On-Chip Spectrometer Applies AI to Laboratory-Grade Sensing

Researchers at the University of California, Davis, have shrunk a lab-grade spectrometer down to the size of a grain of sand, making a tiny spectrometer-on-a-chip that can be integrated into portable devices. The solution addresses the bottleneck posed by the long physical path that results from obtaining measurement of the intensity of each color. The use of a prism or grating, and the long physical light path, makes spectrometers inherently bulky.

“We wanted to take this power [of a spectrometer] out of the lab and put it in your pocket,” said Ahasan Ahamed, a postdoctoral scientist in the lab of Saif Islam, professor and chair of the Department of Electrical and Computer Engineering, and the corresponding paper’s first author. 

In the current work, instead of physically separating each color, the device uses 16 distinct silicon detectors, each engineered to respond slightly differently to incoming light. This is analogous to giving a handful of specialized sensors a mixed drink, with each sensor sampling a different aspect of the drink. The key to deciphering the original recipe is the second part of the invention: artificial intelligence.

The heart of this innovation lies into technological breakthroughs. First, the researchers engineered the surfaces of standard silicon photodiodes with specialized photon-trapping surface textures. Silicon is typically effective at sensing visible light but is poor at sensing near-infrared light, which is critical for many applications. The photon-trapping surface forces near-infrared light to scatter within the thin silicon layer instead of passing straight through. This dramatically increases the likelihood that the silicon absorbs light, making the entire chip sensitive across a broad spectral range.

This microscopic chip uses photon-trapping surface nanostructures and AI to extend silicon's spectral range into the near-infrared, simulating the capabilities of a spectrometer. Courtesy of the University of California, Davis.

Beyond color detection, the design employs high-speed sensors to provide an inherent, ultra-fast capability for measuring photon lifetime. This temporal precision allows the device to capture fleeting light-matter interactions that are invisible to traditional instruments.

Second, the chip uses a powerful fully connected neural network. Since the 16 unique detectors only capture encoded, noisy signals, the AI is trained on thousands of examples to learn the complex, hidden relationship between the detectors' raw outputs and the original, pure light spectrum. The AI addresses this "inverse problem," reconstructing the light spectrum with high accuracy (around 8 nm resolution). This computational method completely removes the need for bulky optics. 

The system has a minimal footprint of 0.4 mm², high sensitivity, and strong noise resistance. The AI augmented chip can maintain signal clarity even in the presence of significant electrical interference, which is a major challenge in portable, low-cost electronics. By extending the sensing range of silicon into the near-infrared spectrum while enabling high performance through machine learning, this technology will bring integrated, real-time hyperspectral sensing across applications ranging from advanced medical diagnostics to environmental remote sensing.

“We are paving the way for a future where your watch or phone doesn’t just take pictures,” Islam said, the project’s principal investigator, “[but] analyzes the chemistry of the world around you.” 

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awarde

AI-Enhanced Biophotonic Nanostructures for Ultra-Sensitive Biomedical Imaging and Precision Therapeutics

 

Abstact

Biophotonics has emerged as a transformative interdisciplinary field combining optics, nanotechnology, and biomedical engineering to enable highly sensitive diagnostic and therapeutic technologies. Recent advances in nanostructured photonic materials and artificial intelligence (AI)-driven signal analysis have significantly improved imaging resolution, disease detection accuracy, and targeted therapeutic delivery.

This research proposes an integrated biophotonic platform based on plasmonic nanostructures, multimodal optical sensing, and AI-assisted data interpretation for early-stage disease diagnostics and precision phototherapy. Experimental simulations demonstrate enhanced photon–tissue interaction efficiency, leading to improved sensitivity in detecting cellular-level abnormalities. The proposed system offers a scalable and non-invasive approach suitable for real-time clinical applications, particularly in oncology, cardiovascular disease monitoring, and infectious disease diagnostics.

 The findings highlight the potential of next-generation intelligent biophotonic systems to revolutionize personalized medicine through rapid diagnostics, targeted treatment, and continuous health monitoring.

