Using deep learning image segmentation to prepare lung masks for CT scans of COVID-19 patients
Company Information
In the face of the global COVID-19 pandemic, a prominent healthcare organization sought to enhance its diagnostic capabilities through advanced medical imaging technology. Partnering with ProCogia, the initiative aimed to develop a cutting-edge deep learning model for lung mask preparation in CT scans of COVID-19 patients. This project was set to pioneer the use of image segmentation in 3D CT scans, significantly improving the accuracy and efficiency of COVID-19 diagnosis and research.
The Challenge
The primary challenge was the accurate segmentation of lung tissues in CT scans of COVID-19 patients, a critical step for diagnosing and understanding the disease’s progression. Existing models lacked the precision required, often failing to differentiate effectively between healthy and diseased lung tissue, especially in severe cases. The project demanded a solution that not only surpassed the performance of benchmark models but also provided a tool for rapid, reliable analysis to support ongoing research and patient care.
Procogia’s Approach
Hypothesis testing
Our approach was to develop the models by posing hypotheses and repeatedly testing them at different points in development (e.g., whether domain-inspired data augmentation would improve model performance)
Iterative approach
We developed several iterations of models using data that was partially annotated using previous iterations and corrected by hand
Domain knowledge
We liaised with the client to inform them of our data augmentation strategy and guide the criteria for model acceptance
Define milestones
We defined several important milestones which guided the model development and performance, including: over 95% average dice score in all cases, over 95% average dice on severe COVID cases, submit abstract based on work to a conference, and host model internally.
The Results
Our model performs with >95% average dice over all test cases, compared to benchmark models, for example, the state-of-the-art work by Hofmanninger, which achieved dice scores of around 90% on our data
Our model can be used to volumetrically distinguish inhalation and exhalation CT scans with >95% F1 score. This allowed other research teams to sort their data automatically, saving days or weeks of tedious manual work
Our paper was accepted at an international biomedical conference to discuss the benefits of the data augmentation approach we used. This gave us a chance to showcase our expertise and data findings any future sizes for which no historical information currently exists.
Our model is hosted internally, providing lung segmentation to the organization, expediting the use of this model by internal company research teams to accelerate their research.
Services Used
Services Used Data Science
We use open source technology to leverage the full potential of your data. Predictive and prescriptive results are actioned using AI and Machine Learning (ML).
Machine Learning
Drawing on statistical methods to enable improvement with experience, ProCogia’s machine learning algorithms predict outcomes and automate processes.
Data Engineering
ProCogia is partnered with the leading cloud providers, enabling our agnostic approach to focus on delivering tailored game-changing solutions for our clients.
Medical Imaging
Whereas off-the-shelf medical imaging models typically use 2D images, ProCogia’s model utilizes higher-quality image segmentation in 3D CT scans to seamlessly distinguish between healthy, diseased and consolidated lung tissue.
Conclusion
ProCogia’s development of a deep learning image segmentation model for COVID-19 CT scans represents a significant leap forward in medical diagnostics and research. By providing highly accurate lung tissue segmentation, the model enhances the understanding of COVID-19’s impact on the lungs, supports the healthcare sector in delivering precise diagnoses, and accelerates the pace of research. This project exemplifies ProCogia’s ability to combine technical innovation with practical application, offering powerful tools that meet urgent global health challenges.
Explore more stories
Dig deeper into data development by browsing our blogs…
Implemented Bookmarking Apps
A leading investment consulting firm faced challenges in managing and sharing complex data analyses through their internal tools developed with Shiny. As the firm sought
Optimizing and Scaling RESTful API for an Investment Consulting Firm
Optimizing and Scaling RESTful API for an Investment Consulting Firm Company Information A prominent investment consulting firm sought to elevate the performance and maintainability
Low-code solutions using Azure & Snowflake
ProCogia implemented low-code solutions using Azure Data Factory and Snowflake for the Performance Data Science team of a Leading Sports Organization
Get in Touch
Let us leverage your data so that you can make smarter decisions. Talk to our team of data experts today or fill in this form and we’ll be in touch.