Does visual phosphenes associate with fraction dose of radiotherapy?

Introduction

Background

In recent years, there have been many studies on the visual phenomenon caused by photon and particle therapy. In 2013, the GSI Helmholtz Centre for Heavy Ion Research (GSI) proved that when high-energy particles hit the retina, they caused a flash. In 2016, a study on phosphenes was carried out in patients treated with choroidal melanoma at the Proton Therapy Center of the Institute d’Orsay Curie (ICPO) in France, Chuard et al. found that the mechanism of visual phosphenes in ocular proton therapy irradiation was related to space radiation (1-3).

Astronauts also reported flashes of different colors, shapes, and movements (optical illusion), and the correlation of dosimetry showed that these visual effects were caused by particles passing through the retina or optic nerve (4). A survey was conducted using a questionnaire administered to 98 National Aeronautics and Space Administration (NASA) and European Space Agency (ESA) astronauts, and it was found that 47 out of 59 respondents experienced sudden light flashes in space before sleep, and some of them indicated that the light flashes disturbed heir sleep (5).

Rationale and knowledge gap

Visual phosphenes, caused by stimuli other than luminance changes, are seen as signs of various diseases in the retina or visual pathways (6). Patients who experienced radiation in the head, neck, eyes, and central nervous system consistently reported that they perceived visual phosphenes (7). The experience of visual phosphenes varies widely among most patients but is predominantly white and blue, with various visual phenomena such as light of different colors, flashes, flickering, and streaks. Furthermore, this is not the case for all irradiated parts in all patients (8).

In oncology, visual phosphenes are also present during radiotherapy in patients with tumors of the central nervous system, eyes, head, and neck (9). The frequency of blue or white light produced during radiotherapy of the brain, head, and neck is usually related to retinal dose and age (10). Optical visual phosphenes can also be produced by photon activation of a photochemical by direct or scattered light beams or Cherenkov radiation (8).

The eye is the window of the human soul and is a critical organ that needs to be handled with great care. In addition, protons can induce neuronal responses sufficient to trigger a conscious sensory illusion. The brain region receiving the highest dose corresponds to the anatomical structure related to each illusion (11,12).

Logistic regression describes the relationship between a binary interest outcome and one or more independent variables. Logistic regression analysis was used to analyze retrospective data and build a prediction model to evaluate the probability of the outcome (13). Many studies (14-18) have used the logistic regression method to analyze and interpret parameters and outcomes in the radiation field. The odds ratio (OR) is a commonly used tool to describe the relationship and significance of each variable, especially in depicting the relationship between radiation volume and the probability of an event. Based on the results for area under the curve (AUC), logistic regression also had good predictive performance.

Objective

The aim of this study was to determine the association between visual phosphenes and fraction doses in different locations during radiotherapy and whether the increase in the fraction dose could be a factor influencing patient induced visual phosphenes. The relationship between the increase in dose and the color of visual phosphenes seen by the patients is another important issue that is explored. We present this article in accordance with the STROBE reporting checklist (available at https://tro.amegroups.com/article/view/10.21037/tro-23-39/rc).

Methods

Dataset

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and approved by the Institutional Review Board of Taipei Veterans General Hospital (2016-09-025C). All methods were performed in accordance with the relevant guidelines and regulations. Because of the retrospective nature of the research, the requirement for informed consent was waived. From November 2016 to March 2017, 101 patients were examined in this study, and 23 patients were excluded because of failure of the light switch test, which was used to test if the patient had consistent visual phosphenes without the influence of light. The participants were included 48 patients who underwent brain irradiation and 30 patients with head and neck cancer who received radiotherapy. Male and female patients in the age group of 20–85 years were considered eligible for participation. Other detailed relevant patient information and treatment methods are shown in Table 1.

Statistical analysis

Logistic regression analyzes the relationship between a categorical outcome of interest and the independent variables, including categorical and numeric variables. Compared to linear regression, logistic regression does not require many assumptions. The outcome of interest included dichotomous or multinomial variables. As the outcome of interest is a binary variable and the independent variable is numerical, logistic regression fits a curve to depict the relationship between the outcome of interest and the independent variable. This curve is S-shaped, and is also called a sigmoid curve (shown as Figure 1).

