Performance metrics of Xception approaches.
<div><p>Skin lesions encompass a variety of skin abnormalities, including skin diseases that affect structure and function, and skin cancer, which can be fatal and arise from abnormal cell growth. Early detection of lesions and automated prediction is crucial, yet accurately identifying...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , , |
| منشور في: |
2024
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1852025707079663616 |
|---|---|
| author | Anwar Hossain Efat (19942283) |
| author2 | S. M. Mahedy Hasan (19942286) Md. Palash Uddin (19139222) Md. Al Mamun (12800084) |
| author2_role | author author author |
| author_facet | Anwar Hossain Efat (19942283) S. M. Mahedy Hasan (19942286) Md. Palash Uddin (19139222) Md. Al Mamun (12800084) |
| author_role | author |
| dc.creator.none.fl_str_mv | Anwar Hossain Efat (19942283) S. M. Mahedy Hasan (19942286) Md. Palash Uddin (19139222) Md. Al Mamun (12800084) |
| dc.date.none.fl_str_mv | 2024-10-24T17:36:23Z |
| dc.identifier.none.fl_str_mv | 10.1371/journal.pone.0309430.t011 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/dataset/Performance_metrics_of_Xception_approaches_/27293978 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Immunology Cancer Computational Biology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified ham1000 dataset demonstrates dominance dispersion remains convolutional neural network approach achieves 94 achieving desired outcomes abnormal cell growth level ensemble approach identify responsible regions facilitating early diagnosis including skin diseases customized transfer learning approach &# 8220 igpa ),&# 8221 determining optimal weights model &# 8217 &# 8220 ensemble learning specific regions early detection skin cancer skin abnormalities machine learning deep learning triple attention surpassing state specifically involves primary objective novel method multiple levels integral component exact focus enhance explainability empirical evaluation el ). currently lacking current studies art methods affect structure |
| dc.title.none.fl_str_mv | Performance metrics of Xception approaches. |
| dc.type.none.fl_str_mv | Dataset info:eu-repo/semantics/publishedVersion dataset |
| description | <div><p>Skin lesions encompass a variety of skin abnormalities, including skin diseases that affect structure and function, and skin cancer, which can be fatal and arise from abnormal cell growth. Early detection of lesions and automated prediction is crucial, yet accurately identifying responsible regions post-dominance dispersion remains a challenge in current studies. Thus, we propose a Convolutional Neural Network (CNN)-based approach employing a Customized Transfer Learning (CTL) model and Triple Attention (TA) modules in conjunction with Ensemble Learning (EL). While Ensemble Learning has become an integral component of both Machine Learning (ML) and Deep Learning (DL) methodologies, a specific technique ensuring optimal allocation of weights for each model’s prediction is currently lacking. Consequently, the primary objective of this study is to introduce a novel method for determining optimal weights to aggregate the contributions of models for achieving desired outcomes. We term this approach “Information Gain Proportioned Averaging (IGPA),” further refining it to “Multi-Level Information Gain Proportioned Averaging (ML-IGPA),” which specifically involves the utilization of IGPA at multiple levels. Empirical evaluation of the HAM1000 dataset demonstrates that our approach achieves 94.93% accuracy with ML-IGPA, surpassing state-of-the-art methods. Given previous studies’ failure to elucidate the exact focus of black-box models on specific regions, we utilize the Gradient Class Activation Map (GradCAM) to identify responsible regions and enhance explainability. Our study enhances both accuracy and interpretability, facilitating early diagnosis and preventing the consequences of neglecting skin lesion detection, thereby addressing issues related to time, accessibility, and costs.</p></div> |
| eu_rights_str_mv | openAccess |
| id | Manara_ab55ddf00a8fa7b37229559c661aa04f |
| identifier_str_mv | 10.1371/journal.pone.0309430.t011 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/27293978 |
