PhD Studentship: AI-Enabled Smart Spectroscopy Pixels: Revealing Microbial Responses to Antibiotics

Spectroscopy has long offered a powerful window into nature, revealing the hidden structure and dynamics of matter through its interaction with light. Extending this window to individual microorganisms could transform understanding of how antibiotics act on living cells. Within apparently uniform populations, bacteria can respond very differently to treatment, with survival strategies concealed by measurements averaged across thousands of cells. Revealing this individuality requires new instruments that combine cellular precision, informative optical measurements and compact design.

The aim of this PhD is to develop intelligent spectroscopy pixels for investigating microbial responses to antibiotics, uniting physics, materials engineering, computer science and bioscience. Recent research demonstrated that a quantum device such as a single two dimensional perovskite photodetector can encode wavelength and light intensity in its nonlinear electrical response, with a compact neural network recovering this information.

This project will extend that principle towards biological sensing, combining advanced photodetectors with established microfluidic platforms that trap individual microorganisms, control antibiotic exposure and follow cellular behaviour over time.

The research will develop sensor quantum materials, optical and electronic interfaces, and learning algorithms to extract biological information from complex signals. Beginning with calibrated optical measurements and progressing towards individual-cell studies, it will investigate whether changes in transmitted or scattered light reveal antibiotic-induced changes in cellular state. Comparison with microscopy, growth, viability and recovery measurements will establish the biological meaning and reliability of these optical fingerprints.

The ambition is to uncover microbial responses early and without added labels, laying foundations for compact tools for future phenotypic antimicrobial-susceptibility assessment. The project offers a distinctive combination of training in quantum technologies, optical instrumentation, bacterial culturing, single-cell microfluidics, neural-network development and rigorous experimental validation. The balance between creating bespoke sensors, interfacing them with living systems and developing algorithms will be shaped by the doctoral researcher’s interests and strengths. This ability to connect quantum devices, instruments, computation and biology will prepare the researcher to contribute at the forefront of computational spectroscopy, AI-enabled biosensing, single-cell biotechnology and intelligent scientific instrumentation. Joint Engineering and Physical Sciences Research Council Doctoral Landscape Award supervision will support a coherent research programme and the development of skills relevant to both academic and industrial research.

References

1. K. J. Riisnaes et al., “Compact Variational Neural Networks for Spectral Inference from a Single Nonlinear 2D Perovskite Photodetector”, arXiv preprint arXiv:2608.27977 (2026).
2. U. Łapińska et al., “Fast bacterial growth reduces antibiotic accumulation and efficacy”, eLife 11, e74062 (2022).
3. Y. Zhang et al., “Single Cell Killing Kinetics Differentiate Phenotypic Bacterial Responses to Different Antibacterial Classes”, Microbiology Spectrum 11, e03667-22 (2023).

Funded by: EPSRC Engineering and Physical Sciences Research Council