Skin Lesion Classification Using Ensembles of Multi-Resolution EfficientNets with Meta Data

TL;DR

This paper presents an ensemble-based approach using multi-resolution EfficientNets and meta data to classify skin lesions, achieving top performance in the ISIC 2019 challenge.

Contribution

It introduces a novel ensemble method combining multi-resolution EfficientNets and meta data fusion for improved skin lesion classification.

Findings

Achieved top-ranked balanced accuracy of 63.6% and 63.4% on the challenge test set.

Effectively handled class imbalance with loss balancing techniques.

Utilized multi-resolution inputs and meta data fusion to enhance classification performance.

Abstract

In this paper, we describe our method for the ISIC 2019 Skin Lesion Classification Challenge. The challenge comes with two tasks. For task 1, skin lesions have to be classified based on dermoscopic images. For task 2, dermoscopic images and additional patient meta data have to be used. A diverse dataset of 25000 images was provided for training, containing images from eight classes. The final test set contains an additional, unknown class. We address this challenging problem with a simple, data driven approach by including external data with skin lesions types that are not present in the training set. Furthermore, multi-class skin lesion classification comes with the problem of severe class imbalance. We try to overcome this problem by using loss balancing. Also, the dataset contains images with very different resolutions. We take care of this property by considering different model input resolutions and different cropping strategies. To incorporate meta data such as age, anatomical site, and sex, we use an additional dense neural network and fuse its features with the CNN. We aggregate all our models with an ensembling strategy where we search for the optimal subset of models. Our best ensemble achieves a balanced accuracy of 74.2% using five-fold cross-validation. On the official test set our method is ranked first for both tasks with a balanced accuracy of 63.6% for task 1 and 63.4% for task 2.