Figure 1 The graph of sigmoid function.

The general formula is shown below:

y=eα+β1x1+⋯+βxxx1+eα+β1x1+⋯+βxxx[1]

The regression output directly generates the estimate and P value. The general equation of binary logistic regression for multiple independent variables is as follows:

log⁡it(y)=log⁡(p1−p)=α+β1x1+⋯+βxxx[2]

where p is the probability of interested outcome, xi is an independent variable, α is an intercept, βi is the coefficient of xi (19,20).

We used logistic regression analysis to examine whether the outcome of seeing visual phosphenes is influenced by difference in fraction doses and locations. Logistic regression was fitted for each of the dichotomous outcomes of seeing visual phosphene and blue visual phosphene. In this study, we analyzed the dose at five different locations as our predictors to examine visual phosphenes. Statistical analyses, including mean, standard deviation, and logistic regression analysis, were conducted using RStudio version 1.2.5001 software (2009–2019 RStudio).

Results

A total of 295 samples from 78 patients (49 men and 29 women) were collected. The average patient age was 55.3 years with a standard deviation of 12.3 years. The patients reported different colored light visual phosphenes during the radiation treatment. In Figure 2, 172 samples are reported without visual phosphenes, and 123 samples with visual phosphenes, including 49, 18, 15, and 41, white, orange, blue, and mixed color records, respectively (Figure 3). The mean and standard deviation of the fractional dose at each of the locations are shown in Table 2. To investigate whether there are significant differences in the means of fraction dose between patients without visual phosphenes and with different color visual phosphenes, an analysis of variance (ANOVA) was used to examine significant differences. The comparison of the mean dose was considered statistically significant if the P value was less than 0.05. Detailed results of the ANOVA, Tukey’s honest significant difference (HSD) tests, and the boxplot of fraction dose in different area between the group without visual phosphenes and with different visual phosphenes are shown in the supplementary material (Tables S1-S10 and Figures S1-S5). During the radiation treatment, patients with blue visual phosphenes are treated with higher doses at each location as compared to those without visual phosphenes or those with other colorful visual phosphenes.

Figure 2 The percentage of patients with/without visual phosphenes.

Figure 3 The distribution of samples with visual phosphenes of color scale intensity map.

Table 3 shows the results of logistic regression analysis of patients with visual and blue visual phosphenes. The results for patients with visual phosphenes, the fractional doses at the retina and visual cortex are considered to have a statistically significant effect on the visual phosphenes. The retinal fraction dose has a positive effect on visual phosphenes. In contrast, the visual cortex fraction dose has the opposite influence. The OR increases to 1.0523 times when the retina fraction dose increases by 1 cGy, which is an increase of 5.23%. On the other hand, the OR reduces to 0.9762 times as the cortex fraction dose increases by 1 cGy which is a decrease of 2.38%. In patients with blue visual phosphenes, only the retinal fraction dose is considered to have a statistically significant effect. The OR increases to 1.0366 times, while the retina fraction dose increases by 1 cGy, which is an increase of 3.66%. Compared to the fractional dose at other locations, the one at the retina plays a key role in determining whether patients experience visual phosphenes.

Figure 4 shows a visualization of the logistic regression analysis of the probability of seeing visual phosphenes and blue visual phosphenes. In this section, we discuss only the relationship between the retinal fraction dose and the probability of outcome. The blue lines show the probabilities for different of retinal fraction doses. The plots show that the probabilities increase with the retinal fraction doses. With the increase in dose, the probability of seeing visual phosphenes also increases (Figure 4A). In Figure 4A, the result shows that the probability is higher than 0.5, when the dose is greater than 31 cGy. However, the probability of seeing blue visual phosphenes is higher than 0.5when the dose is greater than 96 cGy (Figure 4B). The results of the logistic regression analysis are in line with those in Table 3. The increase in the retina fraction dose influences the visual phosphene of patients during radiation treatment; in comparison with other colors, it is likely to make patients sense blue visual phosphene when the dose is increased further. Therefore, patients receiving higher doses are more likely to experience blue visual phosphene than those receiving lower doses.

