Introduction: Cancer is the main public health problem in the world and is already among the main causes of death. Breast CA will be the second most common type of cancer in the Brazilian population between 2020-2022. The pandemic caused by COVID-19 brought several changes in society, affecting health services, one of which was the need to maintain regular treatment for various health conditions, such as cancer. Physical therapy is an essential part of the multidisciplinary treatment of cancer patients. With the new technologies available, it is possible to allow continuous treatment of cancer patients, even during quarantine, and one possibility is the application of information and communication technologies to provide rehabilitation therapy to remote people, called home telerehabilitation. Objective: To verify the effect of virtual reality (VR) training on motor performance, pain and heart rate variability variables in patients diagnosed with breast cancer. Methods: The study will be composed of 30 women over 18 years old with a diagnosis of breast cancer attended at an oncology physical therapy clinic in the city of Juiz de Fora - MG. Participants will undergo training using virtual reality technology, always guided by the researcher. There will be a total of 10 (ten) interventions, with two training sessions per week. The participant will be guided by a researcher who will instruct her online. Five games will be used: Reaction Time, Coincident Timing, Fitts, Genius, MoveHero. Participants will answer some questionnaires and scales such as EORTC QLQ-C30 which assesses quality of life in cancer patients, perceived exertion scale will be applied before and between matches, a mood scale (BRUMS) will be applied at the beginning and at the end of training and Affective Analog Face Scale for pain assessment and analysis of Heart Rate Variability (HRV) that will be performed before, during and after the first and last training, through a heart rate monitor that will be given to each participant and an application cell phone (Elite HRV).

METHOD

1. Study design and location This is a longitudinal study of the randomized clinical trial type. This study will be developed at the Hope Physiotherapy and Esthetics Clinic, headquartered and located in the city of Juiz de Fora, in the state of Minas Gerais.

2. Participants Thirty individuals will participate in the study with a diagnosis of breast CA, previously confirmed by a specialist physician, who attend physiotherapy care at the Hope Physiotherapy and Esthetics clinic in the city of Juiz de Fora-MG, authorized by the owner (ANNEX I).

   To calculate the sample size, an alpha value (probability of error type 1) of 5% was assumed for the two-tailed hypothesis, beta (probability of error type 2) of 20%, test power of 80% and a difference between groups of 10% referring to engine performance data. Thus, at least 30 individuals will be needed (Dodd et al., 2002).

   All participants will be informed about the objectives and procedures of the study and, when they agree, will sign the Informed Consent Form (FICF). The study will be submitted to the Research Ethics Committee (CEP). Resolution 466/2012 of the National Health Council of 10/10/1996 which regulates research involving human beings and the Declaration of Helsinki (1964) will be respected. Data will be electronically stored in databases with restricted and secure access. All data will be encrypted with removal of any information that could identify individuals.

3. Procedures Thirty women diagnosed with breast AC will be selected to analyze the effect on motor performance, pain, perceived exertion and HRV in interventions through VR.

   Ten sessions will be applied (two sessions per week). Initially, the researcher will travel to the participant's residence with all the protective equipment necessary for the delivery of the heart rate capture straps, which will be inside a plastic bag disinfected with 70% alcohol and also to present the consent form and clarified to the participants, as well as the procedures and objectives of the study. After signing the consent form, the following information will be collected through online forms: age, gender, corresponding phase (such as chemotherapy, initial diagnosis, complications, relapse diagnosis), length of stay, associated diseases, medications in use, education . Then, the questionnaires will be applied to the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30), Brunel Mood Scale (BRUMS), BORG scale of perceived exertion, and Analog Affective Faces Scale.

   Then, training with virtual reality will be carried out, in which the HRV will be captured (on days 1, 5 and 7) 10 minutes before performing the tasks in a virtual reality environment (at rest, sitting), for 15 minutes during the tasks in VR (MoveHero and Genius, according to item 5.4.2. D and E) and for another 10 minutes after (at rest, sitting), to assess the autonomic recovery of physical activity. In the 10 days, interventions in VR will be used (MoveHero and Genius and the Borg scale), the HRV and the other evaluations will be carried out on days 1, 5 and 10.

