The classification of rare skin diseases faces significant data scarcity challenges due to the difficulty in acquiring clinical samples and the high cost of annotation, which severely hinders the training of deep neural network–based models. Few-shot learning has emerged as a cutting-edge solution, with its core capability being the identification of novel disease classes using limited annotated samples to mitigate data insufficiency. However, most existing methods fail to fully leverage the statistical information from base classes to calibrate the distribution of few-shot classes, thereby optimizing classifier inputs. Two critical research challenges remain: (1) accurately estimating the true distribution of few-shot classes with minimal samples, and (2) selecting appropriate base class information for effective distribution calibration. To address these challenges, we propose SADC (skin disease classification via adaptive distribution calibration), a new few-shot learning framework incorporating multi-scale feature extraction and adaptive sample calibration. First, our multi-scale feature extraction strategy employs feature descriptor matrices and composite metrics to optimize multi-dimensional, multi-directional feature representations, enabling precise similarity computation between base-class and few-shot samples. Second, the adaptive sample calibration strategy constructs weight matrices based on sample similarity to automatically select optimal base-class samples with adaptive weights for distribution calibration, ensuring alignment between calibrated distributions and true unbiased distributions. Experimental results demonstrated that SADC achieves state-of-the-art performance across three public dermatology datasets (ISIC2018, Derm7pt, and SD198), showing significant improvements over existing methods. The framework's innovation lies in its dual-strategy approach to distribution-aware few-shot learning, advancing the frontier of data-efficient medical image analysis.
Citation: Yin Wen, Yingbo Wu, Zhigao Zeng, Shengqiu Yi, Xinpan Yuan, Yanhui Zhu. Few-shot learning for rare skin disease classification via adaptive distribution calibration[J]. Mathematical Biosciences and Engineering, 2025, 22(12): 3005-3027. doi: 10.3934/mbe.2025111
The classification of rare skin diseases faces significant data scarcity challenges due to the difficulty in acquiring clinical samples and the high cost of annotation, which severely hinders the training of deep neural network–based models. Few-shot learning has emerged as a cutting-edge solution, with its core capability being the identification of novel disease classes using limited annotated samples to mitigate data insufficiency. However, most existing methods fail to fully leverage the statistical information from base classes to calibrate the distribution of few-shot classes, thereby optimizing classifier inputs. Two critical research challenges remain: (1) accurately estimating the true distribution of few-shot classes with minimal samples, and (2) selecting appropriate base class information for effective distribution calibration. To address these challenges, we propose SADC (skin disease classification via adaptive distribution calibration), a new few-shot learning framework incorporating multi-scale feature extraction and adaptive sample calibration. First, our multi-scale feature extraction strategy employs feature descriptor matrices and composite metrics to optimize multi-dimensional, multi-directional feature representations, enabling precise similarity computation between base-class and few-shot samples. Second, the adaptive sample calibration strategy constructs weight matrices based on sample similarity to automatically select optimal base-class samples with adaptive weights for distribution calibration, ensuring alignment between calibrated distributions and true unbiased distributions. Experimental results demonstrated that SADC achieves state-of-the-art performance across three public dermatology datasets (ISIC2018, Derm7pt, and SD198), showing significant improvements over existing methods. The framework's innovation lies in its dual-strategy approach to distribution-aware few-shot learning, advancing the frontier of data-efficient medical image analysis.