Figure 4 Results of probability scale based on logistic regression analysis for (A) the outcome for seeing visual phosphenes and (B) the outcome for seeing blue visual phosphenes. The blue line shows the relationship between the probability and retina fraction dose (cGy). The gray area is the 95% confidence interval.

Discussion

Key findings

Our study aimed to determine the relationship between the fraction dose rate at each location and visual phosphenes. The main objective was to explore the reason for the different colors of visual phosphenes. Logistic regression analysis revealed that the retina and visual cortex doses had a significant effect on visual phosphenes. The retinal fraction dose has a positive effect on visual phosphenes; however, the visual cortex has a negative effect.

As mentioned in the literature, many patients experience flashes during radiotherapy. Retinal dose and age were significantly correlated with the flash rate. This study suggested that the retina is an important factor in flash. The intensity of the light decreased with the treatment area, and the retinal dose decreased from 50% to 0%. Therefore, the occurrence of flash decreased with a reduction in the irradiation area. This shows that the retinal and eye doses affect the flash, which is also highly correlated with the results presented in this paper (10). Only the retinal fraction dose influenced the patient’s perception of blue visual phosphenes. Compared with other colors, it is likely that patients will feel a blue visual illusion, which accounts for 19%. This finding is consistent with other previous studies, which also reported that most colors of proton therapy are blue to purple. Blue and purple are dominant in radiotherapy of the brain or head (7,10).

As seen in Figure 4, we used logistic regression to determine the relationship between the probability and retinal fraction dose rate. We conclude that a higher retinal fraction dose causes visual phosphenes. Blue visual phosphene is caused by a strong increase in the retinal fraction dose. The retinal fraction dose is the most important factor affecting visual phosphenes, and in our team’s previous publication it also has a positive effect on visual phosphenes (21). The effect of the retinal fraction dose is in line with our logistic regression results. de Kruijf et al. determined that the retinal dose can influence the visual phosphenes that occur in patients as well. They assumed that the probability of patients experiencing visual phosphenes could be reduced by decreasing the dose rate for the retina (8). A relationship between the doses for retina and vitreous humor has been found in the literature, with the odds for visual phosphenes increasing exponentially by 1.604 with an increase of 10 cGy dose at the retina. The characteristic colors of the flashes vary, with 80% of them being predominantly blue and violet, which is consistent with the findings in literature on patients with head and neck vision. Patients with cancer treated with high-energy photons predominantly have typical blue flashes. The visual phosphenes in the process of remote photon radiotherapy are mainly due to radiation in the vitreous humor as reported by Cherenkov (9).

Comparison with similar researches

It is known from literature (21) that energy is not the only key to inducing visual phosphenes, and the most important possible reason is the location of radiation-induced visual pathways; this is highly correlated with the statistical results of this study. During radiotherapy, visual phosphenes were affected, with fractional doses to both the retina and visual cortex considered to have a statistically significant effect on visual phosphenes, while fractional doses to the retina had a positive effect on visual phosphenes. The OR increased by 1.0366, while the retina fraction dose increased by 1 cGy, which is an increase of 3.66%. Compared to the fractional dose at other locations, the retina plays a key role in determining whether patients sense the visual phosphenes.

Explanations of findings

In this study, when the fractional retinal dose was increased by 1 cGy, the OR increased by 1.0523, which is an increase of 5.23%. Therefore, it is known that the phenomenon of optic visual phosphenes does not seem to require high doses or energies to be induced in patients during treatment. A study by Kuo et al. showed that subjects experienced visual phosphenes, regardless of whether their eyes were open or closed (21). The results from this study show that the probability is higher than 0.5 when the dose is greater than 31 cGy. However, at doses greater than 96 cGy, the probability of seeing blue visual phosphenes is higher than 0.5. In comparison with other colors, it is likely that the patient will experience blue visual phosphenes, requiring more dose increases rather than energy increases. Flash heavy ions have been reported in treated cancer patients (21), with patients reporting mostly white flashes (10% yellow). The energy of the ion beam was also lower than that of the Cherenkov lamp, indicating that the Cherenkov lamp did not cause this flash. Instead, the primary mechanism could be the deposition of energetically charged particles on the retina (16).