   3.1. Physical Assessment Questionnaires and Tests A) European Organization for Research and Treatment of Cancer Core Quality of Life Questionnaire (EORTC QLQ-C30) The general questionnaire (EORTC QLQ-C30) includes 30 questions related to five functional scales (physical, functional, emotional, social and cognitive), a global health status scale, three symptom scales (fatigue, pain and nausea/vomiting ) and six additional symptom items (dyspnea, insomnia, loss of appetite, constipation, diarrhea and financial difficulties).

   The score is obtained according to the type of answer chosen. The options allowed by the questionnaire are "no" (one point), "little" (two points), "moderately" (three points) or "very" (four points). In the two questions referring to the global health status scale, the options to choose from range from one (very bad) to seven points (very good). Questionnaire scores range from 0 to 100.

   Regarding functional and global health status scales, higher scores are related to better quality of life; however, for the symptom scales, higher scores correspond to a greater presence of that symptom and, consequently, a worse quality of life (Appendix I).

   B) Affective Analog Scale of Faces (FAS):

   This scale assesses the affective magnitude of pain through 9 facial expressions that range from the feeling of joy to the feeling of sadness, which makes it possible to measure the absence of pain to a higher degree of pain (McGrath, 1996).

   Facial expressions are sequentially ordered by the letter interval (AI), in which the AD interval represents the absence of pain, E represents the neutral point and the FI interval represents different degrees of pain intensities, with F being the easily ignored pain indication by the presence of family members, favorable environment, measures of comfort, safety and understanding of the professional; G an intensity of pain that does not interfere with behavior and can be mediated by the child's expression; There is an intensity of pain that interferes with behavior and requires the use of analgesics, and I an intensity in the maximum degree of pain that requires greater intervention by the team, observation of physical and psychological aspects, in addition to appropriate analgesic techniques (Torritesi et al., 1998).

   Thus, individuals choose among the 9 faces, ordered in a scale from A to I, the one that most indicates the intensity of their pain.

   C) Borg's perceived exertion scale:

   It is the classification of perceived exertion used to measure the subjective intensity of exertion. It is based on sensations experienced during exercise, such as muscle fatigue, increased heart rate, and increased breathing (Andrews et al., 2013). It is often used as an alternative to heart rate to measure the intensity of physical activity (BORG, 1970). The Borg scale has a linear relationship with oxygen uptake (BORG, 1982), in addition to being a measure to determine exercise intensity, while evaluating the overall physical and mental effort perceived by the individual on physical activity (BORG, 1998).

   D) Brunel Mood Scale (BRUMS):

   It measures six identifiable affective states (tension, depression, anger, energy, fatigue and confusion.

   use) through a 24-item self-report inventory, with respondents rating a list of adjectives on a 5-point Likert scale from 0 (none) to 4 (extremely) based on subjective feelings. Instructions refer to how participants felt last week, including today's date (Van Wijk; Martin; Hans-Arendse, 2013).

   3.2. Virtual Reality Training Participants will carry out the tasks individually in their homes with the help of a computer on a table and with the online presence of the evaluator responsible for providing instructions and recording the results. Height and distance adjustments will be made according to the individual's needs.

   Before starting the task, the researcher will explain the task verbally and a video will be sent with a demonstration of how the game works. Participants will then complete a single trial to verify that they understand the instructions.

   The following virtual games will be used: Reaction Time (3 adaptation attempts and 10 attempts for analysis), Coincident Timing (20 repetitions for motor skill acquisition and performance improvement), Coincident Timing (5 repetitions + 5 repetitions with level increase of speed for analysis of retention and transfer - motor learning), 5 repetitions for the transfer phase), Fitts (3 attempts in 4 different difficulty indexes), Genius (5 minutes of training with progressive increase in difficulty according to the number of correct answers of the participant), MoveHero (15 minutes of training with music according to the number of spheres selected).

   A) Task 1 - Reaction time Reaction time is a task that consists of evaluating the time the subject spends to match his cognitive and motor response with the stimuli given on the computer screen.