| [1] |
F. Grignaffini, F. Barbuto, L. Piazzo, M. Troiano, P. Simeoni, F. Mangini, et al., Machine learning approaches for skin cancer classification from dermoscopic images: A systematic review, Algorithms, 15 (2022), 438. https://doi.org/10.3390/a15110438 doi: 10.3390/a15110438
|
| [2] |
K. M. Hosny, W. Said, M. Elmezain, M. A. Kassem, Explainable deep inherent learning for multi-classes skin lesion classification, Appl. Soft Comput. , 159 (2024), 111624. https://doi.org/10.1016/j.asoc.2024.111624 doi: 10.1016/j.asoc.2024.111624
|
| [3] |
D. Meedeniya, S. D. Silva, L. Gamage, U. Isuranga, Skin cancer identification utilizing deep learning: A survey, IET Image Process. 18 (2024), 3731-3749. https://doi.org/10.1049/ipr2.13219 doi: 10.1049/ipr2.13219
|
| [4] |
H. Naseri, A. A. Safaei, Diagnosis and prognosis of melanoma from dermoscopy images using machine learning and deep learning: a systematic literature review, BMC Cancer, 25 (2025), 75. https://doi.org/10.1186/s12885-024-13423-y doi: 10.1186/s12885-024-13423-y
|
| [5] |
Y. Song, T. Wang, P. Cai, S. K. Mondal, J. P. Sahoo, A comprehensive survey of few-shot learning: Evolution, applications, challenges, and opportunities, ACM Comput. Surveys, 55 (2023), 1-40. https://doi.org/10.1145/3582688 doi: 10.1145/3582688
|
| [6] |
W. Zeng, Z. Y. Xiao, Few-shot learning based on deep learning: A survey, Math. Biosci. Eng., 21 (2024), 679-711. https://doi.org/10.3934/mbe.2024029 doi: 10.3934/mbe.2024029
|
| [7] | A. Parnami, M. Lee, Learning from few examples: A summary of approaches to few-shot learning. Preprint, (2022), arXiv:2203.04291. |
| [8] | C. Finn, P. Abbeel, S. Levine, Model-agnostic meta-learning for fast adaptation of deep networks, In: Proceedings of the 34th International Conference on Machine Learning, (2017), 1126-1135. |
| [9] | J. Snell, K. Swersky, R. Zemel, Prototypical networks for few-shot learning, Adv. Neural Inform. Process. Syst., (2017), 30. http://arXiv.org/abs/1703.05175 |
| [10] | K. Mahajan, M. Sharma, L. Vig, Meta-dermdiagnosis: Few-shot skin disease identification using meta-learning, In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, (2020), 730-731. https://doi.org/10.1109/CVPRW50498.2020.00373 |
| [11] | X. Li, L. Yu, Y. Jin, C. W. Fu, L. Xing, P. A. Heng, Difficulty-aware meta-learning for rare disease diagnosis, Med. Image Comput. Computer Assisted Intervent., (2020), 357-366. https://doi.org/10.1007/978-3-030-59710-8_35 |
| [12] |
R. Singh, V. Bharti, V. Purohit, A. Kumar, A. K. Singh, S. K. Singh, MetaMed: Few-shot medical image classification using gradient-based meta-learning, Pattern Recogn. , 120 (2021), 108111. https://doi.org/10.1016/j.patcog.2021.108111 doi: 10.1016/j.patcog.2021.108111
|
| [13] | W. Zhu, H. Liao, W. Li, W. Li, J. Luo, Alleviating the incompatibility between cross entropy loss and episode training for few-shot skin disease classification, Med. Image Comput. Computer Assisted Intervent., (2020), 330-339. |
| [14] | R. R. Chowdhury, D. R. Bathula, Influential prototypical networks for few shot learning: A dermatological case study, In: 2022 IEEE 19th International Symposium on Biomedical Imaging, (2022), 1-4. |
| [15] | C. Zhou, M. Sun, L. Chen, A. Cai, J. Fang, Few-shot learning framework based on adaptive subspace for skin disease classification, In: 2022 IEEE International Conference on Bioinformatics and Biomedicine, (2022), 2231-2237. https://doi.org/10.1109/BIBM55620.2022.9995042 |
| [16] |
