Development and characterization of the efficient portable X-ray imaging device based on Raspberry Pi camera

This study presents the development and characterization of a compact, portable, and efficient X-ray imaging system based on a Raspberry Pi camera and scintillation screens, which achieves clinical-grade spatial resolution and demonstrates versatility across different scintillator materials for scientific, educational, and medical applications.

The Gist

Imagine you want to take an X-ray picture of a broken toy or a secret compartment in a suitcase. Usually, you'd need a massive, expensive machine the size of a small car, costing tens of thousands of dollars, and it would need a dedicated room with heavy shielding.

This paper describes a team of scientists who decided to build a "pocket-sized" X-ray camera using parts you can buy at an electronics store. Think of it as building a high-tech camera using a Raspberry Pi (a tiny, $50 computer often used by hobbyists) and a high-quality lens, but with a special twist to see inside objects.

Here is the story of how they did it, explained simply:

1. The Core Idea: Turning Invisible Light into Visible Light

X-rays are invisible to our eyes and to normal cameras. To take a picture, you need something to catch the X-rays and turn them into visible light, like a translator.

• The Translator: They used a special glowing screen (called a scintillator) made of a material called GOS. When X-rays hit this screen, it glows like a neon sign.
• The Camera: Instead of a giant medical detector, they used a Raspberry Pi High-Quality Camera. This is the same camera module used by photographers for macro shots, but here it's used to photograph the glow from the screen.

2. The "Folded" Design: A Clever Detour

There was a big problem: X-rays are dangerous to the camera's sensitive electronics. If you put the camera right next to the screen, the X-rays would fry it.

• The Solution: They built a clever "folded" path. Imagine a hallway where you can't walk straight because of a wall. Instead, you put a mirror in the corner to bounce the view around the wall.
• They used a prism (a glass mirror) to bounce the light from the screen 90 degrees away from the X-ray beam. This keeps the camera safe in a "bunker" (an aluminum case) while still letting it see the image. It's like watching a movie on a screen in the next room through a periscope.

3. Tuning the Camera: Finding the "Sweet Spot"

Just like taking a photo in the dark with your phone, you have to balance the settings.

• ISO (Sensitivity): If you turn the sensitivity too high, the picture gets bright but grainy (noisy).
• Exposure Time: If you leave the shutter open too long, you get a brighter picture, but it might get blurry or the camera might get "tired" (noisy).
• The Result: The team found the perfect balance. They discovered that for X-ray imaging, a specific setting (ISO 400 with a 500ms exposure) gave them the clearest, sharpest image without too much grain.

4. How Sharp is the Picture? (The Resolution Test)

To see if their cheap camera was any good, they tested it against a "resolution target" (a pattern of tiny lines, like a barcode).

• In Normal Light: The camera was incredibly sharp, able to see tiny details (68 lines per millimeter).
• Under X-rays: The image got a bit softer (25 lines per millimeter) because X-rays scatter a bit inside the glowing screen, like light scattering in fog. However, 25 lines per millimeter is still very sharp—it's comparable to the machines used in hospitals for checking broken bones!

5. The Trade-Off: Contrast vs. Clarity

The paper explains a classic balancing act in X-ray imaging:

• Low Energy (50 kV): Think of this as a "soft" X-ray. It creates high contrast (very dark and very bright areas), making it easy to see the difference between materials, like seeing a metal coin inside a plastic bag. However, the image is a bit "noisy" (grainy).
• High Energy (70 kV): Think of this as a "hard" X-ray. It penetrates deeper and creates a cleaner, smoother image (less noise), but the difference between materials becomes fainter (lower contrast).
• The Takeaway: You can choose your setting based on what you are looking at. If you need to see fine details, you go for the cleaner image. If you need to spot a specific object, you go for the high contrast.

6. Why This Matters

The most exciting part is the cost and flexibility.

• Cost: A professional X-ray detector costs as much as a luxury car ($50,000+). This device cost about $570 (excluding the X-ray machine itself, which is a separate piece of equipment).
• Modularity: Because it's built like Lego, they could swap out the glowing screen. They tested it with two other types of screens (LYSO and GAGG) and it worked perfectly with both. This means the same camera could be used to detect neutrons, protons, or different types of radiation just by changing the screen.

The Bottom Line

This paper proves that you don't need a million-dollar lab to do high-quality scientific imaging. By combining a tiny computer, a clever mirror trick, and some open-source software, they built a portable, affordable X-ray camera that is sharp enough for medical and scientific use. It's like turning a smartphone into a professional-grade medical scanner, making advanced imaging accessible to schools, small labs, and researchers everywhere.

Technical Summary

1. Problem Statement

Traditional digital radiography systems, while offering superior dynamic range and dose efficiency compared to film, remain costly, bulky, and inaccessible for small-scale research, educational institutions, and resource-limited settings. Existing solutions often lack the flexibility to be adapted for various excitation sources (e.g., neutrons or protons) without significant re-engineering. There is a critical need for a compact, affordable, and high-resolution X-ray imaging platform that maintains scientific-grade performance while enabling portability and modularity.