The color difference could be caused by the energy of the charged particles; however, this is still unclear (10). Compared with other colors, we found that an increase in the dose is likely to cause patients to feel blue visual phosphenes. From the color scale chart in Figure 3, the intensity of color is divided into 1–5 grades. Different dose and energy might cause patients to see different intensities of color in visual phosphenes.

Strengths and limitations

Based on our statistical results, we cannot clearly determine which color of visual phosphenes can be caused when the dose is high and whether the difference in light intensity is also positively correlated with the dose. In laboratory research on mice, Garcia et al. showed that the rapid sensory perception of radiation reflex can detect the radiation attack reflex in animals, through an electroencephalogram (EEG) (22). Moreover, it is impossible to observe the brain using EEG during radiotherapy (22,23). At the time point when the subjects in this study received the radiation output of high-energy radiotherapy, it was difficult to record the time point of visual perception events using instruments, which is a limitation of this study.

Implications and actions needed

Visual phosphenes are induced by radiotherapy according to our results. In a previous study, Bates et al. observed that visual acuity declined in pediatric survivors of brain tumors who underwent radiotherapy, especially in very young children (24). Moreover, radiotherapy also influences late neurocognitive and delayed verbal memory (25). Radiotherapy used to treat malignant brain tumors is associated with neurotoxicity, including potential irreversible decline in short-term memory and cognitive ability, which may occur as early as 1–4 months after treatment (26,27). To the best of our knowledge, there is no evidence of a relationship between visual phosphenes and visual decline today. Since it has been found that medical history and physiology do not play a significant role in triggering optical illusions, it may be concluded that only the beam characteristics are the most important. However, it cannot be excluded that variables outside the scope of this study (such as stress, fatigue, and diet) may have an impact (28). In the future, we will keep track of patients with visual phosphenes in clinical settings. However, the neurophysiological targets of visual phosphene stimulation remain unclear. Visions generated by transcranial magnetic stimulation are mostly induced in the visual hemifield on the opposite side of the stimulation site (23). It has been reported that protons can induce sufficient neuronal responses (11). In the future, we may use a variety of sensory organs in a similar manner to further study the brain regions receiving the highest dose to analyze the sensory illusion neuronal responses triggered by the corresponding regions. It is possible to analyze the relationship between beam direction and olfactory perception to identify the organs involved (29-33).

To sum up the above discussion we can hypothesize that evoked potentials can be used to assess sensitivity to phosphenes. It can be deduced from the literature and the results of this study that more objective measures of retinal based assessment, increasing the retinal dose fraction will increase the patient’s response to radiation therapy, and the cause of the light sensation caused by proton rays, is believed to be generated by free radicals close to the retina. Photon rays caused by the interaction between protons and substances are not the only ones produced by general radiation therapy. The Compton effect produced by the field, so these excess free radicals are likely to generate evoked potentials, further making them more sensitive to phosphenes. For patients who have experienced phosphenes induced by radiation therapy, the retinal nerve fiber layer becomes thinner Whether the risk is higher, whether it will cause depolarization of nerve action potentials, and whether there is a threshold dose is a direction worthy of subsequent research and exploration.

Conclusions

Visual phosphenes produced by the human body during radiotherapy are induced by radiation rather than by self-suggested hallucinations. In this study, we found that the retina fraction dose had a statistically significant effect on visual phosphenes, and the increase in retinal fraction dose enhances the probability of visual phosphenes in patients during radiotherapy. A higher retinal fraction dose is even involved in the color of the visual phosphenes that patients sensed. Furthermore, visual decline was observed in pediatric patients who underwent radiotherapy. In the future, we will continue to monitor patients who experience visual phosphenes.