 Introduction

Biophotonics integrates photonic science with biological and medical applications to develop advanced technologies for disease detection, imaging, and therapy. Conventional diagnostic methods often face limitations in sensitivity, invasiveness, and response time, creating a demand for rapid and non-destructive biomedical technologies. Optical techniques such as fluorescence imaging, Raman spectroscopy, and plasmonic sensing have demonstrated exceptional capabilities in detecting molecular-level changes in biological tissues.

Recent progress in nanophotonics has enabled the development of engineered photonic nanostructures capable of enhancing light–matter interactions, thereby improving detection sensitivity and imaging contrast. Simultaneously, the integration of artificial intelligence into photonic systems enables automated interpretation of complex optical datasets, significantly improving diagnostic reliability and reducing clinical decision time.

This study explores an AI-assisted biophotonic sensing and imaging framework that combines plasmonic nanostructures, adaptive optical imaging modules, and intelligent signal-processing algorithms to achieve ultra-sensitive disease detection and targeted photonic therapy. The proposed approach aims to advance next-generation biomedical technologies that support early diagnosis, personalized treatment strategies, and continuous health monitoring.

 Methodology (Overview)

The research employs computational photonic modeling, nanostructure design optimization, and AI-based spectral data analysis to evaluate system performance. Plasmonic nanoparticle arrays are engineered to maximize optical field enhancement, while machine-learning algorithms process spectral variations associated with pathological biomarkers. Simulation studies assess imaging resolution, sensitivity improvement, and therapeutic light-delivery efficiency.

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awarde

Laser‑written glass chip pushes quantum communication toward practical deployment

 As quantum computers continue to advance, many of today’s encryption systems face the risk of becoming obsolete. A powerful alternative—quantum cryptography—offers security based on the laws of physics instead of computational difficulty. But to turn quantum communication into a practical technology, researchers need compact and reliable devices that can decode fragile quantum states carried by light.

A new study from teams at the University of Padua, Politecnico di Milano, and the CNR Institute for Photonics and Nanotechnologies shows how this goal can be approached using a simple material: borosilicate glass. As reported in Advanced Photonics, their work demonstrates a high‑performance quantum coherent receiver fabricated directly inside glass using femtosecond laser writing. The approach provides low optical loss, stable operation, and broad compatibility with existing fiber‑optic infrastructure—key factors for scaling quantum technologies beyond the laboratory.

Why glass?

Continuous‑variable (CV) quantum information processing—used in quantum key distribution (QKD) and quantum random number generation (QRNG)—relies on measuring the amplitude and phase of light waves. These measurements require a coherent receiver that combines a weak quantum signal with a stronger reference beam and analyzes their interference.

Most integrated receivers so far have been implemented on silicon. While silicon is well‑established and highly integrable, it suffers from polarization sensitivity and high optical losses, both of which reduce performance and stability in quantum communication systems.

Glass, on the other hand, is naturally polarization‑insensitive, extremely stable, and allows waveguides to be written in three dimensions with very low propagation loss. Using femtosecond laser micromachining, researchers can draw light‑guiding channels directly inside the material, creating compact photonic circuits without the fabrication complexity of semiconductor foundries.

Inside the laser‑written quantum receiver

The team fabricated a fully tunable heterodyne receiver—an essential component for CV‑QKD and CV‑QRNG—by writing the optical circuit directly into the volume of borosilicate glass. The chip includes:

• Fixed and tunable beam splitters
• Thermo‑optic phase shifters for precise electrical control
• Three‑dimensional waveguide crossings
• Polarization‑independent directional couplers

These elements allow the quantum signal and reference beam to interfere in a controlled manner so that two conjugate quadratures can be measured at once. The device also demonstrates:

• Extremely low insertion loss (≈1 dB)
• Polarization‑independent operation
• Common‑mode rejection ratio above 73 dB, indicating strong suppression of classical noise
• High signal‑to‑noise stability over at least 8 hours of operation

These characteristics meet or exceed those of many silicon‑based photonic receivers.

Two quantum technologies on one chip

Because of the chip’s low loss, tunability, and stability, it can support multiple quantum communication tasks without hardware changes. Using the chip as a heterodyne detector, the team implemented a source‑device‑independent QRNG, meaning the system remains secure even if the incoming optical state is untrusted. Thanks to its high noise suppression and stable quadrature measurements, the chip achieved 42.7 Gbit/s secure random bit generation, which represents a record‑high rate for this security model.

