Skip to main content
CryptInfer: enabling encrypted inference on skin lesion images for melanoma detection
ACM International Conference Proceeding Series
  • Nayna Jain, IIIT Bangalore & Ibm Systems
  • Karthik Nandakumar, Mohamed Bin Zayed University of Artificial Intelligence
  • Nalini Ratha, University at Buffalo, The State University of New York
  • Sharath Pankanti, Microsoft Corporation
  • Uttam Kumar, IIIT Bangalore
Document Type
Conference Proceeding

Deep learning models such as Convolutional Neural Networks (CNNs) have shown the potential to classify medical images for accurate diagnosis. These techniques will face regulatory compliance challenges related to privacy of user data, especially when they are deployed as a service on a cloud platform. Fully Homomorphic Encryption (FHE) can enable CNN inference on encrypted data and help mitigate such concerns. However, encrypted CNN inference faces the fundamental challenge of optimizing the computations to achieve an acceptable trade-off between accuracy and practical computational feasibility. Current approaches for encrypted CNN inference demonstrate feasibility typically on smaller images (e.g., MNIST and CIFAR-10 datasets) and shallow neural networks. This work is the first to show encrypted inference results on a real-world dataset for melanoma detection with large-sized images of skin lesions based on the Cheon-Kim-Kim-Song (CKKS) encryption scheme available in the open-source HElib library. The practical challenges related to encrypted inference are first analyzed and inference experiments are conducted on encrypted MNIST images to evaluate different optimization strategies and their role in determining the throughput and latency of the inference process. Using these insights, a modified LeNet-like architecture is designed and implemented to achieve the end goal of enabling encrypted inference on melanoma dataset. The results demonstrate that 80% classification accuracy can be achieved on encrypted skin lesion images (security of 106 bits) with a latency of 51 seconds for single image inference and a throughput of 18,000 images per hour for batched inference, which shows that privacy-preserving machine learning as a service (MLaaS) based on encrypted data is indeed practically feasible.

Publication Date
  • ciphertext packing,
  • Convolutional neural network,
  • homomorphic encryption,
  • melanoma,
  • multi-threading,
  • non-linear activation function,
  • optimization,
  • skin cancer

IR Deposit conditions: non-described

Citation Information
N. Jain, K. Nandakumar, N. Ratha, S. Pankanti and U. Kumar, "CryptInfer: enabling encrypted inference on skin lesion images for melanoma detection", in the First International Conference on AI-ML-Systems, AIMLSystems 2021, (Association for Computing Machinery), no. 13, p. 1-7, Oct. 2021. Available: 10.1145/3486001.3486233