Radiochromic films have long been used as reliable tools in radiation therapy for both dose measurement and dose distribution verification [1,2]. Their near tissue-equivalence, high spatial resolution, and low energy dependency make them suitable for various clinical dosimetry applications, including verification of treatment doses and commissioning of new equipment [3-5]. Among the popular choices, GAFchromic EBT films (Ashland Inc.) have become widely recognized in the field. Since its first introduction in 2004, the GAFchromic EBT film series has undergone multiple advancements, with each new iteration addressing specific limitations observed in its predecessor [6,7]. The latest models—EBT3 and the newly released EBT4 in 2023—demonstrate improvements in accuracy, signal-to-noise ratio (SNR), and ease of handling under ambient light conditions.

A notable improvement in EBT4 is the inclusion of a yellow dye in its active layer, enhancing its ability to absorb blue and ultraviolet light. This addition decreases the film’s sensitivity in the blue channel, minimizing lateral response artifacts, and improves overall accuracy in scanning orientations [8-11]. This structural enhancement allows EBT4 to maintain more consistent results under different conditions than EBT3.

Several recent studies have compared the performance of EBT4 with that of its predecessor, EBT3, across various clinical applications and dosimetric conditions. Akdeniz [12] conducted a comprehensive uncertainty analysis and reported that EBT4 exhibits lower uncertainties than EBT3, particularly at lower dose levels (0–2 Gy), suggesting improved precision and SNR in clinical dosimetry. Furthermore, Chan et al. [13] assessed EBT4’s dose–response characteristics for different energy levels and concluded that EBT4’s response was largely energy-independent across megavoltage beams, similar to EBT3, but displayed minor differences with kilovoltage X-rays, warranting separate calibration for low-energy photons. Palmer et al. [14] further explored EBT4’s application in treatment verification for advanced radiotherapy techniques, such as volumetric modulated arc therapy (VMAT) and stereotactic ablative radiotherapy (SABR), and found that EBT4 provided enhanced accuracy compared with EBT3, particularly with a significant increase in the SNR in the red and green channels at moderate doses. These findings collectively support EBT4’s suitability for clinical deployment in radiation therapy, offering improvements in stability and accuracy while maintaining similar handling procedures as EBT3.

A critical characteristic of radiochromic films is their delayed chemical reaction after irradiation, which causes variations in the pixel values (PVs) depending on the scan timing [1,15]. For instance, with the EBT3 film, a 24-hour scan time for calibration films is recommended to mitigate the effects of postirradiation darkening and to achieve consistent results [16]. Considering the recent release of EBT4, investigating whether the EBT4 film behaves similarly to the EBT3 film in this respect is essential. In particular, the focus of this study was to evaluate the postirradiation net optical density (NetOD) changes over time in the EBT4 film and to compare dose measurements at different scan times between EBT3 and EBT4.

Furthermore, this study assesses the dose–response curves for EBT3 and EBT4 films for different scanning modes, enabling a direct comparison of their sensitivity. This analysis not only provides insight into the dosimetric stability of EBT4 but also establishes whether its dose–response characteristics closely align with those of EBT3. This evaluation is crucial for ensuring the compatibility of EBT4 with existing dosimetric protocols in clinical settings and enhancing treatment verification accuracy.

1. Film preparation and beam exposure

EBT3 (LOT#08162202) and EBT4 (LOT#12192301) GAFchromic films, each measuring 1.4×1.6 cm, were prepared for dose measurements. In total, 792 film samples were used, with three films per dose level at each scanning time point. Following the American Association of Physicists in Medicine Task Group-51 (TG-51) protocol, absolute dose measurements were performed using a Varian TrueBeam linear accelerator, delivering 6-MV photon beams [17-19]. The dose levels were 0, 0.2, 0.5, 0.8, 1.0, 1.5, 2.0, 2.5, 3.0, 5.0, 8.0, and 10 Gy, with calibration films irradiated at the respective doses and scanned 24 hours after irradiation. After irradiation, the films were stored in an environment with no further radiation exposure and maintained at a constant temperature of 20°C and 50% humidity.