J. Xiao, H. Xu, D. Fang, C. Cheng, H. Gao, Boosting and rectifying few-shot learning prototype network for skin lesion classification based on the internet of medical things, Wireless Networks, 29 (2023), 1507-1521. https://doi.org/10.1007/s11276-021-02713-z doi: 10.1007/s11276-021-02713-z
|
| [17] |
Z. Dai, J. Yi, L. Yan, Q. Xu, L. Hu, Q. Zhang, et al., PFEMed: Few-shot medical image classification using prior guided feature enhancement, Pattern Recogn. , 134 (2023), 109108. https://doi.org/10.1016/j.patcog.2022.109108 doi: 10.1016/j.patcog.2022.109108
|
| [18] | D. Chen, Y. Chen, Y. Li, F. Mao, Y. He, H. Xue, Self-supervised learning for few-shot image classification, In: ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing, (2021), 1745-1749. https://doi.org/10.1109/ICASSP39728.2021.9413783 |
| [19] | C. Medina, A. Devos, M. Grossglauser, Self-supervised prototypical transfer learning for few-shot classification, preprint, arXiv:2006.11325. |
| [20] | C. P. Phoo, B. Hariharan, Self-training for few-shot transfer across extreme task differences, preprint, (2020), arXiv:2010.07734. |
| [21] | V. Prabhu, A. Kannan, M. Ravuri, M. Chaplain, D. Sontag, X. Amatriain, Few-shot learning for dermatological disease diagnosis, In: Machine Learning for Healthcare Conference, (2019), 532-552. https://doi.org/10.1016/B978-0-32-399851-2.00022-3 |
| [22] | S. Li, X. Li, X. Xu, K. T. Cheng, Dynamic subcluster-aware network for few-shot skin disease classification, IEEE Transact. Neural Networks Learn. Syst., (2023). https://doi.org/10.1109/TNNLS.2023.3336765 |
| [23] | S. Yang, L. Liu, M. Xu, Free lunch for few-shot learning: Distribution calibration, preprint, (2021), arXiv:2101.06395. |
| [24] | Y. Xian, T. Lorenz, B. Schiele, Z. Akata, Feature generating networks for zero-shot learning, In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2018), 5542-5551. https://doi.org/10.1109/CVPR.2018.00581 |
| [25] | A. Nichol, J. Schulman, Reptile: A scalable metalearning algorithm, preprint, (2018), arXiv:1803.02999. |
| [26] | O. Vinyals, C. Blundell, T. Lillicrap, D. Wierstra, Matching networks for one shot learning, Adv. Neural Inform. Process. Syst., 29 (2016). http://arXiv.org/abs/1606.04080 |
| [27] | F. Sung, Y. Yang, L. Zhang, T. Xiang, P. H. Torr, T. M. Hospedales, Learning to compare: Relation network for few-shot learning, In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2018), 1199-1208. https://doi.org/10.1109/CVPR.2018.00131 |
| [28] | W. Y. Chen, Y. C. Liu, Z. Kira, Y. C. F. Wang, J. B. Huang, A closer look at few-shot classification, preprint, (2019), arXiv:1904.04232. |
| [29] | P. Mangla, N. Kumari, A. Sinha, M. Singh, B. Krishnamurthy, V. N. Balasubramanian, Charting the right manifold: Manifold mixup for few-shot learning, In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, (2020), 2218-2227. https://doi.org/10.1109/WACV45572.2020.9093338 |
| [30] |
W. Fu, J. Chen, L. Zhou, Boosting few-shot rare skin disease classification via self-supervision and distribution calibration, Biomed. Eng. Letters, 14 (2024), 877-889. https://doi.org/10.1007/s13534-024-00383-2 doi: 10.1007/s13534-024-00383-2
|
| [31] | J. Gui, T. Chen, J. Zhang, Q. Cao, Z. Sun, H. Luo, D. Tao, A survey on self-supervised learning: Algorithms, applications, and future trends, IEEE Transact. Pattern Anal. Mach. Intell., (2024). https://doi.org/10.1109/TPAMI.2024.3415112 |
| [32] | E. Jang, C. Devin, V. Vanhoucke, S. Levine, Grasp2vec: Learning object representations from self-supervised grasping, preprint, (2018), arXiv:1811.06964. |
