A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis
Skin lesion classification is a pivotal process in dermatology, enabling the early detection and precise diagnosis of skin diseases, leading to improved patient outcomes. Deep learning has shown great potential for this task by leveraging its ability to learn complex patterns from images. However, d...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10491261/ |
| _version_ | 1827281448061632512 |
|---|---|
| author | S. Remya T. Anjali Vijayan Sugumaran |
| author_facet | S. Remya T. Anjali Vijayan Sugumaran |
| author_sort | S. Remya |
| collection | DOAJ |
| description | Skin lesion classification is a pivotal process in dermatology, enabling the early detection and precise diagnosis of skin diseases, leading to improved patient outcomes. Deep learning has shown great potential for this task by leveraging its ability to learn complex patterns from images. However, diagnostic accuracy is compromised by exclusive reliance on single-modality images. This research work proposes an innovative framework that unifies a Vision Transformer model with transfer learning, channel attention mechanism, and ROI for the accurate detection of skin conditions, including skin cancer. The proposed approach blends computer vision and machine-learning techniques, leveraging a comprehensive dataset comprised of macroscopic dermoscopic images, appended with patient metadata. Compared with conventional techniques, the proposed methodology exhibits significant improvements in various parameters, including sensitivity, specificity, and precision. Moreover, it demonstrates outstanding performance in real-world datasets, reinforcing its potential for clinical implementation. With a remarkable accuracy of 99%, the method outperforms existing approaches. Overall, this investigation underscores the transformative impact of deep learning and multimodal data analysis in the dermoscopic domain, projecting substantial headway into the field of skin lesion analytic diagnosis. |
| first_indexed | 2024-04-24T09:01:27Z |
| format | Article |
| id | doaj.art-144806b19915466784835cfcf30be75a |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-24T09:01:27Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-144806b19915466784835cfcf30be75a2024-04-15T23:00:53ZengIEEEIEEE Access2169-35362024-01-0112507385075410.1109/ACCESS.2024.338534010491261A Novel Transfer Learning Framework for Multimodal Skin Lesion AnalysisS. Remya0https://orcid.org/0000-0003-2391-2013T. Anjali1https://orcid.org/0000-0003-1584-0179Vijayan Sugumaran2https://orcid.org/0000-0003-2557-3182Amrita School of Computing, Amrita Vishwa Vidyapeetham, Amritapuri, Kerala, IndiaAmrita School of Computing, Amrita Vishwa Vidyapeetham, Amritapuri, Kerala, IndiaDepartment of Decision and Information Sciences, Centre for Data Science and Big Data Analytics, School of Business, Oakland University, Rochester, MI, USASkin lesion classification is a pivotal process in dermatology, enabling the early detection and precise diagnosis of skin diseases, leading to improved patient outcomes. Deep learning has shown great potential for this task by leveraging its ability to learn complex patterns from images. However, diagnostic accuracy is compromised by exclusive reliance on single-modality images. This research work proposes an innovative framework that unifies a Vision Transformer model with transfer learning, channel attention mechanism, and ROI for the accurate detection of skin conditions, including skin cancer. The proposed approach blends computer vision and machine-learning techniques, leveraging a comprehensive dataset comprised of macroscopic dermoscopic images, appended with patient metadata. Compared with conventional techniques, the proposed methodology exhibits significant improvements in various parameters, including sensitivity, specificity, and precision. Moreover, it demonstrates outstanding performance in real-world datasets, reinforcing its potential for clinical implementation. With a remarkable accuracy of 99%, the method outperforms existing approaches. Overall, this investigation underscores the transformative impact of deep learning and multimodal data analysis in the dermoscopic domain, projecting substantial headway into the field of skin lesion analytic diagnosis.https://ieeexplore.ieee.org/document/10491261/Skin lesion classificationdermatologydeep learningmultimodal data analysistransfer learningvision transformer |
| spellingShingle | S. Remya T. Anjali Vijayan Sugumaran A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis IEEE Access Skin lesion classification dermatology deep learning multimodal data analysis transfer learning vision transformer |
| title | A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis |
| title_full | A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis |
| title_fullStr | A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis |
| title_full_unstemmed | A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis |
| title_short | A Novel Transfer Learning Framework for Multimodal Skin Lesion Analysis |
| title_sort | novel transfer learning framework for multimodal skin lesion analysis |
| topic | Skin lesion classification dermatology deep learning multimodal data analysis transfer learning vision transformer |
| url | https://ieeexplore.ieee.org/document/10491261/ |
| work_keys_str_mv | AT sremya anoveltransferlearningframeworkformultimodalskinlesionanalysis AT tanjali anoveltransferlearningframeworkformultimodalskinlesionanalysis AT vijayansugumaran anoveltransferlearningframeworkformultimodalskinlesionanalysis AT sremya noveltransferlearningframeworkformultimodalskinlesionanalysis AT tanjali noveltransferlearningframeworkformultimodalskinlesionanalysis AT vijayansugumaran noveltransferlearningframeworkformultimodalskinlesionanalysis |