Acknowledgments

The research group would like to express the highest gratitude to the following researchers who provided assistance for the study. We would like to thanks Yu-Ming Liu, Keng-Li Lan, Yu-Wen Hu, Kuo-Ying Fan and all the medical radiographers of Department of Heavy Particles and Radiation Oncology, Taipei Veterans General Hospital, Taiwan for their assistance during this study, and Professor Jao-Perng Lin from Department of Medical Imaging and Radiological Technology, Yuanpei University of Medical Technology for review of this article and teaching guidance.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tro.amegroups.com/article/view/10.21037/tro-23-39/rc

Data Sharing Statement: Available at https://tro.amegroups.com/article/view/10.21037/tro-23-39/dss

Peer Review File: Available at https://tro.amegroups.com/article/view/10.21037/tro-23-39/prf

Funding: This work was supported by the National Science and Technology Council (NSTC) in Taiwan (No. NSTC113-2221-E-A49-193-MY3), the University System of Taipei Joint Research Program (No. USTP-NTOU-TMU-112-04 to E.C.Y.S.), and the National Science and Technology Council (NSTC) in Taiwan (No. NSTC113-2221-E-227-002 to C.Y.K.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tro.amegroups.com/article/view/10.21037/tro-23-39/coif). Y.W.C. serves as the unpaid associate Editor-in-Chief of Therapeutic Radiology and Oncology from October 2017 to July 2025. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and approved by the Institutional Review Board of Taipei Veterans General Hospital (2016-09-025C). All methods were performed in accordance with the relevant guidelines and regulations. Because of the retrospective nature of the research, the requirement for informed consent was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Chuard D, Anthonipillai V, Dendale R, et al. Mechanisms of phosphene generation in ocular proton therapy as related to space radiation exposure. Life Sci Space Res (Amst) 2016;10:23-8. [Crossref] [PubMed]
2. Schardt D, Kavatsyuk O, Krämer M, et al. Light flashes in cancer patients treated with heavy ions. Brain Stimul 2013;6:416-7. [Crossref] [PubMed]
3. Narici L. Heavy ions light flashes and brain functions: recent observations at accelerators and in spaceflight. New J Phys. 2008;10:075010. [Crossref]
4. Sannita WG, Narici L, Picozza P. Positive visual phenomena in space: A scientific case and a safety issue in space travel. Vision Res 2006;46:2159-65. [Crossref] [PubMed]
5. Fuglesang C, Narici L, Picozza P, et al. Phosphenes in low earth orbit: survey responses from 59 astronauts. Aviat Space Environ Med 2006;77:449-52. [PubMed]
6. Cervetto L, Demontis GC, Gargini C. Cellular mechanisms underlying the pharmacological induction of phosphenes. Br J Pharmacol 2007;150:383-90. [Crossref] [PubMed]
7. Mathis T, Hofverberg P, Caujolle JP, et al. Occurrence of Phosphenes in Patients Undergoing Proton Beam Therapy for Ocular Tumor. Am J Ophthalmol 2018;192:31-8. [Crossref] [PubMed]
8. de Kruijf W, Timmers A, Dekker J, et al. Occurrence and mechanism of visual phosphenes in external photon beam radiation therapy and how to influence them. Radiother Oncol 2019;132:109-13. [Crossref] [PubMed]
9. Robles Díaz JF, Cabrera Moreno CM. Prediction of phosphenes occurrence in patients undergoing photon irradiation to ocular structures due to brain tumor. Arch Soc Esp Oftalmol (Engl Ed) 2022;97:331-6. [Crossref] [PubMed]
10. Mizumoto M, Oshiro Y, Miyamoto T, et al. Light flashes during proton and photon radiotherapy: A multicenter prospective observational study. Tech Innov Patient Support Radiat Oncol 2021;20:41-5. [Crossref] [PubMed]
11. Narici L, Titova E, Obenaus A, et al. Multiple sensory illusions are evoked during the course of proton therapy. Life Sci Space Res (Amst) 2020;26:140-8. [Crossref] [PubMed]
12. Sannita WG, Peachey NS, Strettoi E, et al. Electrophysiological responses of the mouse retina to 12C ions. Neurosci Lett 2007;416:231-5. [Crossref] [PubMed]
