Case Study: Janssen Pharmaceuticals achieves 10X faster deep learning model development for precision cancer medicine with Domino Data Lab

Delivering Precision Medicine for Cancer with 10X Faster Development of Deep Learning Models

Janssen set out to accelerate its precision-medicine programs for cancer by improving how deep learning models are developed and deployed to identify clinical-trial–eligible patients. The team faced a challenge of slow, hard-to-reproduce model training and tuning workflows that limited iteration, collaboration, and the ability to scale patient screening.

They built a flexible, distributed platform to run parallel experiments, automate hyperparameter sweeps, track metrics, and store models centrally for reproducibility and governance, with easy integration into data-science and production workflows. The result: a 10× reduction in training time, optimized models ready for deployment, and the potential to increase the number of patients screened as eligible for trials by 4×.

Janssen Pharmaceuticals

Peter Shen

Data Scientist