   The TRT_S2012 software, built and validated by Crocetta et al., 2017, will be used in a sample of 76 healthy adults. The Software proposes a Simple Total Reaction Time (TRT) test, which consists of the appearance of a yellow square (parameterizable) in the center of the monitor at previously defined time intervals (ranging from 1.5 to 6.5 ms) and, when given the stimulus, the participant should react as quickly as possible, pressing the space bar on the computer keyboard.

   In this activity, the milliseconds that the subject presents of anticipation or delay of movement in response to the stimulus will be recorded. This task will test the speed of central and peripheral information processing and will be used in this study as an instrument to assess cognitive motor training.

   The Software also allows a Mental Fatigue Assessment Test from the TRT (TRT Fatigue). In this test, the stimulus consists of the appearance of a yellow bar that started at a time previously defined as the interval between stimuli (IEE), following a shift, from left to right. As an indication prior to the appearance of the stimulus, a thin vertical bar in black color, simulating a cursor, was presented until the yellow color filled the stimulus bar. The reaction should be to press the space key at the moment the yellow stimulus bar is perceived. The space key must be held down while the yellow color is shown. When the yellow stimulus is stopped, the space key must be released. Yellow stimulus presentation times were previously defined as stimulus presentation time (TAE). This activity will be used in this study as an instrument to assess cognitive motor training.

   B) Task 2 - Coincident Timing The coincident timing task is defined as the perceptual-motor capacity to execute a motor response in synchrony with the arrival of an external object, at a certain point (Belisle, 1963). This is an interception task that involves synchrony between eye and hand movements, requires a movement start, deceleration and "stop" strategy, in addition to accuracy and precision (Fooken et al., 2016; Ohta, 2016). The execution of this first protocol will be performed with the Coincident Timing task using the touchscreen (touch the computer screen).

   To evaluate the virtual task, we chose to use a software developed in partnership with the Information Systems group of the School of Arts, Sciences and Humanities, EACH-USP (first version used by Monteiro et al., 2014) which proposes the 3D coincident timing task on the computer. This activity will be used in this study as an instrument to assess cognitive motor training.

   In the virtual environment, 10 3D circles are displayed on the computer screen that light up (red light) in sequence until reaching the last circle that is considered the target. The participant must touch the computer screen exactly at the moment when the last sphere lights up, so the goal is to match the touch on the screen exactly above the last lit circle (target circle). The software provides immediate feedback on the task's success or error through different sounds and colors, previously demonstrated (Monteiro et al., 2014, 2017; Bezerra et al., 2018). If the participant hits the target at the same time as the stimulus arrives, a green light will turn on around the task - hit feedback, however if the participant delays or advances the movement, a red light will turn on next to the task - feedback from error. The game shows the sequence and speed of lighting the cubes and records the total time of task execution (time taken to touch the sensor); existing errors between stimulus arrival and task completion.

   In the design of the coincident timing study, each participant will perform 20 attempts of the task with the dominant upper limb at moderate speed, that is, 500ms between the lighting of each circle for the Acquisition phase. After acquisition, participants will wait at rest for 5 minutes and then perform five attempts in the Retention phase. As it is considered a simple task, we chose to use short-term retention (Monteiro et al., 2014 and Malheiros et al., 2016). For the Transfer phase, five more attempts will be made with increased speed, that is, 250ms between the lighting of each circle.

   Performance during acquisition, retention and transfer tasks will be evaluated considering the difference between target firing time and sensor touch. For this purpose, performance measures will be used as those related to achieving the "task-coincidence" goal: Constant Error, which reflects the directional trend of the movement, being of delay or anticipation of the response, and calculated using the arithmetic mean of the differences between performance and goal, considering their signs; Absolute Error, which corresponds to the accuracy with which the target of the action was reached, calculated through the arithmetic mean of the absolute differences between the actual performance in each attempt and the target; and Variable Error that identifies the precision of the movement, performing the calculation of each attempt, considering the signs, minus the average of the first five attempts for each participant, then the values are raised to the second power and to the average of each five attempts, extract the square root (Santos et al., 2003).

