Rad AI spent the week showcasing its growing expertise in photon-counting CT detector physics, emphasizing the distinction between charge sharing and fluorescence escape in CdTe and CdZnTe sensors. Across several technical LinkedIn posts, the company warned that conflating these effects can distort detector models, spectral corrections, and material decomposition accuracy.
Claim 30% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The posts stress that charge sharing, driven by electron cloud diffusion, conserves energy across neighboring pixels and can be mitigated with charge summing, while fluorescence escape produces discrete escape peaks that require spectral unfolding. Rad AI highlighted that fluorescence effects above roughly 27 keV are non-trivial at clinical CT energies, with implications for energy bin selection and quantitative imaging performance.
Complementing this discussion, Rad AI detailed simulation work using CERN’s Allpix-squared framework to model charge sharing under CT-relevant conditions. In 100,000 simulated photon interactions, 41.1% involved charge sharing, including 32.3% adjacent bipixel events, with the distribution matching Gaussian charge cloud diffusion expectations.
The company noted that such levels of charge sharing can degrade energy resolution and introduce spectral distortion, potentially affecting diagnostic image quality in photon-counting CT. By quantifying these effects and openly engaging the technical community, Rad AI appears to be investing in high-fidelity detector modeling capabilities rather than announcing immediate commercial products.
From an investor perspective, the week’s activity underscores Rad AI’s focus on R&D-driven differentiation in advanced medical imaging. Deep domain expertise in detector physics and simulation could strengthen its positioning for future partnerships with imaging OEMs, software integrations, and high-end CT workflows as photon-counting systems move toward broader commercialization.
If successfully translated into algorithms, correction software, or design guidance, this physics-informed approach may support better image quality, regulatory readiness, and hospital adoption in spectral CT. Overall, the week highlighted Rad AI’s strategic emphasis on detector physics accuracy as a potential long-term competitive lever in the evolving photon-counting CT market.