SkinSAM: Empowering Skin Cancer Segmentation with Segment Anything Model

Skin cancer is a prevalent and potentially fatal disease that requires accurate and efficient diagnosis and treatment. Although manual tracing is the current standard in clinics, automated tools are desired to reduce human labor and improve accuracy. However, developing such tools is challenging due to the highly variable appearance of skin cancers and complex objects in the background. In this paper, we present SkinSAM, a fine-tuned model based on the Segment Anything Model that showed outstanding segmentation performance. The models are validated on HAM10000 dataset which includes 10015 dermatoscopic images. While larger models (ViT_L, ViT_H) performed better than the smaller one (ViT_b), the finetuned model (ViT_b_finetuned) exhibited the greatest improvement, with a Mean pixel accuracy of 0.945, Mean dice score of 0.8879, and Mean IoU score of 0.7843. Among the lesion types, vascular lesions showed the best segmentation results. Our research demonstrates the great potential of adapting SAM to medical image segmentation tasks.

[1]  Ross B. Girshick,et al.  Segment Anything , 2023, 2023 IEEE/CVF International Conference on Computer Vision (ICCV).

[2]  J. Roper,et al.  Abdomen CT Multi-organ Segmentation Using Token-based MLP-Mixer. , 2022, Medical physics.

[3]  A. Jemal,et al.  Cancer statistics, 2022 , 2022, CA: a cancer journal for clinicians.

[4]  B. Landman,et al.  Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysis , 2021, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Ross B. Girshick,et al.  Masked Autoencoders Are Scalable Vision Learners , 2021, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Muhammad Attique Khan,et al.  A two‐stream deep neural network‐based intelligent system for complex skin cancer types classification , 2021, Int. J. Intell. Syst..

[7]  Qi Tian,et al.  Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation , 2021, ECCV Workshops.

[8]  Deepika Koundal,et al.  Skin Disease Diagnosis: Challenges and Opportunities , 2021, Proceedings of Second Doctoral Symposium on Computational Intelligence.

[9]  A. Dosovitskiy,et al.  MLP-Mixer: An all-MLP Architecture for Vision , 2021, NeurIPS.

[10]  S. Gelly,et al.  An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale , 2020, ICLR.

[11]  John Paoli,et al.  Human–computer collaboration for skin cancer recognition , 2020, Nature Medicine.

[12]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

[13]  Chunyan Miao,et al.  A Survey of Zero-Shot Learning , 2019, ACM Trans. Intell. Syst. Technol..

[14]  Klaus H. Maier-Hein,et al.  nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation , 2018, Bildverarbeitung für die Medizin.

[15]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[16]  Johannes E. Schindelin,et al.  Fiji: an open-source platform for biological-image analysis , 2012, Nature Methods.

[17]  Michael R Hamblin,et al.  CA : A Cancer Journal for Clinicians , 2011 .