2. Scanning procedure and image acquisition

The irradiated films were scanned at specific intervals after exposure to observe postirradiation changes. The scanning times were 1, 2, 4, 6, 8, 12, 24, 48, 168 (7 days), and 336 (14 days) hours. Scanning was performed using an Epson 10000XL flatbed scanner in both transmission and reflection modes with a resolution of 300 dpi. The scanning setup was configured to 48-bit RGB color mode, positive film type, and a 10×10 mm region of interest (ROI). The scanned images were analyzed using RIT113 Classic (version 6.10; Radiological Imaging Technology, Inc.). A 5×5 median filter was applied to reduce image noise and improve the PV accuracy.

3. Dose–response curve and dose calculation

From each scanned image, PVs from the red channel within the defined ROI were extracted. Subsequently, the optical density (OD) was calculated, and the NetOD was determined using the following formula [20,21]:

where $M_{exp}$ represents the mean PV of the exposed film, $M_{bkg}$is the background PV, and $M_{unexp}$ is the PV of the unexposed film. This NetOD calculation enabled us to compare changes in NetOD over time for the EBT3 and EBT4 films in both transmission and reflection scan modes.

Furthermore, for each film type and scan mode, dose–response curves were evaluated across different time intervals and compared with the dose–response curve of the calibration film. Furthermore, the dose sensitivities of the calibration film were assessed and compared for each film type and scan mode. The relationship between NetOD and dose (D) was modeled using the following equation:

The sensitivity of each film was determined as the first derivative of NetOD with respect to dose as follows:

Calibration curves for the EBT3 and EBT4 films were established by fitting the NetOD values to determine dose values, expressed as follows:

Using these calibration curves, the NetOD values obtained at various postirradiation time points were applied to obtain the calculated dose values ($D_t$). The difference between the delivered (D) and calculated ($D_t$) doses was then evaluated in terms of accuracy using the following percentage difference formula:

Dose differences were assessed for the EBT3 and EBT4 films in both scanning modes at each postirradiation time interval.

1. Dose–response curve and sensitivity

The OD is defined as the logarithmic ratio of the light intensity before passing through the film to the intensity after transmission. NetOD is calculated by subtracting the background OD of the unirradiated film from the OD of the irradiated film, representing the degree of color change in the film caused by radiation exposure. NetOD is directly correlated with the absorbed dose, enabling precise dose evaluation in radiochromic film dosimetry [16,22].

Dose sensitivity was defined as the rate of change in NetOD per unit dose and represented as the slope of the NetOD curve as a function of dose. Sensitivity fluctuations refer to the variability in the film’s response across dose changes. Maintaining stable sensitivity over a range of doses is essential for achieving consistent and reliable dose measurements [8,23].

As shown in Fig. 1, changes in NetOD over time were observed for each film type and scan mode. After irradiation, the NetOD values continued to increase until 24 hours after beam exposure, with further gradual increases observed up to 7 days and continued even after this period. In the transmission mode, EBT3 and EBT4 are labeled as EBT3 Transmission and EBT4 Transmission, respectively, whereas, in the reflection mode, EBT3 and EBT4 are labeled as EBT3 Reflection and EBT4 Reflection, respectively.

Figure 1. Net optical density (NetOD) changes over time for the EBT3 and EBT4 films according to scan time and mode. (a) EBT3 transmission represents EBT3 scanned in the transmission mode, (b) EBT4 transmission represents EBT4 scanned in the transmission mode, (c) EBT3 reflection represents EBT3 scanned in the reflection mode, and (d) EBT4 reflection represents EBT4 scanned in the reflection mode. a.u, arbitrary units.