The same device was used to implement a QPSK‑based CV‑QKD protocol, where information is encoded in a four‑point constellation of quantum states. Over a simulated 9.3‑km fiber link, the system reached a 3.2 Mbit/s secret key rate. This performance demonstrates that a glass‑based photonic front end can support state‑of‑the‑art CV‑QKD without the limitations associated with silicon platforms

Platform ready for real‑world deployment

Beyond performance metrics, the work underscores the inherent advantages of glass for integrated quantum photonics:

• Environmental stability: Glass is inert and resistant to thermal and mechanical fluctuations.
• Low‑loss fiber coupling: Waveguides closely match the size of standard telecom fibers.
• 3D design freedom: Circuits can include crossings and complex routing without added scattering.
• Scalability and cost‑effectiveness: Femtosecond laser writing enables rapid prototyping without expensive semiconductor processing steps.

These features support long‑term stability and resilience—qualities that could enable future use in field systems and even space‑based quantum communication missions. The authors emphasize that glass‑based integrated photonics may help bridge the gap between laboratory‑grade experiments and deployable quantum networks.

Leveraging these advantageous properties, the team demonstrated two core applications with the same chip: a source-device-independent QRNG, achieving a record-high secure generation rate of 42.7 Gbit/s, and a QPSK-based CV-QKD system, reaching a 3.2 Mbit/s secure key rate over a simulated fiber link of 9.3 kilometers.

Beyond these achievements, the work highlights the potential of glass-based integrated photonics as a robust and versatile platform for quantum technologies. Glass is inert, stable, and cost-effective, enabling the fabrication of devices inherently resistant to harsh environmental conditions. This novel approach could bridge the gap between laboratory prototypes and deployable quantum communication systems, marking a significant step toward a real-world quantum network infrastructure.

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awarde

Biophotonics imaging transforms studies of neuronal activities

 Yuehan Liu is a fifth-year doctoral candidate affiliated with the Biophotonics Imaging Technology Lab (BIT) advised by Xingde Li. She recently gave a talk at SPIE Photonics West BiOS entitled "Two-photon fiberscope with a proactive optoelectrical commutator for rotational resistance-free neuroimaging in freely-behaving rodents." Her talk focused on the recent progress of non-invasive imaging technologies that could revolutionize the study of brain function and diseases.

Biophotonics is an interdisciplinary field that applies photonics — the branch of physics dealing with the creation, transmission, manipulation and reception of light — to biology-related studies, particularly in neuroscience. At the core of biophotonics is the use of photons and optical imaging techniques to study cells and tissue. Unlike traditional biopsy, which requires the extraction of sample cells for examination, biophotonics allows biological cells to be examined while keeping their integrity so that they can be monitored in real time.

"Biophotonics provides alternatives to traditional imaging methods like X-rays or ultrasound. The technology offers a real-time, non-invasive look at biological tissues — with less risk and higher resolution images than ultrasound — and offers advantages over other techniques that might be harmful or require longer processing times," Liu said in an interview with The News-Letter.

Historically, to image live rodents, researchers have had to fix their heads on a stationary bench top to be examined under microscopes. For studies of non-stationary behavior, their heads have to be fixed the entire time, which is suboptimal. Liu highlighted the critical need for advanced optical imaging tools capable of capturing detailed neuronal activity in live, freely moving mice.

"When neuroscientists edit certain genes in mice that are believed to influence behavior, learning or memory, they want to see how these changes manifest in the mice's behavior," she said. “Traditional methods, which require animals to be immobilized, drastically alter their natural behaviors and lead to skewed or unrepresentative data."

Liu's work with the FiberScope, a two-photon microscope, overcomes the constraints of traditional fixed imaging setups. With its compact design that fits all lenses and sensors inside a tube, the FiberScope is small and lightweight yet does not compromise on its ability to produce fast, high-resolution imaging comparable to standard-size microscopes. 

Moreover, Liu and her team have optimized the FiberScope to provide an enlarged field of view with increased scanning range and speed that allows fast and stable imaging of multiple planes. With it, scientists can now observe over a thousand neurons simultaneously, offering insights into neuronal networks in a naturalistic setting. 

