SpectraScanNet: Enhancing Early Skin Cancer Detection through Spectral Imaging and Deep Learning
Article Sidebar
Main Article Content
Abstract
This study introduces SpectraScanNet, a cutting-edge framework that combines spectral imaging and deep learning to enhance early skin cancer detection. The objective was to overcome the limitations of traditional diagnostic methods, which rely heavily on the subjective expertise of clinicians and standard imaging techniques, resulting in inconsistent accuracy and potential misdiagnosis. SpectraScanNet employs multispectral and hyperspectral imaging to capture the comprehensive spectral data of skin lesions. These data were processed using a customized deep learning model that included normalization, noise reduction, and spectral calibration to ensure accuracy. This methodology focuses on extracting and analyzing spectral features using convolutional layers and spectral attention mechanisms, facilitating precise differentiation between benign and malignant lesions. The performance of SpectraScanNet was evaluated using the HAM10000 dataset, which achieved a diagnostic accuracy of 92.5%, sensitivity of 94.8%, and specificity of 90.3%. These results demonstrated a significant improvement over traditional methods, with an average accuracy of 75.0% and an accuracy of 85.0 %. Robustness analysis across various spectral bands confirmed a consistent performance, particularly in the 600-700 nm range, highlighting the model’s effective use of spectral data for enhanced diagnosis. However, the study acknowledges limitations such as reliance on high-quality spectral data and integration challenges in clinical workflows. Future research will aim to adapt SpectraScanNet to lower-quality images, expand dataset diversity, and streamline the clinical integration process, ensuring broader applicability and improved patient outcomes.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
IJCERT Policy:
The published work presented in this paper is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. This means that the content of this paper can be shared, copied, and redistributed in any medium or format, as long as the original author is properly attributed. Additionally, any derivative works based on this paper must also be licensed under the same terms. This licensing agreement allows for broad dissemination and use of the work while maintaining the author's rights and recognition.
By submitting this paper to IJCERT, the author(s) agree to these licensing terms and confirm that the work is original and does not infringe on any third-party copyright or intellectual property rights.
References
Siegel, R. L., Miller, K. D., & Jemal, A. (2020). Cancer statistics, 2020. CA: A Cancer Journal for Clinicians, 70(1), 7-30.
American Cancer Society. (2020). Skin Cancer Statistics. American Cancer Society.
Wong, C. S. M., Strange, R. C., & Lear, J. T. (2003). Basal cell carcinoma. BMJ, 327(7418), 794-798.
Carli, P., De Giorgi, V., Chiarugi, A., Stante, M., & Giannotti, B. (2004). Automated diagnosis of pigmented skin lesions. Clinical Dermatology, 22(2), 175-182.
Argenziano, G., Soyer, H. P., De Giorgi, V., Piccolo, D., & Carli, P. (2006). Dermoscopy: A tutorial. EDRA.
Kittler, H., Pehamberger, H., Wolff, K., & Binder, M. (2002). Diagnostic accuracy of dermoscopy. The Lancet Oncology, 3(3), 159-165.
Johr, R. H. (2002). Dermoscopy: Alternative melanocytic algorithms—the ABCD rule of dermatoscopy, Menzies scoring method, and 7-point checklist. Clinics in Dermatology, 20(3), 240-247.
Nachbar, F., Stolz, W., Merkle, T., et al. (1994). The ABCD rule of dermatoscopy. Journal of the American Academy of Dermatology, 30(4), 551-559.
Vestergaard, M. E., Macaskill, P., Holt, P. E., & Menzies, S. W. (2008). Dermoscopy compared with naked eye examination for the diagnosis of primary melanoma: a meta-analysis of studies performed in a clinical setting. British Journal of Dermatology, 159(3), 669-676.
Zalaudek, I., Argenziano, G., Soyer, H. P., et al. (2004). Three-point checklist of dermoscopy. Dermatoscopy: an algorithmic method based on pattern analysis. Dermatologic Clinics, 19(3), 313-321.
Rosendahl, C., Cameron, A., McColl, I., & Wilkinson, D. (2011). Dermatoscopy in routine practice: 'Chaos and Clues'. Australasian Journal of Dermatology, 52(2), 109-115.
Kittler, H., Seltenheim, M., Dawid, M., et al. (1998). Diagnostic accuracy of epiluminescence microscopy and the impact of experience on early diagnosis of melanoma. Archives of Dermatology, 134(12), 1550-1555.
Esteva, A., Kuprel, B., Novoa, R. A., et al. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115-118.
Phillips, M., Marsden, H., Jaffe, W., & Matin, R. N. (2019). Reporting the accuracy of machine learning models in dermatology: A framework for confidence. JAMA Dermatology, 155(10), 1249-1251.
Tschandl, P., Rosendahl, C., & Kittler, H. (2020). The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data, 5(1), 180161.
Thissen, M., de Haas, E. R. M., Nijsten, T., et al. (2007). Ongoing debate: which is the best surgical excision margin for basal cell carcinoma? Journal of the European Academy of Dermatology and Venereology, 21(4), 563-570.
Yudovsky, D., & Pilon, L. (2010). Hyperspectral imaging in medical diagnostics. Lasers in Surgery and Medicine, 42(5), 361-371.
Panasyuk, S. V., Yang, S., Faller, D. V., et al. (2007). Medical hyperspectral imaging to facilitate residual tumor identification during surgery. Cancer Biology & Therapy, 6(3), 439-446.
Lu, G., & Fei, B. (2014). Medical hyperspectral imaging: a review. Journal of Biomedical Optics, 19(1), 010901.
John, A., Niemeijer, M., Garvin, M. K., et al. (2010). Automated detection of diabetic retinopathy using optical coherence tomography. Journal of the American Academy of Dermatology, 62(6), 979-987.
Zuzak, K. J., Schaeberle, M. D., Lewis, E. N., & Levin, I. W. (2002). Visible reflectance hyperspectral imaging: characterization of a noninvasive, in vivo system for determining tissue perfusion. Analytical Chemistry, 74(9), 2021-2028.
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.
Litjens, G., Kooi, T., Bejnordi, B. E., et al. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60-88.
Gulshan, V., Peng, L., Coram, M., et al. (2016). Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA, 316(22), 2402-2410.
McKinney, S. M., Sieniek, M., Godbole, V., et al. (2020). International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 89-94.
Brinker, T. J., Hekler, A., Enk, A. H., et al. (2019). Deep neural networks are superior to dermatologists in melanoma image classification. European Journal of Cancer, 119, 11-17.
Esteva, A., Chou, K., Yeung, S., et al. (2019). Deep learning-enabled medical computer vision. NPJ Digital Medicine, 2(1), 71.Liu, Y., Jain, A., Eng, C., et al. (2019). A deep learning system for differential diagnosis of skin diseases. Nature Medicine, 25(6), 900-908.
Tschandl, P., Rosendahl, C., & Kittler, H. (2018). The HAM10000 dataset: A large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data, 5, 180161.