2. Methodology and Instrument Configuration

The authors developed a portable indirect-conversion X-ray imaging system using commercially available, low-cost components.

• Core Hardware:

  • Controller: Raspberry Pi 4 Model B (providing control, data acquisition, and remote operation via SSH/VNC).
  • Sensor: Raspberry Pi High-Quality (HQ) Camera module featuring a Sony IMX477 back-illuminated CMOS sensor (12.3 MP, 1.55 µm pixel pitch). This sensor offers low read noise (~3 e- rms) and high quantum efficiency.
  • Scintillator: A Gd₂O₂S:Tb (GOS) screen, chosen for its emission spectrum (450–700 nm) which aligns well with the camera's peak sensitivity.
  • Optical Assembly: A custom "folded" optical path using a 15×15×15 mm optical prism and a 16 mm C-mount lens. This design deflects the light path by 90°, protecting the radiation-sensitive camera sensor from direct X-ray exposure and allowing for a compact form factor.
  • Housing: An aluminum enclosure shields the electronics and reduces ambient light noise.

• Characterization Methods:

  • Optical Verification: Tested under ambient light using a resolution target (THORLABS R2L2S2P) to verify focusing.
  • Noise Analysis: Measured read noise and dark current across ISO settings (100–800) and exposure times (100–2000 ms) using dark frames.
  • X-ray Testing: Performed using a clinical X-ray source at 50 kV and 70 kV. Parameters varied included tube current (mA) and exposure time (ms) to achieve different mAs levels.
  • Performance Metrics:
    • Spatial Resolution: Quantified using the Slanted-Edge method to derive the Modulation Transfer Function (MTF).
    • Contrast & SNR: Calculated using Regions of Interest (ROIs) to determine Signal-to-Noise Ratio (SNR) and image contrast.
    • Modularity: Validated by swapping the GOS screen with LYSO:Ce and GAGG:Ce scintillators.

3. Key Contributions

• Low-Cost High-Performance Platform: Demonstrated that a Raspberry Pi-based system can achieve spatial resolution comparable to clinical systems at a hardware cost of approximately $570 (excluding the X-ray source).
• Optimized Indirect Detection: Successfully implemented a folded optical path with a prism to protect the sensor, a novel configuration for portable Pi-based X-ray systems.
• Systematic Parameter Optimization: Provided a comprehensive analysis of the trade-offs between ISO, exposure time, and X-ray tube voltage (kV) to maximize image quality.
• Modularity: Proved the system's versatility by successfully imaging with different scintillation materials (GOS, LYSO, GAGG), suggesting applicability for neutron or proton imaging.

4. Key Results

• Spatial Resolution (MTF):
  • Under ambient light, the system achieved an MTF20 of 68 lp/mm, confirming excellent optical alignment.
  • Under X-ray irradiation, the MTF20 was 25 lp/mm at both 50 kV and 70 kV. This resolution is comparable to many clinical radiography systems.
  • Observation: Resolution improved slightly at 70 kV compared to 50 kV because higher-energy photons penetrate deeper into the scintillator, reducing light scatter within the phosphor layer.
• Noise Performance:
  • The camera exhibited low intrinsic noise. At ISO 100, the standard deviation was minimal. Even at ISO 800 with long exposure (2000 ms), the deviation remained below 1.6% of the mean, confirming suitability for scientific imaging.
• Contrast and SNR Trade-offs:
  • 50 kV: Produced higher image contrast due to dominant photoelectric absorption but resulted in lower SNR due to lower photon flux.
  • 70 kV: Produced higher SNR due to increased photon flux and penetration but resulted in lower contrast as Compton scattering became dominant and the energy transfer efficiency per photon decreased.
• Optimal Settings: The study identified ISO 400 with a 500 ms exposure as a balanced setting, offering approximately 73% contrast with manageable noise levels.

5. Significance and Impact

• Accessibility: This work democratizes access to high-resolution digital radiography, making it feasible for educational institutions, small labs, and field applications where budget and space are constraints.
• Scientific Utility: The system's ability to operate with various scintillators opens doors for multi-modal imaging (e.g., neutron radiography) without redesigning the core electronics.
• Safety: The remote operation capabilities (SSH/VNC) allow operators to control the system from a safe distance, adhering to the ALARA (As Low As Reasonably Achievable) principle for radiation safety.
• Future Applications: The platform is positioned for use in non-destructive testing (NDT), materials characterization, and medical diagnostics in resource-limited environments, bridging the gap between expensive clinical equipment and low-cost prototypes.

In conclusion, the authors successfully validated a scientific-grade, portable, and modular X-ray imaging system that challenges the notion that high-resolution radiography requires expensive, bulky infrastructure.