Skin cancer classification integrating improved ResNet50 with ensemble classifier

doi:10.11772/j.issn.1001-9081.2025040513

Abstract

Abstract:

As a malignant tumor with continuously rising incidence rates globally， skin cancer requires early and precise diagnosis to reduce the mortality rate. To address the challenges of insufficient model performance for clinical requirements and low diagnostic accuracy in minority categories of skin cancers， a model of an improved ResNet50 integrated with an ensemble classifier was proposed. Firstly， hair-induced noise was eliminated through grayscale black-hat threshold processing and Telea algorithm， and Synthetic Minority Over-sampling TEchnique （SMOTE） was used to balance class distribution. Secondly， the deep-level features were extracted using the ResNet50 model， with a soft attention module combining spatial and channel attention mechanisms introduced to focus on skin lesion regions. Finally， an ensemble classifier integrating random forest， eXtreme Gradient Boosting （XGBoost）， Light Gradient Boosting Machine （LightGBM）， K-Nearest Neighbor （K-NN）， and Support Vector Machine （SVM） via soft voting was employed， and the proposed model was applied for early diagnosis of skin cancers. The results of three separate experiments on the HAM10000， ISIC2019 and ISIC2020 datasets indicate that the proposed model improves the accuracy to （98.33±0.03）%，（96.15±0.06）% and （99.19±0.02）%， respectively. Compared with current mainstream networks， the proposed model exhibits superior feature extraction and classification capabilities， and is helpful to improve early diagnostic effects.

Key words: skin cancer classification, ResNet50, Synthetic Minority Over-sampling TEchnique (SMOTE), ensemble learning, attention mechanism

摘要：

皮肤癌是全球发病率持续攀升的恶性肿瘤之一，它的早期精准诊断对降低死亡率至关重要。针对现有模型难以满足临床要求及少数类别识别皮肤癌精度低的问题，提出一种融合改进ResNet50与集成分类器的模型。首先，通过灰度黑帽阈值处理和Telea算法去除毛发噪声，再使用合成少数过采样技术（SMOTE）平衡类别分布；其次，采用ResNet50模型提取深层特征，并引入融合空间与通道注意力的软注意力模块聚焦皮肤病变区域；最后，将随机森林、极限梯度提升（XGBoost）、轻量级梯度提升机（LightGBM）、K-近邻（K-NN）和支持向量机（SVM）作为集成分类器，通过软投票法进行集成，进行皮肤癌的早期诊断。在HAM10000、ISIC2019和ISIC2020数据集上的3次独立实验结果表明，所提模型将准确率分别提升到（98.33±0.03）%、（96.15±0.06）%和（99.19±0.02）%，相较于当前主流网络具有更优的特征提取与分类能力，有助于提升早期诊断效果。

关键词: 皮肤癌分类, ResNet50, 合成少数过采样技术, 集成学习, 注意力机制

CLC Number:

TP391.4

Chuandong QIN, Zhiqiang SUO. Skin cancer classification integrating improved ResNet50 with ensemble classifier[J]. Journal of Computer Applications, 2026, 46(4): 1354-1362.

秦传东, 索志强. 融合改进的ResNet50与集成分类器的皮肤癌分类[J]. 《计算机应用》唯一官方网站, 2026, 46(4): 1354-1362.

