# Skin Lesion Diagnosis using Ensembles, Unscaled Multi-Crop Evaluation and Loss Weighting

arXiv: https://arxiv.org/abs/1808.01694 · HTML: https://mlsanity.com/paper/1808.01694/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Aug 2018.

## Built on by (most cited first, 6 of 6)

- [Single Model Deep Learning on Imbalanced Small Datasets for Skin Lesion Classification](https://mlsanity.com/paper/2102.01284/) (2021)
- [Out-of-Distribution Detection in Dermatology using Input Perturbation and Subset Scanning](https://mlsanity.com/paper/2105.11160/) (2021)
- [Low-cost and high-performance data augmentation for deep-learning-based skin lesion classification](https://mlsanity.com/paper/2101.02353/) (2021)
- [Synthetic Generation of Dermatoscopic Images with GAN and Closed-Form Factorization](https://mlsanity.com/paper/2410.05114/) (2024)
- [An interpretable imbalanced semi-supervised deep learning framework for improving differential diagnosis of skin diseases](https://mlsanity.com/paper/2211.10858/) (2022)
- [Progressive Class-Wise Attention (PCA) Approach for Diagnosing Skin Lesions](https://mlsanity.com/paper/2306.07300/) (2023)

## Abstract (arXiv)

In this paper we present the methods of our submission to the ISIC 2018 challenge for skin lesion diagnosis (Task 3). The dataset consists of 10000 images with seven image-level classes to be distinguished by an automated algorithm. We employ an ensemble of convolutional neural networks for this task. In particular, we fine-tune pretrained state-of-the-art deep learning models such as Densenet, SENet and ResNeXt. We identify heavy class imbalance as a key problem for this challenge and consider multiple balancing approaches such as loss weighting and balanced batch sampling. Another important feature of our pipeline is the use of a vast amount of unscaled crops for evaluation. Last, we consider meta learning approaches for the final predictions. Our team placed second at the challenge while being the best approach using only publicly available data.