"It's a revolutionary technique that allows us to watch neurons firing in living and freely moving rodents in real time, which is a game changer for studying the brain's communication pathways and animal behavior at the cellular level," Liu explained.

The work is laborious, and the success of such sophisticated optical instruments demands extreme precision for each minute component. 

"Making instruments like the FiberScope is more than designing a theoretical optical system; it demands a meticulous level of dexterity and precision that brings it to reality. The lenses we work with are tiny, and every single one needs to be aligned with exacting accuracy. A slight misalignment would render the entire system futile," she said.

Currently, the FiberScope is mostly used to aid basic neuroscience studies, diagnose diseases and create disease models. Yet, the implications of Liu's research extend beyond academic interest, as biophotonics holds promise for enhancing clinical practices, particularly in guiding neurosurgical procedures and improving early disease detection. Still, there is a long way to go from animal tests to being able to apply this technology to humans. 

"We're always pushing the boundaries of what's possible with biophotonics. It’s truly exciting that our research can facilitate neurological research to advance our understanding of the brain," Liu said.

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awarde

New CRISPR tool spreads through bacteria to disable antibiotic resistance genes

 Antibiotic resistance (AR) has steadily accelerated in recent years to become a global health crisis. As deadly bacteria evolve new ways to elude drug treatments for a variety of illnesses, a growing number of "superbugs" have emerged, ramping up estimates of more than 10 million worldwide deaths per year by 2050.Scientists are looking to recently developed technologies to address the pressing threat of antibiotic-resistant bacteria, which are known to flourish in hospital settings, sewage treatment areas, animal husbandry locations, and fish farms.

University of California San Diego scientists have now applied cutting-edge genetics tools to counteract antibiotic resistanceThe laboratories of UC San Diego School of Biological Sciences Professors Ethan Bier and Justin Meyer have collaborated on a novel method of removing antibiotic-resistant elements from populations of bacteria.

The researchers developed a new CRISPR-based technology similar to gene drives, which are being applied in insect populations to disrupt the spread of harmful properties, such as parasites that cause malaria. The new Pro-Active Genetics (Pro-AG) tool called pPro-MobV is a second-generation technology that uses a similar approach to disable drug resistance in populations of bacteria.

"With pPro-MobV we have brought gene-drive thinking from insects to bacteria as a population engineering tool," said Bier, a faculty member in the Department of Cell and Developmental Biology. "With this new CRISPR-based technology we can take a few cells and let them go to neutralize AR in a large target population."

In 2019 Bier's lab collaborated with Professor Victor Nizet's group (UC San Diego School of Medicine) to develop the initial Pro-AG concept, in which a genetic cassette is introduced and copied between the genomes of bacteria to inactivate their antibiotic-resistant components. The cassette launches itself into an AR gene carried on plasmids, circular types of DNA that replicate within cells, thereby restoring sensitivity of the bacteria to antibiotic treatments

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awarde

Biophotonic probes for bio-detection and imaging

 Sensitive detection and imaging in bio-microenvironment is highly desired in biophotonic and biomedical applications. However, conventional photonic materials inevitably show incompatibility and invasiveness to bio-systems. To address this issue, Scientists in China reviewed recent progresses of biophotonic probes, including bio-lasers, biophotonic waveguides, and bio-microlenses, made from biological entities with inherent biocompatibility and minimal invasiveness, with applications for bio-detection and imaging. These biophotonic probes open up entirely new windows for biophotonic researches and biomedical applications.

The rapid development of biophotonics and biomedical sciences makes a high demand on photonic structures that are capable of manipulating light at small scales for sensitive detection of biological signals and precise imaging of cellular structures in bio-microenvironment. Unfortunately, conventional photonic structures based on artificial materials (either inorganic or toxic organic) inevitably show incompatibility and invasiveness when interfacing with biological systems.

The design of biophotonic probes from the abundant natural materials, particularly biological entities such as virus, cells and tissues, with the capability of multifunctional light manipulation at target sites can greatly increase the biocompatibility and minimizes the invasiveness to biological microenvironment.In a new paper published in Light Science & Application, a team of scientists, led by Professor Baojun Li and Professor Hongbao Xin from Institute of Nanophotonics, Jinan University, China, reviewed the intriguing progresses of emerging biophotonic probes made from biological entities, such as virus, bacteria, cells and tissues, for bio-detection and imaging.