   C) Task 4 - Fitts The software that simulates the task is "FittsReciprocalAimingTask v.1.0 (Horizontal)", in the public domain (http://okazaki.webs.com - available from the internet on 09/01/2010) (Fernani et al., 2017) and was developed by Victor Hugo Alves Okazaki who presents the task proposed by Fitts' law in a virtual (computer) environment. This activity will be used in this study as an instrument to assess cognitive motor training.

   The task that assesses the speed and precision of movement relationship, based on Fitts' Law, consists of performing manual movements directed at a target (blue bars), in 3 difficulty indices (ID) with increasing demands for precision. The movement time is obtained through the division between pre-established seconds for the task (10) and the number of touches performed on the target.

   Before starting the task, the performer must place one of his hands on the mouse, and the cursor (viewed on the computer) must be in an intermediate point between the two targets (sidebars). After the initial position, the performer must use the mouse to move the cursor and click on the targets, two bars, which are arranged parallel vertically, with dimensions and distance determined according to each iID, alternately and as quickly as possible , after the audible alarm is triggered by the computer to start the task. After 10 seconds a new audible alarm indicates the end of the task. The total movement time will be recorded, obtained by dividing pre-established seconds for the task (10) and the number of touches (10/number of touches).

   The task consists of three attempts in each of the difficulty indices (ID 2, ID4a, ID4b and ID6), with each index changing the width and distance of the bars with an increasing level of difficulty. ID4 was applied in two ways (ID4a and ID4b): in ID4a the distance and width of bars are larger, in ID4b the distance between bars and width are smaller, keeping the same ID.

   For data analysis, the relationship between speed and accuracy of movement described by the mathematical equation proposed by Fitts will be considered, where there is a log-linear relationship between movement time and task difficulty, with targets of varying sizes and distances requiring different times to reach goals (Boyd et al., 2009).

   As the size of the target is reduced, or as the distance becomes greater, the movement speed decreases so that the movement is accurate. Due to the intrinsic information between the tathe target (D) and the distance between them (L), the log2 equation (2D/L) provides a difficulty index (DI) for the targeting skills where, the higher the ID, the more difficult the task. More difficult tasks will require more movement time (Fitts, 1954).

   D) Task 5 - Genius Game software created in partnership with the Information Systems group of the School of Arts, Sciences and Humanities, EACH-USP, will be used. Among the different games presented by the software, a game that allows motor and cognitive stimuli was chosen, the game presents a memory task where a sequence of colors is emitted by the computer (task based on the game commercially known as "Genius") so that the individual then repeat in order to obtain the correct sequence generated. The level of difficulty increases with each hit of the subject. The virtual interface provides reporting of reaction time and time between touches of each button.

   This activity will be used in this study as an instrument to assess memory performance as well as a component of cognitive motor training.

   E) Task 6 - MoveHero Software developed at the School of Arts, Sciences and Humanities of the University of São Paulo will also be used. The game features balls that fall, in four imaginary columns on the computer screen, in the rhythm of a song chosen by the researcher. The task is not to let the balls fall. However, the balls can only be touched when they reach four circles placed in parallel (at two levels of height), two to the left and two to the right of the participant (0 0 \o/ 0 0), called targets 1, 2 , 3 and 4 as viewed from left to right.

   The game captures the participant's movements through a webcam, not requiring physical contact to perform the task, so the participant must move their arms, at a distance of one and a half meters from the computer screen, or through the touch screen that requires contact on the computer screen in the inner space of the drawn circles. The participant must wait for the balls to fall, until they start to overlap one of the target circles. Therefore, the game requires the participant to have a strategy of anticipating movement to reach the balls within these circles.

   The game offers hit feedback through a numbering (+1) that appears next to the sphere that is successfully hit inside the target, in addition, the total score is visible in the upper left corner of the screen, with 10 points for each right.

   This activity will be used in this study as a memory performance assessment instrument as well as a component of cognitive motor training.

   3.3. Heart Rate Variability (HRV) The analysis of the HRV will follow the guidelines of the Task Force of the European Society of Cardiology and North American Society of Pacing and Electrophysiology (TFESC & NASPE, 1996).