Figures/Tables 26

References 29

[1]	SHAIKH M A， DUAN T， CHAUHAN M， et al. Attention based writer independent verification［C］// Proceedings of the 17th International Conference on Frontiers in Handwriting Recognition. Piscataway： IEEE， 2020： 373-379.
[2]	PADHY S， DASH S， KUMAR N， et al. Temporal integration of ResNet features with LSTM for enhanced skin lesion classification［J］. Results in Engineering， 2025， 25： No.104201.
[3]	NAWAZ K， ZANIB A， SHABIR I， et al. Skin cancer detection using dermoscopic images with convolutional neural network［J］. Scientific Reports， 2025， 15： No.7252.
[4]	ROTEMBERG V， KURTANSKY N， BETZ-STABLEIN B， et al. A patient-centric dataset of images and metadata for identifying melanomas using clinical context［J］. Scientific Data， 2021， 8： No.34.
[5]	TSCHANDL P， ROSENDAHL C， KITTLER H. The HAM10000 dataset， a large collection of multi-source dermatoscopic images of common pigmented skin lesions［J］. Scientific Data， 2018， 5： No.180161.
[6]	SABIR R， MEHMOOD T. Classification of melanoma skin cancer based on Image Data Set using different neural networks［J］. Scientific Reports， 2024， 14： No.29704.
[7]	KHAN M A， ZHANG Y D， SHARIF M， et al. Pixels to classes： intelligent learning framework for multiclass skin lesion localization and classification［J］. Computers and Electrical Engineering， 2021， 90： No.106956.
[8]	KASSEM M A， HOSNY K M， FOUAD M M. Skin lesions classification into eight classes for ISIC 2019 using deep convolutional neural network and transfer learning［J］. IEEE Access， 2020， 8： 114822-114832.
[9]	CODELLA N C F， GUTMAN D， CELEBI M E， 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）［C］// Proceedings of the IEEE 15th International Symposium on Biomedical Imaging. Piscataway： IEEE， 2018： 168-172.
[10]	HERNÁNDEZ-PÉREZ C， COMBALIA M， PODLIPNIK S， et al. BCN20000： dermoscopic lesions in the wild［J］. Scientific Data， 2024， 11： No.641.
[11]	何雪英，韩忠义，魏本征. 基于深度卷积神经网络的色素性皮肤病识别分类［J］. 计算机应用， 2018， 38（11）： 3236-3240.
	HE X Y， HAN Z Y， WEI B Z. Pigmented skin lesion recognition and classification based on deep convolutional neural network［J］. Journal of Computer Applications， 2018， 38（11）： 3236-3240.
[12]	ALHUDHAIF A， ALMASLUKH B， ASEERI A O， et al. A novel nonlinear automated multi-class skin lesion detection system using soft-attention based convolutional neural networks［J］. Chaos， Solitons and Fractals， 2023， 170： No.113409.
[13]	THWIN S M， PARK H S， SEO S H. A trustworthy framework for skin cancer detection using a CNN with a modified attention mechanism［J］. Applied Sciences， 2025， 15（3）： No.1067.
[14]	OZDEMIR B， PACAL I. An innovative deep learning framework for skin cancer detection employing ConvNeXtV2 and focal self-attention mechanisms［J］. Results in Engineering， 2025， 25： No.103692.
[15]	CHENG H， LIAN J， JIAO W. Enhanced MobileNet for skin cancer image classification with fused spatial channel attention mechanism［J］. Scientific Reports， 2024， 14： No.28850.
[16]	OZDEMIR B， PACAL I. A robust deep learning framework for multiclass skin cancer classification［J］. Scientific Reports， 2025， 15： No.4938.
[17]	HALDER A， DALAL A， GHARAMI S， et al. A fuzzy rank-based deep ensemble methodology for multi-class skin cancer classification［J］. Scientific Reports， 2025， 15： No.6268.
[18]	NATHA P， TERA S P， CHINTHAGINJALA R， et al. Boosting skin cancer diagnosis accuracy with ensemble approach［J］. Scientific Reports， 2025， 15： No.1290.
[19]	DORATHI JAYASEELI J D， BRISKILAL J， FANCY C， et al. An intelligent framework for skin cancer detection and classification using fusion of Squeeze-Excitation-DenseNet with Metaheuristic-driven ensemble deep learning models［J］. Scientific Reports， 2025， 15： No.7425.
[20]	DAS A， MOHANTY M N. Design of stacked ensemble classifier for skin cancer detection［J］. Multimedia Tools and Applications， 2025， 84（28）： 34705-34724.
[21]	AKILANDASOWMYA G， NIRMALADEVI G， SUGANTHI S， et al. Skin cancer diagnosis： leveraging deep hidden features and ensemble classifiers for early detection and classification［J］. Biomedical Signal Processing and Control， 2024， 88（Pt C）： No.105306.
[22]	TAN M， LE Q V. EfficientNet： rethinking model scaling for convolutional neural networks［C］// Proceedings of the 36th International Conference on Machine Learning. New York： JMLR.org， 2019： 6105-6114.
[23]	SANDLER M， HOWARD A， ZHU M， et al. MobileNetV2： inverted residuals and linear bottlenecks［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 4510-4520.
[24]	HE K， ZHANG X， REN S， et al. Deep residual learning for image recognition［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 770-778.
[25]	HUANG G， LIU Z， VAN DER MAATEN L， et al. Densely connected convolutional networks［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2017： 2261-2269.
[26]	SZEGEDY C， LIU W， JIA Y， et al. Going deeper with convolutions［C］// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2015： 1-9.
[27]	DONG H， KOTENKO I， DONG S. A lightweight vision Transformer with weighted global average pooling： implications for IoMT applications［J］. Complex and Intelligent Systems， 2025， 11（5）： No.215.
[28]	KAUR R， GHOLAMHOSSEINI H， LINDÉN M. Advanced deep learning models for melanoma diagnosis in computer-aided skin cancer detection［J］. Sensors， 2025， 25（3）： No.594.
[29]	WANG F， JU M， ZHU X， et al. A geometric algebra-enhanced network for skin lesion detection with diagnostic prior［J］. The Journal of Supercomputing， 2025， 81： No.327.

数据集	AKIEC	BCC	BKL	DF	MEL	NV	VASC	SCC
HAM10000	327	514	1 099	115	1 113	6 705	142
ISIC2019	867	3 323	2 624	239	4 522	12 875	253	628

数据集	AKIEC	BCC	BKL	DF	MEL	NV	VASC	SCC
HAM10000	327	514	1 099	115	1 113	6 705	142
ISIC2019	867	3 323	2 624	239	4 522	12 875	253	628

项目	配置	项目	配置
操作系统	Windows11 64 位	Python	3.11.0
处理器	i9-14900	PyTorch	2.5.1
显卡	NVIDIA RTX4060	Torchvision	0.20.1

项目	配置	项目	配置
操作系统	Windows11 64 位	Python	3.11.0
处理器	i9-14900	PyTorch	2.5.1
显卡	NVIDIA RTX4060	Torchvision	0.20.1

类别	精确率	召回率	F1分数	Support
AKIEC	1.00	1.00	1.00	2 048
BCC	0.99	1.00	0.99	1 993
BKL	0.97	0.98	0.97	1 954
DF	1.00	1.00	1.00	2 071
MEL	0.96	0.96	0.96	2 017
NV	0.97	0.95	0.96	2 040
VASC	1.00	1.00	1.00	1 958