Optics-Free Bioluminescence Detection on a CMOS Camera Chip

Abstract

This thesis explores the development of a compact, optics free and low-cost biosensing platform based on direct on-chip detection of bioluminescence using a complementary metal–oxide–semiconductor (CMOS) image sensor. This work is motivated by the need and challenge of combining the high sensitivity of optical biosensing in laboratories with the simplicity and portability required for point-of-care applications. More specifically, we calibrated a commercial CMOS camera sensor and evaluated it as a quantitative detector for low-light measurements by characterizing its responsivity, linearity, noise floor, and saturation behavior under controlled LED illumination. In parallel, we expressed, purified and characterized a bioluminescent reporter system based on NanoLuc luciferase using UV–Vis spectroscopy and SDS–PAGE analysis. Lastly, we integrated these two elements into a biosensing device using accessible methods, including 3D-printed components and a PDMS sample well positioned directly above the sensor surface area. Baseline experiments demonstrated successful direct detection of NanoLuc bioluminescence down to 10nM, using the CMOS platform, while a microscopy-based setup served as a reference system for comparison. To improve detection sensitivity under weak-signal conditions below 10 nM, we proposed and demonstrated a signal enhancement approach based on lock-in amplification using a microfluidic modulation platform in a microscope-based setup. These findings highlight the potential of modulation-based signal processingfor future highly sensitive and portable optical biosensing applications.

Popular Abstract

During the COVID-19 pandemic, rapid antigen tests introduced many people to biosensors, although (knowingly or not) people had already been using biosensors such as pregnancy tests or smartwatches that monitor heartbeat for years. In fact, any system that combines a biological recognition element with a transducer that converts detection into a readable signal is, by definition, a biosensor. To explain this with the COVID-19 test example, in this case the biological recognition element is an antibody tagged with a gold nanoparticle, that specifically binds to molecules from the virus (important note: please do not open a COVID-19 test trying to retrieve the gold nanoparticles). When enough of these virus - antibody - gold systems accumulate on the test strip, they form the visible red line indicating a positive result. Interestingly, gold nanoparticles are not actually gold-colored at all, but can appear red, blue, or green depending on their size and arrangement (the bright colors in stained-glass church windows are a centuries-old example of this effect). Back to the transducer element, which, in this case, is simply the human eye, that detects and interprets the red line. The human eye is remarkably sensitive in low-light conditions (and probably the cheapest detector you can find — it comes for free!). It can usually tell us whethersomething is present or not, and sometimes even whether there is “a little” or “a lot”
of it. However, the eye is not truly a quantitative detector. In many medical situations, doctors care not only about whether a biomarker exists, but also about its concentration and how it changes over time, since small changes can reveal whether a disease is progressing or whether a treatment is working. Today, highly sensitive quantitative biological measurements are already possible using advanced laboratory techniques such as Polymerase Chain Reaction (PCR) and fluorescence microscopy. The problem is that these methods often require expensive equipment, specialized laboratories, and trained personnel. Because of this, they are usually only used once symptoms become serious enough to justify complex testing. This is not ideal for diseases such as cancer or sepsis, which may begin with extremely small biological changes and vague symptoms that initially look no different from a common cold or simple fatigue. Of course, it would be impossible, both practically and economically, to send every patient with generic symptoms for highly specialized laboratory analysis. This creates a gap between highly sensitive laboratory diagnostics and simple point-of-care tests that can be used quickly and routinely in everyday healthcare. This thesis explores a possible middle ground by combining bioluminescence with a CMOS camera, similar to the ones found in smartphones, to create a simple optical biosensing platform. In this system, bioluminescence acts as the biological signal generation mechanism, while the CMOS camera, adapted for low-light detection, converts light signals into measurable digital information. Bioluminescence produces light directly through a chemical reaction, similar to how fireflies glow in nature. Since the light is generated internally, there is no need for external light sources such as lasers or complicated optical components and alignment, making the system potentially simpler, cheaper and more portable, than most techniques used in laboratories nowadays. To investigate this idea, we expressed a highly bright bioluminescent protein called NanoLuc and integrated it into a custombuilt sensing platform positioned directly above the CMOS camera (see super cool Fig. IV.2a). The results of this thesis showed that relatively weak bioluminescent signals could be detected directly on the sensor. In addition, a proof-of-concept signal enhancement approach, called lock-in amplification, was explored to investigate whether sensitivity could be further improved under extremely low-light conditions.
In the long term, systems like this could contribute to portable biosensors capable of detecting diseases earlier and more routinely outside specialized laboratories. Beyond medical diagnostics, similar technologies could also find applications in areas such as food safety and environmental monitoring. Imagine a future where biosensors like these become as integrated into everyday life as pregnancy tests are today. Making reliable scientific measurements more accessible in everyday life could not only improve quality of life, but also help strengthen public trust and interest in science at a time when misinformation and pseudoscience are increasingly widespread.

• author: Christoforidou, Anna
• supervisor: Heiner Linke; Nils Gustafsson
• organization: Solid State Physics; Department of Physics
• course: PHYM03 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Biosensing, Bioluminescence, Point-of-Care device, CMOS camera chip.
• language: English
• id: 9230333
• date added to LUP: 2026-06-11 12:34:57
• date last changed: 2026-06-11 12:34:57