   The volunteers will be instructed (online) to put on the collection belt given to them. The strap should be placed on the chest, and the heart rate receiver (Elite HRV app on the cell phone) will be positioned close to the volunteer. The HRV will be recorded after the initial assessment at rest for 10 minutes, during training with virtual reality for 10 minutes and for 10 minutes after the activity in VR. For analysis of HRV data at rest and during tasks, 256 consecutive RR intervals will be used.

   HR will be recorded beat by beat throughout the protocol by Elite HRV and the RR intervals recorded by the monitor will be transferred (.txt file) to the Kubios HRV® program (Kubios HRV v.1.1 for Windows, Biomedical Signal Analysis Group, Department of Applied Physics, University of Kuopio, Finland).

   Digital filtering will be performed in moderate mode in the program itself (Kubios HRV®) to eliminate premature ectopic beats and artifacts, and only series with more than 95% of sinus beats will be included in the study (Vanderlei et al., 2008).

   The analysis of HRV will be performed using linear methods (time and frequency domains) and through non-linear methods that will be analyzed using the Kubios HRV® software.

   A) Linear Methods (Time and Frequency Domain) In the time domain, the RMSSD, pNN50 and SDNN indexes will be used. The RMSSD index is defined as the square root of the mean square of the differences between adjacent normal RR intervals. (Marães et al., 2003).

   Where: RR = R-R intervals; N = number of RR intervals in the selected data series.

   The pNN50 index, in turn, is a sensitive and easily interpretable marker of parasympathetic SNA modulation, defined as the percentage of successive differences in the R-R interval whose absolute value exceeds 50ms. The SDNN, on the other hand, reflects the participation of both ANS branches and corresponds to the standard deviation of the mean of all normal RR intervals, expressed in ms (Pumprla et al., 2002; Ribeio and Moraes Filho, 2005; Vanderlei et al., 2009).

   For the analysis of HRV in the frequency domain, low frequency spectral components (LF - range from 0.04 to 0.15 Hertz) in absolute units and high frequency (HF - variation range from 0.15 to 0.4) will be used Hertz), in normalized units, and the ratio between these components (LF/HF), which represents the relative value of each spectral component in relation to the total power, minus the very low frequency components (VLF). The algorithm used for the spectral analysis will be the fast Fourier transform - FFT (256 s window with 50% overlap).

   To obtain the spectral indices, a graph (tachogram) is formed that expresses the variation of RR intervals as a function of time. The tachogram contains an apparently periodic signal that oscillates in time and is processed by mathematical algorithms such as the Fast Fourier Transform (FFT). The FFT method is used to obtain an estimate of HRV spectral power during stationary phases of the experiment in order to allow comparisons between study results. It allows the tachogram signal to be recovered even after transformation by the FFT, which demonstrates the objectivity of the technique, since information is not lost during the process. The ease of application of this method and good graphical presentation are the main reasons for its greater use (Carvalho, 2009; Vanderlei et al., 2009).

   B) Nonlinear Methods (Trends Debugged Fluctuations Analysis and Poincaré Graph) For the analysis of HRV by non-linear methods, the Poincaré graph (components SD1, SD2 and SD1/SD2 ratio), DFA (Trends Debugged Fluctuation Analysis) will be used.

   The DFA quantifies the presence or absence of fractal correlation properties of RR intervals and has been validated by time series data. This method calculates the root mean square fluctuation of the integral and debugs the time series, allowing detection of the intrinsic self-similarity embedded in the non-stationary time series.

   The DFA plot is not strictly linear, but consists of two distinct regions of distinct curves, separated by a point, suggesting that there is a short-term fractal scale exponent (α1) during periods of 4-11 beats (or 4 to 13), and a long-term exponent (α2), for longer periods (greater than 11 beats) (Carvalho et al, 2009).

   Values of α1 close to 0.5 are associated with white noise (random signal; there is no correlation between values), while values close to 1.5 are associated with Brownian noise (strongly correlated behavioral signals). Values close to 1.0 are characteristic of fractal-like processes associated with the dynamic behavior of time series generated by complex systems, such as the autonomic regulation of sinus rhythm in a healthy subject (Godoy et al., 2007).