Fig. 2 compares the dose–response curves of the calibration film with those obtained at various postirradiation time points. Consistent with the increase in NetOD shown in Fig. 1, the dose–response curve also varied over time, particularly up to 24 hours after irradiation. Films scanned 24 hours after irradiation exhibited the closest match to the calibration film’s dose–response curve, indicating that scanning at 24 hours after irradiation yields the most accurate results compared with the calibration baseline.

Figure 2. (a) EBT3 transmission, (b) EBT4 transmission, (c) EBT3 reflection, and (d) EBT4 reflection. Dose–response curves for the EBT3 and EBT4 films according to scan time and mode. NetOD, net optical density; a.u, arbitrary units.

Fig. 3 presents dose sensitivity differences between EBT3 and EBT4. EBT3 demonstrated higher dose sensitivity than EBT4 at doses below approximately 4 Gy. However, EBT4 exhibited a more stable dose sensitivity response across the dose range, with less variability in sensitivity relative to dose changes than EBT3.

Figure 3. Dose sensitivity for the EBT3 and EBT4 films according to scan mode for the calibration film.

2. Dose difference

Table 1 and Fig. 4 present the comparison between the delivered and calculated doses, determined by applying the NetOD values for each film type and scan mode to the respective calibration curve. Table 1 presents the dose difference across a 0–10 Gy range for the EBT3 and EBT4 films in both scan modes at various time intervals after irradiation. The films scanned at 24 hours after irradiation, which corresponded to the calibration curve scan time, exhibited the highest accuracy across both scan modes and both film types. In particular, the mean dose difference and standard deviation for films scanned at 24 hours after irradiation were as follows: EBT3 transmission, −0.06±1.34; EBT4 transmission, −0.30±1.02; EBT3 reflection, −0.02±0.97; EBT4 reflection, −0.03±1.01.

Table 1 . Mean dose differences for the EBT3 and EBT4 films in the 0–10 Gy range across both scan modes.

Film type		Transmission mode						Reflection mode
		EBT3			EBT4			EBT3			EBT4
Time (h)		Mean (%)	SD (%)		Mean (%)	SD (%)		Mean (%)	SD (%)		Mean (%)	SD (%)
1		2.31	1.87		3.25	−0.27		3.06	1.11		3.63	1.44
2		2.46	1.61		2.47	2.07		1.88	2.83		2.30	1.79
4		2.13	1.58		1.92	2.00		1.92	2.54		1.96	1.14
6		1.80	1.76		1.70	0.50		1.84	1.54		1.16	1.10
8		1.82	0.89		1.09	1.46		1.34	1.27		0.64	1.03
12		1.13	0.98		0.50	1.57		0.62	1.06		0.30	1.12
24		−0.06	1.34		−0.30	1.02		−0.02	0.97		−0.03	1.01
48		−0.80	0.97		−1.11	1.47		−1.01	1.14		−1.19	0.93
168		−1.79	0.87		−1.20	1.44		−1.84	1.40		−2.27	0.91
336		−3.07	1.46		−2.33	2.16		−2.63	1.20		−2.85	0.92

Figure 4. (a) EBT3 transmission, (b) EBT4 transmission, (c) EBT3 reflection, and (d) EBT4 reflection. Box plot showing the mean dose difference between the delivered and calculated doses for the EBT3 and EBT4 films across scan modes over time. The graph illustrates the distribution of dose differences in the 0–10 Gy range as a function of time.

As illustrated in Fig. 4, the dose difference for the EBT3 and EBT4 films across the two scan modes exhibited fluctuations within approximately ±4% over the 1–336-hours postirradiation period. This variability highlights the effects of scanning time on dose measurement accuracy, particularly within the first several hours to days after irradiation.

