Superconducting Nanowire Single-Photon Detectors for Enhanced Biomedical Imaging

TL;DR

This work addresses the photon-starved nature of biophotonics and proposes superconducting nanowire single-photon detectors (SNSPDs) as a detection-path solution to extend imaging depth and fidelity. It frames SNSPDs against traditional detectors, outlining their operating principle, key metrics, and how they enable innovations in SWIR/NIR-II imaging, single-photon spectroscopy, and time-resolved fluorescence techniques, including quantum-enabled modalities. Core contributions include a systematic discussion of current opportunities, such as SWIR deep-tissue imaging, DCS and spectroscopy, high-resolution FLIM, and quantum-inspired approaches, as well as a candid assessment of integration barriers—optical coupling, large-area arrays, cryogenics, polarization, and dynamic range. The paper highlights practical progress toward clinical translation through large SNSPD arrays, compact cryogenic solutions, and device geometries that mitigate polarization sensitivity and latching, arguing that SNSPDs can enable gentle, ultralow-dose imaging and new biomedical capabilities. Overall, SNSPDs promise near-unity detection efficiency and picosecond timing across broad spectral ranges, potentially redefining photon-limited biomedical imaging and enabling clinical translation as system integration matures, including synergy with structured-light excitation and wavefront shaping.

Abstract

Significance: Superconducting nanowire single-photon detectors (SNSPDs; also known as SSPDs) show enormous promise for low-light biomedical imaging by offering exceptional sensitivity, picosecond timing resolution, and broad spectral coverage. Aim: This perspective evaluates the role of SNSPDs by comparing their performance with other photon-counting detectors for emerging biomedical imaging applications. Approach: We outline the need for ultrasensitive detectors for biophotonics, summarize SNSPD operating principles and compare their performance with established photon-counting devices. We highlight applications in which SNSPDs enable new imaging capabilities and discuss system-level challenges and technological developments that are critical to future applications, including clinical translation. Results: SNSPDs offer advantages in signal-to-noise ratio, temporal precision, and detection bandwidth, enabling deeper tissue imaging, high-precision fluorescence lifetime measurements, and quantum-enhanced imaging modalities. Advances in scalable arrays, cryogenic miniaturization, and improved signal collection are reducing barriers to widespread adoption. Conclusions: SNSPDs are poised to transform photon-limited biomedical imaging. As device performance and system integration continue to advance, their adoption in imaging platforms is expected to accelerate. Combining SNSPDs with advancements in the excitation pathway, such as structured-light excitation with Bessel beams, aberration correction, and wavefront shaping, shows promise for delivering unprecedented imaging capabilities and broadening both the preclinical and clinical utility of these detectors.

Figures (5)

• Figure 1: Working principle of an SNSPD at a macroscopic level, looking at a cross-section of the nanowire. (a) The nanowire is stabilized below the critical temperature and biased with a current (depicted as white arrows), keeping it in a superconducting state. (b) A photon is absorbed, creating a small resistive hotspot. (c) The current is diverted around the hotspot, flowing along the outer edge of the nanowire. (d) As the current diversion continues, the local current density surrounding the hotspot increases until it exceeds the critical current density required for superconductivity. (e) This results in the formation of a resistive barrier across the entire width of the nanowire. The resistance rises rapidly, resulting in a measurable voltage pulse. The current flow is blocked, and the external circuit is used to shunt the bias current. (f) The reduced current allows the resistive region to cool and collapse, returning the nanowire to a fully superconducting state. This diagram was inspired by Natarajan et al. natarajan_superconducting_2012 and Lau et al. lau_superconducting_2023 respectively.
• Figure 2: SWIR confocal imaging using an SNSPD. (a) Volumetric images of blood vessels in an intact mouse head (from scalp to cortex), acquired with 5 $\mu\text{m}$ axial scan increments. A 1.65 $\mu$m laser was used for excitation, and fluorescence was collected in the 1.8-2 $\mu$m window. Imaging was performed 30 minutes after intravenous injection of quantum dots. (b) High-resolution confocal images at various depths using PMT and SNSPD detectors. (i) PMT with 1.32 $\mu$m excitation and 1.5-1.7 $\mu$m collection. (ii) SNSPD with the same excitation and collection wavelengths as the previous panel. (iii) SNSPD with 1.65 $\mu$m excitation and 1.8-2 $\mu$m collection. All lasers operated at 28.5 mW at the mouse head surface. Results are representative of three mice (BALB/c, female, 3 weeks old). (c) Stitched xz scan (4,500 $\mu \text{m}$$\times$ 512 $\mu \text{m}$) of mouse colon tissue using SNSPD with 1 mW excitation. The 512 $\mu \text{m}$ imaging depth spans the mucosa, submucosa, and muscular layers. Data are composed of 16 individual xy-plane scans. (d) Stitched xy scan (2.3 mm $\times$ 0.95 mm) of colon tissue at a depth of approximately 100 $\mu \text{m}$. Credit: Panels a-b adapted with permission from wang_vivo_2022 © Springer Nature Ltd.; panels c-d adapted with permission from liu_superconducting_2024 © Optica Publishing Group.
• Figure 3: (a) Schematic of the near-infrared (NIR) confocal microscope equipped with wavefront shaping and superconducting nanowire single-photon detector readout. The excitation path consists of a 1.31 $\mu$m continuous-wave laser, beam expander, spatial light modulator (SLM), dichroic mirror (DM), xy galvanometric scanner (G), and objective lens (Obj), which focuses light onto the sample. The collection path comprises the objective (Obj), x-y galvanometric scanner (G), dichroic mirror (DM), polarizing beam splitter (PBS), achromatic doublet, single-mode fiber (SMF), and SNSPD. The excitation path is represented in orange and the emission path is represented in red. (b) Depth-resolved confocal fluorescence images of a monolayer of NIR quantum dots overlaid with successive layers of scattering phantom tissue added in 400 $\mu\text{m}$ increments. Images were acquired under Bessel beam and Gaussian beam illumination. Scale bar 40 $\mu$m
• Figure 4: In vivo through-skull vasculature imaging of a cerebral ischemia mouse, where thrombosis was photochemically induced. (a) In vivo time-lapse imaging. Scale bar 1 mm. (b) The maximum intensity projection image of the cerebral vasculature from depths of 0 mm to 2.5 mm. Scale bar 1 mm. Panels a-b adapted with permission from Liao et al.liao_depth-resolved_2020 © Optica Publishing Group.
• Figure 5: NIR-II confocal fluorescence lifetime imaging demonstrated using an SNSPD with a quantum efficiency centered at 1064 nm. Cultured C6 rat glioma cells were stained with (a,c) indocyanine green (ICG; fluorescence lifetime of 144.8 ps) or (b,d) human serum albumin-ICG. Panels (a) and (b) show NIR-II fluorescence intensity images, in which the two fluorophores are indistinguishable. Panels (c) and (d) display color-coded fluorescence lifetime images, allowing clear differentiation between the two fluorophores. Panels (e) and (f) present fluorescence decay curves of the fluorophores in cellular and solution environments. Scale bars in (a)–(d) are $20~\mu$m, figure adapted with permission from Yu et al © Optica Publishing Group yu_intravital_2020.