SizeID.bio
Advanced particle analytics for pharmaceutical investigations and comparability
Automated Particle Raman Analysis
Routine particle counting tells you how many particles are present. SizeID.bio adds morphology, chemical identity and source-relevant comparison, helping you understand what the particles are and what to investigate next.
Small sample sets are typically reported within 1–3 business days, including final review.
Population → Targeted Selection → Raman Analysis → Confirmed Identity
From a measured population to chemically confirmed particles
We do not only show isolated spectra. We connect a measured particle population to chemically confirmed individual particles.
52,410 total detected particles · 834 particles in the selected ≥5 µm analytical range · 250 Raman targets
The large-area scan detects the complete optical population. For this example, the Raman study focused on a defined ≥5 µm subset.
Targeted subset by particle length: 5–10 µm: 576 · 10–25 µm: 206 · ≥25 µm: 52
Trace one particle from the population to its identity
Particle 11292: Raman-confirmed polypropylene. Length 38.97 µm; width 23.24 µm. Locator image, particle detail and measured Raman spectrum remain linked.
CHEMICAL POPULATION
From individual spectra to a chemical population fingerprint
80.8% of the Raman-analyzed particles were PET, revealing a dominant material signature across the investigated population.
Supplier A · n = 156 Raman-analyzed particles
PET · 126 · 80.8%
Polystyrene · 17 · 10.9%
Epoxy · 7 · 4.5%
Cellulose · 4 · 2.6%
Polyethylene · 1 · 0.6%
Unknown ester · 1 · 0.6%
Identifying a particle as cellulose, polypropylene or PET may only be the first answer. SizeID.bio can compare unknown particles with customer-supplied process, packaging and device references to determine which reference fingerprint best explains the finding.
Reference-study result: six closely related cellulose source classes were distinguished with 96.4% accuracy on the held-out test dataset.
Unknown cellulose-type particle
Raman fingerprint and image-derived context guide reference comparison.
Reference library
Cellulose
Cotton fiber
Swab
Filter fiber
Lab-coat fiber
Paper fiber
Raman data → reference-trained classification
Probable source class
96.4%
Test-set classification accuracy
Balanced accuracy 96.7%
Cohen’s κ 0.957
Reference study using automated Raman to discriminate closely related cellulose source classes. Reported test metrics: CNN-LDA-PCA Raman model. Combined Raman and image-derived analysis is discussed separately as a future approach in the poster.
Source attribution strength depends on the available reference library, sampling design and process context.
Traditional database search
“What material is it?”
Fast first-pass chemical identification and a useful troubleshooting overview. Structurally similar cellulose sources can be difficult to distinguish reliably.
Reference-trained ML fingerprinting
“Which known source class does it most resemble?”
Higher discrimination between closely related reference materials, with additional training, planning and maintenance. Particularly useful for controlled in-house contamination libraries.
Reference study: Markus Lankers and Lia Ivanov, “Pinpointing Contamination Sources: Evaluating AI supported Spectroscopic Data Analysis Methods for Cellulose Fiber Contamination”, mibiC GmbH & Co. KG.
APPLICATIONS
Start with the particle question you need to answer
Foreign Particle Investigation
An unexpected count increase? Identify the material classes and focus the next investigation.
Parenteral & Combination Drug Products
Compare batches and material references to investigate product, packaging and device-related particles.
OINDP & MDRS
Build high-depth Test-vs.-Reference evidence for API size, morphology and composite structures.
DEEPMORPH
In nasal-suspension validation, DeepMorph was benchmarked against >30,000 fully Raman-mapped particles and achieved 95.0% particle-level classification accuracy.