They systematically reviewed three biophotonic probes with different optical functions, i.e., biological lasers for light generation, cell-based biophotonic waveguides for light transportation, and bio-microlenses for light modulation.To realize their potential biomedical applications of photonic probes, effective control and modulation of light generation are particularly important in various biochemical environments.

In this regard, the unique properties of light emitted by lasers, including high intensity, directionality and monochromatic emission, have rendered lasers one of the most useful tools in biomedical applications. Unlike traditional laser devices, bio-lasers utilize biological entities such as cells, tissues and virus, as part of the cavity and/or gain medium in a biological system. Bio-lasers can be categorized into three types, i.e., cell lasers, tissue lasers and virus lasers.

These bio-lasers avoid the biohazards of conventional laser devices. Since their optical output is tightly related to the biological structures and activities of the biological systems, bio-lasers can serve as highly sensitive tools in a range of biomedical applications, including cellular tagging and tracking, diagnostics, intracellular sensing, and novel imaging. For example, whispering gallery modes (WGM) microdisks with slightly different diameters resulted in obviously different lasing output spectra. Intracellular cell lasers realized by incorporating these microdisks into cells enabled tagging and tracking of individual cells from large cell populations at the same time.

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awardee

Novel technique to characterise chemical composition and structure of samples

 Raman spectroscopy is an essential technique used in the study of materials – including nanostructures – and biological systems to analyse their composition. Its applications range from the medical industry to planetary explorations.

Despite their popularity as a non-destructive, fast and efficient tool for the identification and verification of various substances, Raman spectrometers have historically been bulky and expensive. In an effort to make them smaller, affordable and capable of delivering actionable results, an initiative under the EU-funded IoSense project has developed a new system with an on-chip technology. It can be used to produce handheld scanners or even be incorporated into a smartphone.

A news release by project partner Interuniversitair Micro-Electronica Centrum (imec) states that existing handheld products in the market "fail to reach the desired performance for high-end applications largely because of the limited scaling capacity of conventional dispersive Raman spectrometry whereby scattered light is focused on a slit." It adds: "Scaling while maintaining high spectral resolution (< 1nm) means reducing the size of the slit which immediately limits the optical throughput. Thanks to a new concept for which a patent is pending imec has now been able to overcome this performance barrier."

The news item notes that "both high optical throughput and high spectral resolution can be reached in a miniaturized device," thanks to the "massive parallelization of waveguide interferometers integrated monolithically on top of a CMOS image sensor." It further says: "This novel system is built in imec's SiN [silicon nitride] biophotonics platform which guarantees robustness and compatibility with high-volume manufacturing."

Diverse applications

According to Pol Van Dorpe, principal member of the technical staff at imec, the areas where the new technology could be implemented include "food analysis, melanoma detection, or skin hydration. In the medical domain, we see opportunities for in-line measurements during surgery or endoscopy. And for space exploration, the ability to perform material analysis with a compact system is of tremendous value."

Raman spectroscopy, which uses the inelastic scattering of light falling on a material, is named after Sir Chandrasekhara Venkata Raman, the recipient of the Nobel Prize for Physics in 1930. The technique involves the analysis of vibrational, rotational and other low-frequency modes in a system. Light interacts with matter in various ways, transmitting through some materials, while reflecting or scattering off others. Both the material and the wavelength of the light have an impact on this interaction. Spectroscopy refers to the study of this light.

The IoSense (Flexible FE/BE Sensor Pilot Line for the Internet of Everything) project that supported part of imec's work was set up to develop "the base for increased manufacturing capacity for discrete and integrated sensors and sensor system solutions in Europe including design development and test for different key application oriented supply chains" as stated on the project website. IoSense targets several areas such as smart mobility society energy and health.

World Biophotonics Research Awards
Visit: biophotonicsresearch.com
Nominate Now: https://biophotonicsresearch.com/award-nomination/?ecategory=Awards&rcategory=Awardee