Deep Transfer Learning to Improve Classification Accuracy in Dermoscopic Images of Skin Cancer
Abstract
Benign and malignant cancers are the most common skin cancer types. It is essential to know skin cancer symptoms with an early diagnosis to provide an appropriate treatment and reduce the mortality rate. Dermoscopic image is one of the diagnostic media that many researchers have developed. It provides more optimal results in computational-based diagnosis than visual detection. Deep learning and transfer learning are two models that have been used successfully in computational-based analysis, although optimization is still needed. In this study, transfer learning was used to separate dermoscopic images of skin cancer into two categories: benign and malignant. This study used 2,000 images to increase previous research’s accuracy conducted on the Kaggle public dataset containing 3,297 images. Two pretrained models, namely VGG-16 and residual network (ResNet)-50, were compared and used as feature extractors. Fine-tuning was conducted by adding a flatten layer, two dense layers with the ReLU activation function, and one dense layer with the Softmax activation function to classify images into two categories. Hyperparameter tuning on the optimizer, batch size, learning rate, and epoch were performed to get each model’s best performance parameter combination. Before hyperparameter tuning, the model was tested by resizing the input image using different sizes. The results of model testing on the VGG-16 model gave the best performance at an image size of 128 × 128 pixels with a combination of Adam parameters as an optimizer, batch size of 64, learning rate of 0.001, and epoch of 10 with an accuracy value of 91% and loss of 0.2712. The ResNet-50 model yielded better accuracy of 94% and a loss of 0.2198 using the optimizer parameter RMSprop, batch size of 64, learning rate of 0.0001, and epoch of 100. The results indicate that the proposed method provides good accuracies and can assist dermatologist in the early detection of skin cancer.
References
WHO (2022) “Cancer,” [Online], https://www.who.int/news-room/fact-sheets/detail/cancer, access date: 7-Jun-2022.
M.S. Ali et al., “An Enhanced Technique of Skin Cancer Classification Using Deep Convolutional Neural Network with Transfer Learning Models,” Mach. Learn. Appl., Vol. 5, pp. 1–8, Sep. 2021, doi: 10.1016/j.mlwa.2021.100036.
M.A. Kassem, K.M. Hosny, dan M.M. Fouad, “Skin Lesions Classification into Eight Classes for ISIC 2019 Using Deep Convolutional Neural Network and Transfer Learning,” IEEE Access, Vol. 8, pp. 114822–114832, Jun. 2020, doi: 10.1109/ACCESS.2020.3003890.
M.R. Hasan et al., “Comparative Analysis of Skin Cancer (Benign vs. Malignant) Detection Using Convolutional Neural Networks,” J. Healthc. Eng., Vol. 2021, pp. 1–17, 2021, doi: 10.1155/2021/5895156.
P.P. Naik, “Cutaneous Malignant Melanoma: A Review of Early Diagnosis and Management,” World J. Oncol., Vol. 12, No. 1, pp. 7–19, Feb. 2021, doi: 10.14740/wjon1349.
S. Bechelli dan J. Delhommelle, “Machine Learning and Deep Learning Algorithms for Skin Cancer Classification from Dermoscopic Images,” Bioeng., Vol. 9, No. 3, pp. 1–18, Mar. 2022, doi: 10.3390/bioengineering9030097.
T.J. Brinker et al., “Skin Cancer Classification Using Convolutional Neural Networks: Systematic Review,” J. Med. Internet Res., Vol. 20, No. 10, pp. 1–8, Oct. 2018, doi: 10.2196/11936.
P.M.M. Pereira et al., “Skin Lesion Classification Enhancement Using Border-Line Features – The Melanoma vs Nevus Problem,” Biomed. Signal Process. Control, Vol. 57, pp. 1-8, Mar. 2020, doi: 10.1016/j.bspc.2019.101765.
G. Argenziano et al., “Dermoscopy Improves Accuracy of Primary Care Physicians to Triage Lesions Suggestive of Skin Cancer,” J. Clin. Oncol., Vol. 24, No. 12, pp. 1877–1882, Apr. 2006, doi: 10.1200/JCO.2005.05.0864.