To evaluate the effects of time on dose differences, analysis of variance was performed for each film type and scanning mode. The analysis revealed statistically significant variations in dose differences over time across all film modes, including EBT3 Transmission, EBT4 Transmission, EBT3 Reflection, and EBT4 Reflection (P<0.001 for all modes). This indicates that scanning time is a critical factor that influences dose measurement accuracy for both the EBT3 and EBT4 films, regardless of the scanning mode.

To assess the equivalence of performance between the EBT3 and EBT4 films in the reflection and transmission modes, paired t-tests were performed. The dose differences did not significantly differ between the EBT3 and EBT4 films in either the reflection (P=0.666) or transmission (P=0.745) mode. These results suggest that both film types demonstrate comparable performance in terms of dose differences across scanning modes.

This study investigated the dosimetric characteristics of EBT3 and EBT4 radiochromic films under different scanning modes and postirradiation times, particularly focusing on dose response, sensitivity, and postirradiation stability. The results indicate that despite minor differences in sensitivity, both EBT3 and EBT4 exhibited similar dose–response curves across transmission and reflection scanning modes, supporting EBT4 as a viable successor to EBT3 in clinical applications.

After irradiation, the films were stored under controlled conditions at 20°C and 50% relative humidity, ensuring a radiation-free environment to minimize variations in dose responses caused by external environmental factors. These carefully controlled storage conditions were implemented to ensure that the observed dose–response stability of the EBT3 and EBT4 films reflected their intrinsic properties rather than the influence of external storage factors. Future studies should investigate the impact of varying storage conditions on the long-term dosimetric stability of radiochromic films, which can offer valuable insights and practical guidelines for optimizing storage practices in clinical and research settings.

EBT4 demonstrated lower variability in sensitivity to dose changes than EBT3, probably because of the improved linearity of its dose–response curve. This finding is consistent with those reported by Akdeniz [12], who observed that EBT4 exhibits lower uncertainty in the low-dose range (0–2 Gy). Although EBT4’s improved uncertainty in this range offers distinct advantages, note that EBT3 demonstrated a steeper dose–response gradient, which may be beneficial for applications requiring higher sensitivity to small dose variations. This dual perspective highlights the complementary strengths of EBT3 and EBT4 in different clinical scenarios.

Furthermore, Palmer et al. [14] demonstrated that EBT4 offers an enhanced SNR and improved dose prediction accuracy, particularly in advanced radiotherapy techniques, such as VMAT and SABR. Consistently, this study confirmed that EBT4 maintains stable dose sensitivity across a wide dose range, supporting Palmer et al.’s findings [14]. These characteristics suggest that EBT4 provides safer and more effective treatment verification for radiotherapy techniques requiring steep dose gradients.

Thus, the reduced sensitivity variability observed in this study highlights the use of EBT4 in various precision radiotherapy applications where accurate dose delivery is essential while minimizing radiation exposure to surrounding healthy tissues. For instance, in pediatric cancer therapy, where precise low-dose delivery to sensitive and developing tissues is critical, the enhanced stability and reduced variability of EBT4 enable more accurate and reliable dose measurements [24]. Furthermore, in advanced radiotherapy techniques, such as stereotactic radiosurgery and stereotactic body radiation therapy (SBRT) [25], which require the delivery of high radiation doses with exceptional precision, the stable dosimetric properties of EBT4 ensure consistent and reliable dose verification. Furthermore, in left-sided breast cancer treatments, where protecting critical cardiovascular structures, such as the heart and coronary arteries, is crucial [26], and in ocular tumor therapies (e.g., retinoblastoma or uveal melanoma), which demand precise dose modulation to the eye and optic nerve [27], EBT4 facilitates accurate and reproducible dose measurements, thereby improving safety and efficiency in radiotherapy.