| publishDate | 2024 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Performance metrics of Xception approaches.Anwar Hossain Efat (19942283)S. M. Mahedy Hasan (19942286)Md. Palash Uddin (19139222)Md. Al Mamun (12800084)ImmunologyCancerComputational BiologyBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedham1000 dataset demonstratesdominance dispersion remainsconvolutional neural networkapproach achieves 94achieving desired outcomesabnormal cell growthlevel ensemble approachidentify responsible regionsfacilitating early diagnosisincluding skin diseasescustomized transfer learningapproach &# 8220igpa ),&# 8221determining optimal weightsmodel &# 8217&# 8220ensemble learningspecific regionsearly detectionskin cancerskin abnormalitiesmachine learningdeep learningtriple attentionsurpassing statespecifically involvesprimary objectivenovel methodmultiple levelsintegral componentexact focusenhance explainabilityempirical evaluationel ).currently lackingcurrent studiesart methodsaffect structure<div><p>Skin lesions encompass a variety of skin abnormalities, including skin diseases that affect structure and function, and skin cancer, which can be fatal and arise from abnormal cell growth. Early detection of lesions and automated prediction is crucial, yet accurately identifying responsible regions post-dominance dispersion remains a challenge in current studies. Thus, we propose a Convolutional Neural Network (CNN)-based approach employing a Customized Transfer Learning (CTL) model and Triple Attention (TA) modules in conjunction with Ensemble Learning (EL). While Ensemble Learning has become an integral component of both Machine Learning (ML) and Deep Learning (DL) methodologies, a specific technique ensuring optimal allocation of weights for each model’s prediction is currently lacking. Consequently, the primary objective of this study is to introduce a novel method for determining optimal weights to aggregate the contributions of models for achieving desired outcomes. We term this approach “Information Gain Proportioned Averaging (IGPA),” further refining it to “Multi-Level Information Gain Proportioned Averaging (ML-IGPA),” which specifically involves the utilization of IGPA at multiple levels. Empirical evaluation of the HAM1000 dataset demonstrates that our approach achieves 94.93% accuracy with ML-IGPA, surpassing state-of-the-art methods. Given previous studies’ failure to elucidate the exact focus of black-box models on specific regions, we utilize the Gradient Class Activation Map (GradCAM) to identify responsible regions and enhance explainability. Our study enhances both accuracy and interpretability, facilitating early diagnosis and preventing the consequences of neglecting skin lesion detection, thereby addressing issues related to time, accessibility, and costs.</p></div>2024-10-24T17:36:23ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1371/journal.pone.0309430.t011https://figshare.com/articles/dataset/Performance_metrics_of_Xception_approaches_/27293978CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/272939782024-10-24T17:36:23Z |
| spellingShingle | Performance metrics of Xception approaches. Anwar Hossain Efat (19942283) Immunology Cancer Computational Biology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified ham1000 dataset demonstrates dominance dispersion remains convolutional neural network approach achieves 94 achieving desired outcomes abnormal cell growth level ensemble approach identify responsible regions facilitating early diagnosis including skin diseases customized transfer learning approach &# 8220 igpa ),&# 8221 determining optimal weights model &# 8217 &# 8220 ensemble learning specific regions early detection skin cancer skin abnormalities machine learning deep learning triple attention surpassing state specifically involves primary objective novel method multiple levels integral component exact focus enhance explainability empirical evaluation el ). currently lacking current studies art methods affect structure |
| status_str | publishedVersion |
| title | Performance metrics of Xception approaches. |
| title_full | Performance metrics of Xception approaches. |
| title_fullStr | Performance metrics of Xception approaches. |
| title_full_unstemmed | Performance metrics of Xception approaches. |
| title_short | Performance metrics of Xception approaches. |
| title_sort | Performance metrics of Xception approaches. |
| topic | Immunology Cancer Computational Biology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified ham1000 dataset demonstrates dominance dispersion remains convolutional neural network approach achieves 94 achieving desired outcomes abnormal cell growth level ensemble approach identify responsible regions facilitating early diagnosis including skin diseases customized transfer learning approach &# 8220 igpa ),&# 8221 determining optimal weights model &# 8217 &# 8220 ensemble learning specific regions early detection skin cancer skin abnormalities machine learning deep learning triple attention surpassing state specifically involves primary objective novel method multiple levels integral component exact focus enhance explainability empirical evaluation el ). currently lacking current studies art methods affect structure |