Prototypical Clustering Networks for Dermatological Disease Diagnosis

Abstract

We consider the problem of image classification for the purpose of aiding doctors in dermatological diagnosis. Dermatological diagnosis poses two major challenges for standard off-the-shelf techniques: First, the data distribution is typically extremely long tailed. Second, intra-class variability is often large. To address the first issue, we formulate the problem as low-shot learning, where once deployed, a base classifier must rapidly generalize to diagnose novel conditions given very few labeled examples. To model diverse classes effectively, we propose Prototypical Clustering Networks (PCN), an extension to Prototypical Networks that learns a mixture of prototypes for each class. Prototypes are initialized for each class via clustering and refined via an online update scheme. Classification is performed by measuring similarity to a weighted combination of prototypes within a class, where the weights are the inferred cluster responsibilities. We demonstrate the strengths of our approach in effective diagnosis on a realistic dataset of dermatological conditions.

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Prototypical Networks for Few-shot Learning
• A systematic study of the class imbalance problem in convolutional neural networks
• Matching Networks for One Shot Learning
• One-shot Learning with Memory-Augmented Neural Networks
• The Devil is in the Tails: Fine-grained Classification in the Wild
• Gaussian Prototypical Networks for Few-Shot Learning on Omniglot

Cited by in corpus (6)

• A deep learning system for differential diagnosis of skin diseases
• Does Your Dermatology Classifier Know What It Doesn't Know? Detecting the Long-Tail of Unseen Conditions
• Meta-Learning with Fewer Tasks through Task Interpolation
• No True State-of-the-Art? OOD Detection Methods are Inconsistent across Datasets
• Federated Reconnaissance: Efficient, Distributed, Class-Incremental Learning
• Using Multitask Learning to Improve 12-Lead Electrocardiogram Classification