Minimal sourced and lightweight federated transfer learning models for skin cancer detection
Abstract One of the most fatal diseases that affect people is skin cancer. Because nevus and melanoma lesions are so similar and there is a high likelihood of false negative diagnoses challenges in hospitals. The aim of this paper is to propose and develop a technique to classify type of skin cancer...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-82402-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585819610677248 |
|---|---|
| author | Vikas Khullar Prabhjot Kaur Shubham Gargrish Anand Muni Mishra Prabhishek Singh Manoj Diwakar Anchit Bijalwan Indrajeet Gupta |
| author_facet | Vikas Khullar Prabhjot Kaur Shubham Gargrish Anand Muni Mishra Prabhishek Singh Manoj Diwakar Anchit Bijalwan Indrajeet Gupta |
| author_sort | Vikas Khullar |
| collection | DOAJ |
| description | Abstract One of the most fatal diseases that affect people is skin cancer. Because nevus and melanoma lesions are so similar and there is a high likelihood of false negative diagnoses challenges in hospitals. The aim of this paper is to propose and develop a technique to classify type of skin cancer with high accuracy using minimal resources and lightweight federated transfer learning models. Here minimal resource based pre-trained deep learning models including EfficientNetV2S, EfficientNetB3, ResNet50, and NasNetMobile have been used to apply transfer learning on data of shape $$\:\:224\times\:224\times\:3$$ . To compare with applied minimal resource transfer learning, same methodology has been applied using best identified model i.e. EfficientNetV2S for images of shape $$\:\:32\times\:32\times\:3$$ . The identified minimal and lightweight resource based EfficientNetV2S with images of shape $$\:32\times\:32\times\:3$$ have been applied for federated learning ecosystem. Both, identically and non-identically distributed datasets of shape $$\:32\times\:32\times\:3$$ have been applied and analyzed through federated learning implementations. The results have been analyzed to show the impact of low-pixel images with non-identical distributions over clients using parameters such as accuracy, precision, recall and categorical losses. The classification of skin cancer shows an accuracy of IID 89.83% and Non-IID 90.64%. |
| format | Article |
| id | doaj-art-fcae7e60c9a64df8abaff2fba7a2bc33 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-fcae7e60c9a64df8abaff2fba7a2bc332025-01-26T12:29:53ZengNature PortfolioScientific Reports2045-23222025-01-0115111810.1038/s41598-024-82402-xMinimal sourced and lightweight federated transfer learning models for skin cancer detectionVikas Khullar0Prabhjot Kaur1Shubham Gargrish2Anand Muni Mishra3Prabhishek Singh4Manoj Diwakar5Anchit Bijalwan6Indrajeet Gupta7Chitkara University Institute of Engineering Technology, Chitkara UniversityChitkara University Institute of Engineering Technology, Chitkara UniversityChitkara University Institute of Engineering Technology, Chitkara UniversityChandigarh Engineering College, Chandigarh Group of Colleges, JhanjeriSchool of Computer Science Engineering and Technology, Bennett UniversityCSE Department, Graphic Era Deemed to be UniversityFaculty of Electrical and Computer Engineering, Arba Minch UniversitySchool of Computer Science and AI, SR UniversityAbstract One of the most fatal diseases that affect people is skin cancer. Because nevus and melanoma lesions are so similar and there is a high likelihood of false negative diagnoses challenges in hospitals. The aim of this paper is to propose and develop a technique to classify type of skin cancer with high accuracy using minimal resources and lightweight federated transfer learning models. Here minimal resource based pre-trained deep learning models including EfficientNetV2S, EfficientNetB3, ResNet50, and NasNetMobile have been used to apply transfer learning on data of shape $$\:\:224\times\:224\times\:3$$ . To compare with applied minimal resource transfer learning, same methodology has been applied using best identified model i.e. EfficientNetV2S for images of shape $$\:\:32\times\:32\times\:3$$ . The identified minimal and lightweight resource based EfficientNetV2S with images of shape $$\:32\times\:32\times\:3$$ have been applied for federated learning ecosystem. Both, identically and non-identically distributed datasets of shape $$\:32\times\:32\times\:3$$ have been applied and analyzed through federated learning implementations. The results have been analyzed to show the impact of low-pixel images with non-identical distributions over clients using parameters such as accuracy, precision, recall and categorical losses. The classification of skin cancer shows an accuracy of IID 89.83% and Non-IID 90.64%.https://doi.org/10.1038/s41598-024-82402-xSkin cancerFederated learningConvolutional neural networkLesionsDiseaseTransfer learning |
| spellingShingle | Vikas Khullar Prabhjot Kaur Shubham Gargrish Anand Muni Mishra Prabhishek Singh Manoj Diwakar Anchit Bijalwan Indrajeet Gupta Minimal sourced and lightweight federated transfer learning models for skin cancer detection Scientific Reports Skin cancer Federated learning Convolutional neural network Lesions Disease Transfer learning |
| title | Minimal sourced and lightweight federated transfer learning models for skin cancer detection |
| title_full | Minimal sourced and lightweight federated transfer learning models for skin cancer detection |
| title_fullStr | Minimal sourced and lightweight federated transfer learning models for skin cancer detection |
| title_full_unstemmed | Minimal sourced and lightweight federated transfer learning models for skin cancer detection |
| title_short | Minimal sourced and lightweight federated transfer learning models for skin cancer detection |
| title_sort | minimal sourced and lightweight federated transfer learning models for skin cancer detection |
| topic | Skin cancer Federated learning Convolutional neural network Lesions Disease Transfer learning |
| url | https://doi.org/10.1038/s41598-024-82402-x |
| work_keys_str_mv | AT vikaskhullar minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT prabhjotkaur minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT shubhamgargrish minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT anandmunimishra minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT prabhisheksingh minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT manojdiwakar minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT anchitbijalwan minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection AT indrajeetgupta minimalsourcedandlightweightfederatedtransferlearningmodelsforskincancerdetection |