Aicendence

Cytology reimagined

We enable automated cytological evaluations using Deep Learning Computer Vision.

Our Impact

A woman in a lab operating a microscope.

Problem

In diagnostics, analyzing pathology samples, especially cytology, is crucial but fraught with challenges. Manual processing not only requires a significant time commitment and is prone to errors, but hospitals and labs are also facing an immense shortage of skilled professionals.

A woman in a lab investigating a slide in front of a monitoring showing a digital representation of it.

Solution

Our solution is a digital platform providing easy and fast access to expert knowledge. Using Computer Vision and Machine Learning, our platforms enables automated and more accurate analysis of samples, improving throughput, availability and consistency.

An eye, visible through glasses. In the reflection you can see code lines on a monitor.

Vision

OUR DEDICATION IS

Our research

is our passion. We started by investigating lung diseases. Through cutting-edge, machine-learning-based biomarker extraction, we develop the next generation of cytological evaluation, allowing us to provide rare expert knowledge at any time.

ML based differential cell counts

ML based sub-phenotyping

New diagnostic options for bright field microscopy

The image shows the Cyto-ML workflow. From scanning the prepared slide to multi step classification from the AI models.

An illustration of the digital workflow. A circular arrow pointing from a glass slide, to scanner to a monitor.

All digital workflow

Digitisation creates opportunities

We believe that the future of cytology and pathology is digital. So are we. Our research creates solutions for the future - all digital - laboratory.

An illustration of a two stage AI system. First one separating rough cell classes and the second one subcategorising them.

Multi-stage specialized ML

divide and conquer

Foundation models provide incredible opportunities. Though, driven by their generalised data foundation, they can't deliver the precision necessary for high risk application. We focus on developing a chain of highly specialised ML models that perform small reliable actions in our analysis pipeline.

Publications

Our research that accompanies our enthusiasm for cytology and machine learning.

Publications

Abstract: Machine-learning based BAL cytology of lung transplant patients

Michael Gerckens, Carlo Muemmler, Johannes Strodel, Gerald Burgstaller, Jürgen Behr, Nikolaus Kneidinger

European Respiratory Journal Suppl. 2023
‍

Our team

Our interdisciplinary team consists of experts for pulmonary disease and respiratory cytology in both human and veterinary medicine, as well as exceptional computer scientists for deep learning computer vision.

A picture showing the team member Johannes Strodel.

Johannes Strodel, M.Sc.

Software Development & ML

A picture showing the team member Moritz Koch.

Moritz Koch, M.Sc.

Software Development & ML

Advisory board

Deputy Director of the Polyclinic V for Pneumology at the LMU University Hospital

Head of the Lung Transplantation and Interstitial Lung Diseases Departments· Supervises the second most lung transplants in Germany

Cited thousands of times in the world's most renowned journals

20+ years' experience in study design as a physician and scientist