Lia Use Cases

Embryology, digital pathology and more. See some of the example usecases for LIA.

Embryology

A major part of an embryologist’s work is performed in a laboratory and involves repetitive, tiring and labour intensive tasks. Image analysis automation offers a convenient way to relieve embryologists from manual work and enable them to focus on the patient and her family. LIA can be easily integrated into the microscopic embryo examination workflow and significantly speed up the evaluation process while maintaining human expert-level accuracy.
Cognexa developed an automated & cognitive human embryo annotation and evaluation system that provides embryologists with additional, objective, consistent and interpretable clinical data which can assist their decision-making and improve IVF treatment results. Our solution resulted in fewer cycles needed for successful conception which not only decreased the emotional stress of patients experienced during the IVF process, but also lowered their level of health risks, and eased the financial burden put on them and on health insurers.

Endocrine pathology

LIA can be utilized to perform analysis of microscopic images to automatically evaluate the quality of pancreatic islets. This approach reliably identifies pancreatic islets among other types of tissue and carries out an estimation of their total mass. This is a crucial piece of information when choosing the right donors and acceptors for islet transplantation.
In partnership with an established healthcare institution in Prague, Cognexa has developed a deep-learning network built and trained in cxflow that automatically segments and evaluates the quality of pancreatic (Langerhans) islets and which can now be accessed via www.islenet.com. The software automatically scans, analyzes and evaluates all images and promptly generates reports for the endocrinologist, thus saving hundreds of hours of manual lab work.

Digital pathology

Deep learning has demonstrated its usability in the automated analysis of digital images generated by whole-slide imaging systems, as well as in the fields of quantitative pathology and digital histology.
Digital pathology offers the potential to improve diagnostic efficiency and reduce turnaround time to report clinical cases. There is also the potential for increased output quality by reducing error rates through automated systems designed for:
  • Object quantification
  • Segmentation based on morphology
  • Rare object identification
LIA is a software application that can be adjusted and trained to perform any visual evaluation of whole-slide images and other digital image types. In addition, our team can also provide support for any of the three steps involved in digital image processing:
  1. Image storage and management.
  2. Image manipulation and editing.
  3. Image viewing, transmission and sharing.

Dermatopathology

Very often clinical dermatologists, unsure of the diagnosis, send biopsy samples to the laboratory where trained dermatopathologists carry out further examinations. These skin samples are collected and examined under a microscope or are subject to other molecular tests. That process reveals the histology of the disease and renders a specific diagnostic interpretation. In some cases, additional specialized testing needs to be performed including immunofluorescence, immunohistochemistry, and electron microscopy.
LIA can be trained and used in a laboratory to enhance and streamline the image analysis process by making it more efficient, less time-consuming and less burdensome. LIA allows you to develop customized algorithms in just days, thus taking your dermatopathology to the next level.

Free consultation

Our consultants will help you identify the steps of your process where LIA could be useful and propose a custom solution designed to work in your specific case and to easily integrate into your workflow. For a free initial consultation, please contact us here...

Contact us
X
*
*
*
*
*
*Required fields
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Subscribe to newsletter