Beyond the 100-cell differential: what a Munich hematology lab wants from automation
Recorded at the Scopio Labs workshop, EuroMedLab 2025. Dr. Mathias Brügel, Deputy Director of the Institute of Laboratory Medicine at LMU Hospital Munich, on why manual microscopy still sets the ceiling in hematology, what digital morphology did and did not fix, and the first preliminary signals from the Complete Blood Morphology (CBM®) multi-center study.
Watch the full session
Chapters
- 12:23 Why manual microscopy limits the lab
- 27:28 Beyond decision support: what an autonomous analyzer is
- 30:36 Why counting more cells changes sensitivity
- 33:30 From cell-oriented to disease-oriented morphology
- 38:34 Q&A: why LMU has not adopted pre-classification systems
A lab where almost nothing is routine
Some numbers set the context for everything Dr. Brügel said. LMU Hospital Munich has about 2,000 beds. Its Institute of Laboratory Medicine runs roughly 10 million tests a year, of which about 500,000 are blood counts. Around 35,000 of those end in a manual microscopy review.
That last number is the one that matters. It is a large volume of skilled work, and at a university hospital with a heavy oncology and hematology load, most of it lands on samples that are genuinely abnormal.
The problem with manual microscopy is not that it is slow
The usual case for automation in hematology is a throughput case. Dr. Brügel made a different one.
"Manual microscopy is not standardized. It relies on subjective classification, varies between observers, counts too few cells, is labor-intensive and time-consuming, and is limited by the human eye."
Read that list again and notice the order. Labor and time come fifth and sixth. The first three are about quality: no standard, subjective classification, variation between observers. The fourth, counting too few cells, is the statistical limit that sits underneath all of them. A 100-cell or 200-cell differential is a very small sample of what is on the slide, and everything downstream inherits that sampling error.
What Full-Field™ imaging is and why it matters
What digital morphology fixed at LMU, and what it did not
LMU was a site in Scopio's 2024 evaluation of the Full-Field™ Peripheral Blood Smear Application, with Dr. Brügel as principal investigator. In that evaluation, blast flagging reached 91% sensitivity at pre-classification and 95% after verification, and hands-on time came down by 20 to 30%.
But Dr. Brügel was direct about why LMU has not adopted a pre-classification system for daily work:
"We don't use today's digital imaging systems, because with so many pathological samples, every sample would still go to the review station."
This is the structural limit of decision support. A system that pre-classifies cells and waits for an expert to verify them makes the bench faster. It does not remove the bench. In a lab where nearly every sample is pathological, a faster bench that every sample still passes through is a smaller win than it sounds.
Full-Field™ Peripheral Blood Smear Application
A different category, not a faster version of the same thing
Dr. Brügel framed the CBM® Analyzer as a change of category rather than an improvement in degree:
"The idea is the first digital morphology analyzer that goes beyond decision support instruments and works as a true automatic morphology analyzer."
The mechanism behind that claim is sample size. The CBM® Analyzer sits downstream of the CBC analyzer and analyzes 1,000 WBCs, 10,000 RBCs and platelets across 50 fields of view per sample, roughly 10 times more cells than the current diagnostic standard. That expanded sample size is what makes autonomous reporting of routine samples statistically defensible, and it is what routes the samples that need expert attention, such as blasts, schistocytes and plasma cells, to a morphologist.
Complete Blood Morphology (CBM®) Analyzer
The first preliminary signals from the multi-center study
Dr. Brügel presented early data from the CBM multi-center study, in which LMU is a participating site. This data came from the training phase of the study and should be read as early insights, not final results.
The observation he highlighted: on samples where more than 1,000 cells were analyzed, agreement for blasts between the analyzer and a reviewer was 0.94, while agreement between two reviewers on the same material was 0.87.
That comparison is worth sitting with. The reference standard in morphology is another expert, and two experts looking at the same slide do not fully agree with each other. Counting more cells is one of the few levers that moves that number.
"Analyzing many more cells may capture populations that smaller manual reviews miss. The CBM analyzer has the potential to improve sensitivity in detecting uncommon or emerging morphologies."
He showed rare-cell examples from the study to illustrate the point: morphologies that are present on the slide but unlikely to be sampled in a 100-cell or 200-cell count.
From cell-oriented to disease-oriented analysis
Asked where this ends up, Dr. Brügel did not talk about turnaround time.
"The most interesting thing for the future is a switch from cell-oriented morphology analysis to a disease-oriented process."
Today a differential returns proportions of cell types, and a person assembles those proportions into a clinical picture. With thousands of quantified cells per sample and consistent measurement across shifts and sites, the raw material exists for morphology-based biomarkers: patterns across a population of cells rather than labels on individual ones. That is potential, not a cleared claim, and it is the reason the number of cells analyzed is the number to watch.
And on the technology itself:
"I think it will really change the future."
FAQ
What is the difference between digital cell morphology and autonomous morphology analysis?
Why is a 100-cell differential considered a limitation?
How many cells does the CBM® Analyzer analyze per sample?
Is the CBM data presented in this session final?
Is the CBM® Analyzer available today?
Who is Dr. Mathias Brügel?
See what 1,000 cells per sample looks like in your lab
Get in touch to learn about your lab's needs and see the CBM® Analyzer in a 30-minute session with our team.
Recorded at the Scopio Labs workshop, EuroMedLab 2025. CBM data presented in this session is preliminary and was collected during the training phase of the multi-center study.
Blast flagging sensitivity and hands-on time figures refer to the evaluation of Scopio's Full-Field™ Peripheral Blood Smear Application at LMU Hospital Munich, as presented by Dr. Brügel, and not to the CBM® Analyzer.
Scopio's CBM® is CE-IVDR certified. Pending FDA clearance in the US.
Final outputs may vary based on sample variability and final product design.
Scopio Labs' Full-Field Peripheral Blood Smear application is CE marked and FDA-cleared, and it's commercially available across the U.S., UK and Europe and other territories.