S.M. Jaisakthi, P. Mirunalini, dan C. Aravindan, “Automated Skin Lesion Segmentation Of Dermoscopic Images Using Grabcut And K‐Means,” IET Comput. Vis., Vol. 12, No. 8, pp. 1088–1095, Dec. 2018, doi: 10.1049/iet-cvi.2018.5289.
R. Agustina, R. Magdalena, dan N.K.C. Pratiwi, “Klasifikasi Kanker Kulit Menggunakan Metode Convolutional Neural Network dengan Arsitektur VGG-16,” ELKOMIKA J. Tek. Energi Elektr. Tek. Telekomun. Tek. Elektron., Vol. 10, No. 2, pp. 446–457, Apr. 2022, doi: 10.26760/elkomika.v10i2.446.
V. Anand et al., “An Enhanced Transfer Learning Based Classification for Diagnosis of Skin Cancer,” Diagnostics, Vol. 12, No. 7, Jul. 2022, doi: 10.3390/diagnostics12071628.
Y. Wu et al., “Skin Cancer Classification With Deep Learning: A Systematic Review,” Front. Oncol., Vol. 12, pp. 1–20, Jul. 2022, doi: 10.3389/fonc.2022.893972.
M.A. Kassem, K.M. Hosny, R. Damaševičius, dan M.M. Eltoukhy, “Machine learning and Deep Learning Methods for Skin Lesion Classification and Diagnosis: A Systematic Review,” Diagnostics, Vol. 11, No. 8, pp. 1–29, Jul. 2021, doi: 10.3390/diagnostics11081390.
A. Saber et al., “A Novel Deep-Learning Model for Automatic Detection and Classification of Breast Cancer Using the Transfer-Learning Technique,” IEEE Access, Vol. 9, pp. 71194–71209, May 2021, doi: 10.1109/ACCESS.2021.3079204.
R. Refianti, A.B. Mutiara, dan R.P. Priyandini, “Classification of Melanoma Skin Cancer Using Convolutional Neural Network,” Int. J. Adv. Comput. Sci., Appl., Vol. 10, No. 3, pp. 409–417, Mar. 2019, doi: 10.14569/IJACSA.2019.0100353.
S. Saidah, I.P.Y.N. Suparta, and E. Suhartono, “Modifikasi Convolutional Neural Network Arsitektur GoogLeNet dengan Dull Razor Filtering untuk Klasifikasi Kanker Kulit,” J. Nas. Tek. Elekt., Teknol. Inf., Vol. 11, No. 2, pp. 148–153, May 2022, doi: 10.22146/jnteti.v11i2.2739.
H.A. Haenssle et al., “Man Against Machine: Diagnostic Performance of a Deep Learning Convolutional Neural Network for Dermoscopic Melanoma Recognition in Comparison to 58 Dermatologists,” Ann. Oncol., Vol. 29, No. 8, pp. 1836–1842, Aug. 2018, doi: 10.1093/annonc/mdy166.
T.J. Brinker et al., “A Convolutional Neural Network Trained with Dermoscopic Images Performed on Par With 145 Dermatologists in a Clinical Melanoma Image Classification Task,” Eur. J. Cancer, Vol. 111, pp. 148–154, Apr. 2019, doi: 10.1016/j.ejca.2019.02.005.
S. Guan and M. Loew, “Breast Cancer Detection Using Transfer Learning In Convolutional Neural Networks,” 2017 Appl. Imag. Pattern Recognit. Workshop (AIPR), 2017, pp. 1–8, doi: 10.1109/AIPR.2017.8457948.
M. Hussain, J.J. Bird, dan D.R. Faria, “A Study on CNN Transfer Learning for Image Classification,” in Advances in Computational Intelligence Systems, A. Lotfi et al., Eds., Cham, Switzerland: Springer Cham, 2019, pp. 191–202, doi: 10.1007/978-3-319-97982-3_16.
