ETH Zurich researchers build PLATON detector that tracks particles in 3D with a single scintillator block

Researchers at ETH Zurich have developed a particle detector called PLATON that could replace millions of individual detector components with a single block of light-producing material, using a light-field camera, highly sensitive photon sensors, and AI to reconstruct particle paths in fast, detailed 3D. In conventional particle detectors, scintillator material is divided into vast numbers of small active segments to pinpoint where a charged particle passed and how it interacted. PLATON instead uses an unsegmented scintillator block. When a particle traverses it, the resulting light is captured by a light-field camera and photon sensors, and machine-learning algorithms reconstruct the three-dimensional trajectory. Simulations suggest the approach could match or surpass the performance of today's best detectors while being far easier to scale. The technology may also lead to sharper positron emission tomography (PET) medical scans. The work addresses a long-standing challenge in particle physics: detecting weakly interacting particles such as neutrinos and certain dark matter candidates requires large detector volumes with fine spatial resolution, but segmenting scintillators into millions of channels adds complexity and cost. By combining a monolithic scintillator with computational imaging and AI-based reconstruction, PLATON aims to decouple detector mass from readout channel count. ETH Zurich researchers say the detector concept could be valuable for future neutrino experiments, dark matter searches, and collider calorimetry, where scaling up traditional segmented designs becomes impractical.