13. Zabor EC, Reddy CA, Tendulkar RD, et al. Logistic Regression in Clinical Studies. Int J Radiat Oncol Biol Phys 2022;112:271-7. [Crossref] [PubMed]
14. Yafeng L, Jing W, Jiawei Z, et al. Construction and Verification of a Radiation Pneumonia Prediction Model Based on Multiple Parameters. Cancer Control 2021;28:10732748211026671. [Crossref] [PubMed]
15. Robertson JM, Söhn M, Yan D. Predicting grade 3 acute diarrhea during radiation therapy for rectal cancer using a cutoff-dose logistic regression normal tissue complication probability model. Int J Radiat Oncol Biol Phys 2010;77:66-72. [Crossref] [PubMed]
16. Takeyama T, Sasaki N. Multivariable Logistic Regression Models of X-Ray Thoracic Spinous Process Osseous Changes Findings and Body Measurement Factors Associated With Defined Over-riding of the Dorsal Spinous in Riding Horses. J Equine Vet Sci 2022;109:103839. [Crossref] [PubMed]
17. Moran A, Daly ME, Yip SSF, et al. Radiomics-based Assessment of Radiation-induced Lung Injury After Stereotactic Body Radiotherapy. Clin Lung Cancer 2017;18:e425-31. [Crossref] [PubMed]
18. Saffari SE, Löve Á, Fredrikson M, et al. Regression models for analyzing radiological visual grading studies--an empirical comparison. BMC Med Imaging 2015;15:49. [Crossref] [PubMed]
19. Park HA. An introduction to logistic regression: from basic concepts to interpretation with particular attention to nursing domain. J Korean Acad Nurs 2013;43:154-64. [Crossref] [PubMed]
20. Pandis N. Logistic regression: Part 1. Am J Orthod Dentofacial Orthop 2017;151:824-5. [Crossref] [PubMed]
21. Kuo CY, Lee CC, Lee YL, et al. Visual light perceptions caused by medical linear accelerator: Findings of machine-learning algorithms in a prospective questionnaire-based case-control study. PLoS One 2021;16:e0247597. [Crossref] [PubMed]
22. GARCIA J. Electroencephalographic responses to ionizing radiation. Science 1963;140:289-90. [Crossref] [PubMed]
23. Tani N, Hirata M, Motoki Y, et al. Quantitative analysis of phosphenes induced by navigation-guided repetitive transcranial magnetic stimulation. Brain Stimul 2011;4:28-37. [Crossref] [PubMed]
24. Bates JE, Indelicato DJ, Morris CG, et al. Visual decline in pediatric survivors of brain tumors following radiotherapy. Acta Oncol 2020;59:1257-62. [Crossref] [PubMed]
25. Zureick AH, Evans CL, Niemierko A, et al. Left hippocampal dosimetry correlates with visual and verbal memory outcomes in survivors of pediatric brain tumors. Cancer 2018;124:2238-45. [Crossref] [PubMed]
26. Chang EL, Wefel JS, Hess KR, et al. Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol 2009;10:1037-44. [Crossref] [PubMed]
27. Gondi V, Tomé WA, Mehta MP. Why avoid the hippocampus? A comprehensive review. Radiother Oncol 2010;97:370-6. [Crossref] [PubMed]
28. Narici L, De Martino A, Brunetti V, et al. Radicals excess in the retina: A model for light flashes in space. Radiation Measurements 2009;44:203-5. [Crossref]
29. Sagar SM, Thomas RJ, Loverock LT, et al. Olfactory sensations produced by high-energy photon irradiation of the olfactory receptor mucosa in humans. Int J Radiat Oncol Biol Phys 1991;20:771-6. [Crossref] [PubMed]
30. Hara N, Isobe A, Yamada K, et al. Unusual visual and olfactory perceptions during radiotherapy sessions: an investigation of the organs responsible. J Radiat Res 2021;62:718-25. [Crossref] [PubMed]
31. Mizumoto M, Oshiro Y, Sumiya T, et al. Olfactory Sensations During Proton and Photon Radiotherapy: A Multicenter Prospective Observational Study. Cureus 2022;14:e22964. [Crossref] [PubMed]
32. Mizumoto M, Oshiro Y, Miyamoto T, et al. Light flash and odor during proton beam therapy for pediatric patients: a prospective observational study. Front Oncol 2022;12:863260. [Crossref] [PubMed]
33. Hara N, Oobuchi J, Isobe A, et al. Generation of ozone during irradiation using medical linear accelerators: an experimental study. Radiat Oncol 2022;17:39. [Crossref] [PubMed]