| [33] | S. Gidaris, P. Singh, N. Komodakis, Unsupervised representation learning by predicting image rotations, preprint, (2018), arXiv:1803.07728. |
| [34] | J. W. Tukey, Exploratory Data Analysis, Reading, MA: Addison-Wesley, 2 (1977), 131-160. |
| [35] | X. Wei, W. Du, H. Wan, W. Min, Feature distribution fitting with direction-driven weighting for few-shot images classification, In: Proceedings of the AAAI Conference on Artificial Intelligence, 37 (2023), 10315-10323. https://doi.org/10.1609/aaai.v37i9.26228 |
| [36] | L. Chen, Y. Gu, Y. Guo, F. Dong, D. Jiang, Y. Chen, DDC: Dynamic distribution calibration for few-shot learning under multi-scale representation, Knowledge-Based Syst., (2025), 113030. https://doi.org/10.1016/j.knosys.2025.113030 |
| [37] | M. Arjovsky, S. Chintala, L. Bottou, Wasserstein generative adversarial networks, In: International Conference on Machine Learning, (2017), 214-223. |
| [38] |
A. Brauer, J. E. Shockley, On a problem of Frobenius, J. Reine Angew. Math. , 211 (1962), 215-220. https://doi.org/10.1515/crll.1962.211.215 doi: 10.1515/crll.1962.211.215
|
| [39] |
M. L. Menéndez, J. A. Pardo, L. Pardo, M. D. C. Pardo, The Jensen-Shannon divergence, J. Franklin Institute, 334 (1997), 307-318. https://doi.org/10.1016/S0016-0032(96)00063-4 doi: 10.1016/S0016-0032(96)00063-4
|
| [40] |
X. Liu, K. Zhou, P. Yang, L. Jing, J. Yu, Adaptive distribution calibration for few-shot learning via optimal transport, Inform. Sci. , 611 (2022), 1-17. https://doi.org/10.1016/j.ins.2022.07.189 doi: 10.1016/j.ins.2022.07.189
|
| [41] | N. C. F. Codella, D. Gutman, M. E. Celebi, B. Helba, M. A. Marchetti, S. W. Dusza, et al., Skin lesion analysis toward melanoma detection: A challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), hosted by the International Skin Imaging Collaboration (ISIC), In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), (2018), 168-172. |
| [42] |
J. Kawahara, S. Daneshvar, G. Argenziano, G. Hamarneh, Seven-point checklist and skin lesion classification using multitask multimodal neural nets, IEEE J. Biomed. Health Inform. , 23 (2018), 538-546. https://doi.org/10.1109/JBHI.2018.2824327 doi: 10.1109/JBHI.2018.2824327
|
| [43] | X. Sun, J. Yang, M. Sun, K. Wang, A benchmark for automatic visual classification of clinical skin disease images, In: Computer Vision-ECCV 2016: 14th European Conference, Proceedings, Part VI 14, (2016), 206-222. https://doi.org/10.1007/978-3-319-46466-4_13 |
| [44] | S. Zagoruyko, N. Komodakis, Wide residual networks, preprint, (2016), arXiv:1605.07146. |
| [45] | D. P. Kingma, J. Ba, Adam: A method for stochastic optimization, preprint, (2014), arXiv:1412.6980. |
| [46] | L. Van der Maaten, G. Hinton, Visualizing data using t-SNE, J. Mach. Learn. Res., 9 (2008). |
| [47] | L. Gamage, U. Isuranga, S. De Silva, D. Meedeniya, Melanoma skin cancer classification with explainability, In: Proceedings of the 2023 3rd International Conference on Advanced Research in Computing (ICARC), (2023), 30-35. https://doi.org/10.1109/ICARC57651.2023.10145622 |
| [48] |
L. Gamage, U. Isuranga, D. Meedeniya, S. De Silva, Melanoma skin cancer identification with explainability utilizing mask guided technique, Electronics, 13 (2024), 680. https://doi.org/10.3390/electronics13040680 doi: 10.3390/electronics13040680
|
Figures(7) / Tables(7)
Yin Wen, Yingbo Wu, Zhigao Zeng, Shengqiu Yi, Xinpan Yuan, Yanhui Zhu. Few-shot learning for rare skin disease classification via adaptive distribution calibration[J]. Mathematical Biosciences and Engineering, 2025, 22(12): 3005-3027. doi: 10.3934/mbe.2025111
DownLoad: