The X-ray sensitivity of radiology instruments is limited by the materials used in their detectors. A material from the perovskite family of semiconductors could allow lower doses of X-rays to be used for medical imaging. See Letter p.87
X-ray imaging is essential for medical applications, but the images produced can be too grainy and blurry for certain purposes. It would also be desirable to lower the radiation doses required, to reduce the risk of harming patients. On page 87, Kim et al.1 report that a semiconductor previously used in solar cells can be repurposed in X-ray detectors, and can produce sharper images using lower doses than are possible with currently available detectors. Moreover, the process used to make the detectors should allow the devices to be manufactured more cheaply than existing detectors.
The use of X-rays for medical imaging is justified only if the risk of harm caused by the radiation is exceeded by the benefit of acquiring diagnostic information. Technical developments in X-ray imaging aim to improve diagnostic certainty or reduce the radiation dose, or both. During the digital revolution in radiology2, the development of instruments called flat-panel X-ray detectors both improved image sharpness and reduced graininess. This increased the diagnostic value of the imaging, while making it possible to reduce the radiation dose compared with that required in the first available digital X-ray imaging procedure, known as computed radiography3.
Flat-panel X-ray detectors are currently based either on the direct-conversion method using amorphous selenium (a semiconductor), or on the indirect-conversion method, which uses the luminescent material caesium iodide (CsI). In the direct method, X-ray energy is converted by the semiconductor into many electrons, which are collected on electrodes. The electrodes form part of an integrated circuit that reads out the image. In the indirect method, photons of light are created as an intermediate signal that is subsequently converted into electrons. A drawback of the latter approach is that the light can scatter and blur the image.
Amorphous selenium has been used as the semiconductor in direct-conversion flat-panel detectors to obtain mammograms4. This application demonstrated that such detectors can produce much sharper images than can indirect-conversion instruments. But although selenium can absorb the low-energy X-rays used for mammography, it cannot adequately absorb the higher-energy X-rays needed for more-general applications. Moreover, neither amorphous selenium nor CsI can be used at the fundamental quantum-noise level — the dose at which individual quanta of X-rays are used to construct images, and which is lower than doses used by currently available flat-panel X-ray detectors.
Kim et al. therefore turned to a material called methylammonium lead triiodide (MAPbI3), which belongs to a remarkable family of semiconductors known as perovskites. The first solar cell made using a perovskite was reported just eight years ago5. Perovskites have since emerged as some of the most promising semiconducting materials for making solar cells, light-emitting diodes and lasers6. Their characteristics make them suitable for applications involving light from the near-infrared to the visible regions of the spectrum.
Unlike most other solar-cell materials, perovskites contain elements (lead and iodine) that have a high atomic number7,8, which makes them effective absorbers of the highly penetrating X-rays used in general radiology. The authors demonstrated that layers of MAPbI3 almost 1 millimetre thick — as needed to absorb X-rays in detectors — can be made easily using a printing method that works at a relatively low temperature. This method is cheaper than the high-temperature processes needed to prepare CsI or amorphous selenium, and should greatly reduce the cost of making imaging systems.
The authors then constructed an X-ray detector using 10-cm × 10-cm samples of MAPbI3 (Fig. 1). As expected, the material has a high sensitivity to X-rays — almost ten times higher than that of amorphous selenium or CsI. Using the same approach, it should be possible to make detectors that have much larger areas, as required for other medical procedures such as chest X-rays4. The sensitivity and processability of this perovskite could therefore reduce the radiation dose needed for medical X-rays to the fundamental quantum and spatial resolution limits.
More work on MAPbI3 X-ray detectors is now needed, for three main reasons. First, thermal activation of electrons in the material can produce a spurious signal known as dark current. This could be reduced by altering the material's electronic structure — by partly replacing iodine atoms with bromine or chlorine atoms6. Second, the sharpness of the images produced should be further increased, by, for example, increasing the applied electric field. And third, studies are needed to establish that the material is stable and long-lived enough to be made into a practical device.
“The potential improvements associated with using MAPbI3 could make a big difference to medical procedures.”
Nevertheless, the potential improvements associated with using MAPbI3 could make a big difference to medical procedures. For example, stents are often used to hold open blocked coronary arteries, the blood vessels that feed the heart muscle. The stents are inserted non-invasively using catheters, guided by fluoroscopy (an X-ray video technique)4. X-ray detectors made using MAPbI3 should greatly lower the radiation dose used in this procedure, and improve the images of the fine wires that make up the stent. More broadly, the advantages of MAPbI3 detectors could lead to the replacement of existing technologies, and could fuel innovation in other X-ray imaging techniques, such as computed tomography.
Kim, Y. C. et al. Nature 550, 87–91 (2017).
Reiner, B. I., Siegel, E. L. & Siddiqui, K. J. Digit. Imaging 16, 324–330 (2003).
Rowlands, J. A. Phys. Med. Biol. 47, R123–R166 (2002).
Yorkston, J. & Rowlands, J. in Handbook of Medical Imaging Vol. 1 (eds Van Metter, R. L., Beutel, J. & Kundel, H. L.) 223–328 (SPIE Press, 2000).
Kojima, A., Teshima, K., Shirai, Y. & Miyasaka, T. J. Am. Chem. Soc. 131, 6050–6051 (2009).
Sutherland, B. R. & Sargent, E. H. Nature Photon. 10, 295–302 (2016).
Yakunin, S. et al. Nature Photon. 9, 444–449 (2015).
Shrestha, S. et al. Nature Photon. 11, 436–440 (2017).
Related links in Nature Research
About this article
Journal of Semiconductors (2020)
Efficient X-ray Attenuation Lead-Free AgBi2I7 Halide Rudorffite Alternative for Sensitive and Stable X-ray Detection
The Journal of Physical Chemistry Letters (2020)
Applied Physics Letters (2020)
Environmentally friendly CsPbBr3 QDs multicomponent glass with super-stability for optoelectronic devices and up-converted lasing
Journal of the European Ceramic Society (2020)
Light: Science & Applications (2020)