An important characteristic of both the EBT3 and EBT4 films is the delayed postirradiation chemical reaction effect, as noted in previous studies [1,15]. Both films exhibited an increase in NetOD within the first 24 hours after irradiation, followed by gradual stabilization over time. This delayed response highlights the necessity of standardized postirradiation scan times to ensure accurate dosimetry. Our findings demonstrate that the dose differences of both films varied significantly over time, regardless of the scanning mode. This observation is consistent with the findings of previous studies highlighting the role of delayed chemical reactions in NetOD stabilization [15]. Furthermore, films scanned at 24 hours after irradiation exhibited dose–response curves most consistent with the calibration reference values. The statistically significant time-dependent variations observed across all modes (P<0.001) emphasize the critical importance of allowing sufficient time for NetOD stabilization and scanning the films at the same postirradiation time as the calibration films to ensure reliable and accurate dose measurements.

This study was conducted using a specific scanning equipment (Epson 10000XL) with a fixed scanning protocol. As previous studies have shown, scanner resolution and uniformity can significantly affect dosimetric accuracy, with potential variability across scanner models [22,28,29]. To minimize these effects, this study employed a fixed protocol and scanned the films at the center of the scanner bed to reduce positional variability. Furthermore, the results were based on a specific lot of GAFchromic EBT4 films (Lot# 12192301). Lot-to-lot variations in terms of physical and chemical properties may influence dose response, suggesting that future studies should evaluate inter-lot variability to ensure reproducibility and broader applicability of findings.

In conclusion, EBT4 demonstrated comparable stability, accuracy, and ease of use to EBT3, while offering notable advantages, including reduced sensitivity fluctuations and improved linearity in dose response. These features make EBT4 particularly well-suited for clinical applications requiring precise low-dose delivery, such as pediatric cancer therapy, and for advanced radiotherapy techniques, such as VMAT and SABR, which require high-precision dose verification due to steep dose gradients.

The lack of significant differences in dose measurements between EBT3 and EBT4 in the reflection (P=0.666) and transmission (P=0.745) modes indicates that the two film types exhibit comparable dosimetric performance. This finding is consistent with those of previous studies, which highlighted EBT4’s ability to achieve dose measurement accuracy on par with EBT3, while providing additional benefits, such as enhanced linearity and reduced sensitivity variability [12,14].

These results support the use of EBT4 as a reliable and effective tool for clinical dosimetry, with the potential to improve dose measurement consistency and compatibility with established dosimetric protocols. Consequently, EBT4 presents itself as a robust and versatile option for various radiotherapy applications, particularly in scenarios requiring accurate and reproducible dose verification.

This study provides a comprehensive comparison of the characteristics of EBT3 and EBT4 radiochromic films for various scanning modes and postirradiation times. The results demonstrated that EBT4 achieved dose measurement accuracy comparable to that of EBT3 while exhibiting reduced sensitivity fluctuations and improved linearity throughout the dose range. The consistent postirradiation NetOD trends observed in both films further highlight the stability of EBT4, supporting its reliable integration into existing clinical protocols.

The reduced sensitivity variability of EBT4 makes it particularly suitable for precise and consistent dosimetry at low to moderate doses. This makes EBT4 highly advantageous for applications requiring high accuracy, such as pediatric radiotherapy, SBRT, and left-sided breast cancer treatments, where precision is paramount.

In conclusion, EBT4 is a valuable and advanced alternative to EBT3, addressing the needs of clinical dosimetry and treatment verification with enhanced performance. Future research should investigate its performance across different energy levels and advanced radiotherapy techniques to further solidify its role in clinical radiation therapy.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors.

The authors have nothing to disclose.

All relevant data are within the paper and its Supporting Information files.

Conceptualization: Jin Dong Cho. Data curation: Jin Dong Cho. Formal analysis: Jin Dong Cho. Investigation: Jin Dong Cho. Methodology: Jin Dong Cho. Supervision: Heerim Nam. Validation: Jin Dong Cho, Su Chul Han. Visualization: Jin Dong Cho. Writing – original draft: Jin Dong Cho. Writing – review & editing: Su Chul Han, Jason Joon Bock Lee, Hyebin Lee, Heerim Nam.