regresses within 6 months and decreases in size.
The embolization method has a number of contraindications:
- Large tumors, over 25 weeks of pregnancy.
- Pedicled nodules or submucosal fibroids larger than 5 cm.
- Pregnancy.
- Acute inflammatory processes in the pelvis.
- Individual intolerance to contrast media.
- Suspicion or presence of a tumor of the pelvic organs.
After treatment, there is a decrease in tumor size by 50%, their softening, change in location or loss.
Regression during pregnancy
Fibroids can be detected in a woman of any age, regardless of the number of pregnancies. This is due to the development of the disease in women of childbearing age and lack of preparation for pregnancy. Detection of the disease during pregnancy is not an indication for its termination, but can complicate the course of pregnancy and childbirth.
Numerous studies show that in the first 10 weeks of pregnancy, myomatous nodes can increase in size. Subsequently, with an increase in hormones - gestagens and a decrease in estrogens, a natural regression occurs. Unfortunately, this phenomenon is temporary and can last in the postpartum period while the woman is breastfeeding. After the resumption of menstruation, the nodes return to their original appearance.
Almost all methods, except arterial embolization, lead to temporary regression of fibroids. An effective treatment method is myomectomy. It involves removing only the nodes and preserving the woman’s reproductive function.
Lasso regression
In lasso regression, as in ridge regression, we add an offset condition to the optimization function in order to reduce collinearity and therefore model variance. But instead of quadratic offset, we use absolute value offset:
min || Xw - y ||² + z|| w ||
There are several differences between ridge regression and lasso that restore differences in the properties of L2 and L1 regularizations:
- Built-in feature selection is considered a useful feature that is present in the L1 norm but absent in the L2 norm. Feature selection results from the L1 norm, which produces sparse coefficients. For example, suppose a model has 100 coefficients, but only 10 of them have coefficients other than zero. Accordingly, “the remaining 90 predictors are useless in predicting the desired value.” The L2 norm produces undischarged coefficients and cannot perform feature selection. Thus, lasso regression can be said to perform "parameter selection" since variables not selected will have a total weight of 0.
- Sparseness means that a small amount of the input data in a matrix (or vector) has a value other than zero. The L1 norm produces a large number of coefficients with zero value or very small values with some large coefficients. This is related to the previous point, which stated that the lasso performs property selection.
- Computational efficiency: the L1 norm does not have an analytical solution, unlike the L2 norm. This allows efficient computation of L2 norm solutions. However, L1 norm solutions do not have sparse properties, which allows them to be used with sparse algorithms for more efficient calculations.
Uterine moma in regression stage
Myoma is detected in the later stages, when the node reaches a large size and clinical symptoms appear. Women complain of menstrual dysfunction, chronic pain in the lower abdomen or lower back, prolonged and heavy bleeding during menstruation, leading to severe anemia. Large fibroids increase the volume of the abdomen, make urination difficult, and lead to a feeling of incomplete emptying of the bladder. Patients report difficulty defecating and constipation. Women's reproductive function is impaired and they cannot become pregnant for a long time.
The next step, after diagnosis, is the choice of treatment method. In this case, the patient wants to avoid surgical treatment and use drug therapy in order to reduce the size of the tumor and bring it into regression.
Uterine fibroids can regress (shrink) not only with the use of special medications, but also for natural reasons.
When tumor regression occurs:
- Menopause period.
- Postmenopause.
- Pregnancy.
- Taking hormonal drugs.
- Use of antitumor drugs in the treatment.
The doctor may consider it necessary to use drug treatment for a patient with fibroids in case of imminent menopause. This will make it possible to convert the drug-induced regression of the tumor into a natural one and avoid surgery.
Elastic Net Regression
Elastic net is a hybrid of lasso and ridge regression methods. It uses both L1 and L2 regularization, considering the efficiency of both methods.
min || Xw - y ||² + z_1|| w || + z_2|| w ||²
A practical advantage of using lasso regression and ridge regression is that it allows the elastic net to inherit some of the stability of ridge regression when rotated.
A few important points about elastic net regression:
- It creates the conditions for a group effect when variables are highly correlated, rather than setting some of them to zero, like the lasso method.
- There are no restrictions on the number of selected variables.
Uterine fibroids: risk factors, symptoms, types of treatment
Uterine fibroids are a benign nodular formation that occurs in its muscle layer and affects immature cells. Today, this disease is considered one of the most common, since according to statistics, it affects up to 28% of women of childbearing age and up to 42% of women in perimenopause.
Myoma, endometriosis, endometrial hyperplastic processes are hormone-dependent diseases that occur against the background of disruption of the hormonal regulation of the menstrual cycle.
The tumor can occur in various parts of the uterus, which is what the classification of fibroids is based on: intramural (in the thickness of the muscle), submucosal and subperitoneal - however, the disease often manifests itself in several types at once, and the nodes are characterized by different locations and sizes.