S. Tammina, “Transfer Learning Using VGG-16 with Deep Convolutional Neural Network for Classifying Images,” Int. J. Sci., Res. Publ., Vol. 9, No. 10, pp. 143–150, Oct. 2019, doi: 10.29322/ijsrp.9.10.2019.p9420.
Shallu dan R. Mehra, “Breast Cancer Histology Images Classification: Training from Scratch or Transfer Learning?” ICT Express, Vol. 4, No. 4, pp. 247–254, Dec. 2018, doi: 10.1016/j.icte.2018.10.007.
D.M. Wonohadidjojo, “Perbandingan Convolutional Neural Network pada Transfer Learning Method untuk Mengklasifikasikan Sel Darah Putih,” Ultimics J. Tek. Inform., Vol. 13, No. 1, pp. 51–57, Jun. 2021, doi: 10.31937/ti.v13i1.2040.
Z. Aliyah, A. Arifianto, and F. Sthevanie, “Classifying Skin Cancer in Digital Images Using Convolutional Neural Network with Augmentation,” Indonesia J. Comput. (Indo-JC), Vol. 5, No. 2, pp. 55–66, Sep. 2020, doi: 10.21108/indojc.2020.5.2.455.
A. Abubakar, M. Ajuji, and I.U. Yahya, “Comparison of Deep Transfer Learning Techniques in Human Skin Burns Discrimination,” Appl. Syst. Innov., Vol. 3, No. 2, pp. 1–15, Apr. 2020, doi: 10.3390/asi3020020.
A.V. Ikechukwu, S. Murali, R. Deepu, and R. C. Shivamurthy, “ResNet-50 vs VGG-19 vs Training from Scratch: A Comparative Analysis of the Segmentation and Classification of Pneumonia from Chest X-Ray Images,” Glob. Transit. Proc., Vol. 2, No. 2, pp. 375–381, Nov. 2021, doi: 10.1016/j.gltp.2021.08.027.
R.A. Khan, A. Crenn, A. Meyer, and S. Bouakaz, “A Novel Database of Children’s Spontaneous Facial Expressions (LIRIS-CSE),” Image, Vis. Comput., Vol. 83–84, pp. 61–69, Mar.–Apr. 2019, doi: 10.1016/j.imavis.2019.02.004.
C. Nwankpa, W. Ijomah, A. Gachagan, and S. Marshall, “Activation Functions: Comparison of Trends in Practice and Research for Deep Learning,” 2018, arXiv:1811.03378.
X. Qin et al., “A Faster R-CNN Based Method for Comic Characters Face Detection,” 2017 14th IAPR Int. Conf. Doc. Anal., Recogn. (ICDAR), 2017, pp. 1074–1080, doi: 10.1109/ICDAR.2017.178.
M. Korpusik and J. Glass, “Convolutional Neural Networks and Multitask Strategies for Semantic Mapping of Natural Language Input to a Structured Database,” 2018 IEEE Int. Conf. Acoust. Speech, Signal Process. (ICASSP), 2018, pp. 6174–6178, doi: 10.1109/ICASSP.2018.8461769.
E. Solovyeva and A. Abdullah, “Binary and Multiclass Text Classification by Means of Separable Convolutional Neural Network,” Inventions, Vol. 6, No. 4, Oct. 2021, doi: 10.3390/inventions6040070.
K. Manasa and D.G.V. Murthy, “Skin Cancer Detection Using VGG-16,” Eur. J. Mol., Clin. Med., Vol. 8, No. 1, pp. 1419–1426, Jan. 2021.
M. Faruk and N. Nafi’iyah, “Klasifikasi Kanker Kulit Berdasarkan Fitur Tekstur, Fitur Warna Citra Menggunakan SVM dan KNN,” Telematika, Vol. 13, No. 2, pp. 100–109, Aug. 2020, doi: 10.35671/telematika.v13i2.987.
N. Khasanah et al., “Skin Cancer Classification Using Random Forest Algorithm,” J. Ilm. Sisfotenika, Vol. 11, No. 2, pp. 137–147, Jul. 2021, doi: 10.30700/jst.v11i2.1122.
© Jurnal Nasional Teknik Elektro dan Teknologi Informasi, under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License.
1.png)
