Imeka Harnesses AI and Nextflow to Gain Insights From Brain Scans to Help Select Treatments
Read the full case studyAim
Scalable automated pipelines for image analysis and GPU-based machine learning algorithms to accelerate analysis and help pharmaceutical companies select optimal treatments.Challenges
- →Support for containerized applications.
- →Critical requirement for reproducibility.
- →Large datasets, many parallel compute steps.
- →Compute-intensive GPU-based workloads.
— Felix Morency, CTO at Imeka
Solution
Nextflow
Manage containerized workflows.
PyTorch
Train predictive models
Seqera Services
Implement solutions quickly and easily overcome technical hurdles
Results
- →Increased productivity: By using fast, high-throughput pipelines that maximize parallelism, Imeka has been able to increase their research productivity. Also, the need for manual intervention is minimized saving time and avoiding errors. Nextflow has helped automate the end-to-end analysis process including image processing and analysis, quality checks, statistical processing, and inference.
- →Simplified integration. Nextflow’s support for existing scripting languages, containers, and its dataflow programming model enabled Imeka to easily adapt existing workflows and containers to the Nextflow environment. Features such as caching and resume functionality simplified pipeline development, by enabling developers to quickly identify and resolve issues without the need to restart long-running analysis pipelines from scratch.
- →Regulatory compliance. By using Nextflow, and taking advantage of its close integration with GitHub and support for versioned containers, Imeka is able to ensure reproducibility and auditability — critical for meeting strict regulatory requirements.
- →Improved quality. By expressing their workflows in Nextflow, Imeka has been able to automate a variety of processes and gain confidence in the validity of datasets and results. They have also been able to increase the quality of their predictive models by training them on the higher-quality datasets, ultimately resulting in better treatment options for patients.
About
Imeka is based in Sherbrooke, Quebec, Canada and has a brain imaging technology platform specializing in microstructure and white matter connectivity. Imeka offers pharmaceutical companies the ability to combine diffusion MRI imaging and AI to map white matter integrity and understand the central nervous system morphology of patients. Imeka’s technology platform offers pharmaceutical and biotech companies the possibility to reconstruct the brain’s fiber network, providing insights into the white matter and the central nervous system, including the spinal cord.
To learn more, visit https://www.imeka.ca/