Startup introduces a range of SPAD-based image sensors

Singular Photonics emerged from stealth mode, launching a new generation of image sensors based on SPADs. A spin-out from the University of Edinburgh lab of digital imaging pioneer Professor Robert Henderson, Singular is one of the first companies to bring advanced computation to SPAD-based image sensing, enabling in-pixel and cross-pixel storage and computations at the lowest light levels to reveal previously invisible details of the material world and its photon events.

SPADs use the 'avalanche' effect in semiconductors to convert light directly into an electrical current without cooling or amplifying. While most commercial SPAD-based image sensors have been limited to time-resolved counting of photons, the company's core innovation lies in complex layers of computation beneath 3D-stacked SPAD sensors, comparable to the way FPGAs and GPUs revolutionised parallel computing by conducting high-speed, localised processing.

Prof Henderson leads the University of Edinburgh's CMOS Sensors and Systems Group. In 2005, he designed one of the first SPAD image sensors in nanometer CMOS technologies, leading to the first time-of-flight sensors in 2013, which today perform an autofocus-assist feature in more than a billion smartphones worldwide.

"There can be no doubt that SPAD sensors are the future of digital imaging, but their use to date in commercial devices hasn't extended much beyond time-resolved counting of photons," said Prof Henderson. "Computational cleverness can be the difference. We are building next-generation imaging sensors, where the computation is done digitally at the pixel level – exactly where the photons arrive."

Simultaneously capturing depth and temporal dimensions to generate 4D images that deliver deep, data-rich insights, the company's noiseless sensors allow more information to be extracted from light, supporting applications ranging from consumer and automotive electronics to the scientific and medical fields. The company's approach transforms SPAD sensors into 3D stacked computational engines capable of performing a wide range of sophisticated tasks, such as real-time photon counting, timing, and advanced processing techniques, including in-pixel histograms, statistical analysis and autocorrelation.

The company is launching two sensors, both of which are available now:

Andarta, developed in collaboration with tech giant Meta, has a miniature form factor combined with high sensitivity and is optimised for usage in several medical imaging modalities. The sensor supports multiple modes of operation, including in-pixel autocorrelation measurements, and is a significant step closer to SPAD integration in the wearables space. For example, Andarta allows monitoring of the cerebral blood flow rate, monitoring rapid fluctuations in light as it passes through tissue, at depths not currently possible with current sensors.

Sirona, the company's first product, is a 512-pixel SPAD-based line sensor capable of time-correlated single photon counting (TCSPC) and enabling Raman spectroscopy, fluorescence lifetime imaging microscopy (FLIM), time-of-flight, and quantum applications. With on-chip histogramming and time-binning capability, the sensor has the potential to revolutionise spectroscopy applications.

"We are in a unique position where we already have commercially available products and are generating revenue in our first year of incorporation," said Shahida Imani, CEO of Singular Photonics. "With new, even more advanced sensors coming to the market in 2025, we are well positioned to lead the SPAD-driven imaging revolution."