   The Poincaré Graph, on the other hand, is a quantitative method of analysis, based on changes in the sympathetic or parasympathetic modulation of heart rate over subsequent intervals, without the need for the data stationarity property. The Poincaré is a diagram where each RR interval is represented as a function of RR (i-T), where i is the interval and T is a predefined delay used for an RR signal. Visual inspection of the diagram has been widely used in the analysis of HRV, where the Poincaré diagram can be analyzed quantitatively to calculate the standard deviations of the RR interval distances. These standard deviations are called SD1 and SD2. This analysis does not require pre-processing or data stability, which makes it especially interesting (Godoy et al., 2007).

   For quantitative analysis of the plot, the following indices will be calculated: SD1 (standard deviation of instantaneous beat-to-beat variability in the short term), SD2 (long-term standard deviation of continuous RR intervals) and the SD1/SD2 ratio (Brunetto et al ., 2005). The qualitative analysis of the plot will be done through the analysis of the figures formed by its attractor, which were described by Tulppo et al. (1998) in: 1) Figure in which an increase in the dispersion of RR intervals is observed with an increase in the intervals, characteristic of a normal plot. 2) Figure with small global beat-to-beat dispersion and no long-term increase in RR interval dispersion. 3) Complex or parabolic figure, in which two or more distinct edges are separated from the main body of the plot, with at least three points included at each edge.

4. Data analysis For data analysis descriptive statistics will be performed to characterize the sample and the results will be presented with mean and standard deviation values.

   For inferential analysis of tasks in VR, as dependent variables, for "Coincident Timing" and "MoveHero" the error measures in milliseconds will be considered (Constant errors that evaluate the directional trend of the movemento, Absolute that demonstrates the accuracy of movement and Variable that identifies the precision of the movement), for TR the time and milliseconds, for Genius the number of sequences performed at the beginning and end of the task time, as well as the time between each touch the spheres.

   For Fitts, the analysis will be performed by obtaining the values of b0 (intercept), b1 (inclination) and r2 (index of determination) for each participant, considering each attempt in the three difficulty indices. The MANOVA One-way will be used to compare the means of the groups in the three variables of interest (b0, b1 and r2). To verify associations between performance on the motor scale, age, slope of the straight line (b1), intercept (b0) and movement time dispersion data (r2), Pearson's correlation coefficient will be used. ANOVA will be performed to explore the effects of the difficulty index on movement time (TM) and multiple regression analysis to find which factors influence the TM.

   For the analysis of the HRV indices, the Student t-test or Mann-Whitney test will also be used for intergroup analyses.

   To analyze the influence of independent variables (age, disease stage, medications, etc.) on the dependent variables, the linear regression test will be performed.

   A significance level of 0.05 (5%) will be defined and all intervals constructed throughout the work will have 95% of statistical confidence.

   The statistical program will be SPSS (StatisticalPackage for Social Sciences), version 26.0.

5. Dissemination and evaluation VR results regarding motor function, pain and HRV will be analyzed through descriptive and inferential statistical analysis and disseminated in scientific articles, which will be sent to specific journals in the area with a high impact factor.

Furthermore, the results will be presented in book chapters and at international conferences. Thus, health and education professionals will have more scientific evidence to guide their activities in patients with AC.

Inclusion Criteria:

• Female patients over 18 years old with a diagnosis of breast AC, who agree to participate in the research and present an understanding compatible with the execution of the procedures, will be invited to participate in the study.

Exclusion Criteria:

• Exclusion criteria will be non-acceptance to participate in the research; presence of neurological diseases (infectious diseases, spinal cord injuries, epilepsy) or orthopedic diseases (orthopedic deformities, muscle strains, sprains, dislocations, fractures and pain conditions) in a period of occurrence; inability to play games on the computer due to motor limitations (fatigue, deformity or muscle weakness), motivational or difficulty in understanding a simple command; metallic skull implants; use of hearing aids; and use of anticonvulsant drugs or muscle relaxants.

We plan to share IPD through mendley database