Leveraging swin transformer with ensemble of deep learning model for cervical cancer screening using colposcopy images DOI Creative Commons
D. Dakshayani Himabindu, E. Laxmi Lydia, M. Rajesh

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 6, 2025

Cervical cancer (CC) is the leading cancer, which mainly affects women worldwide. It generally occurs from abnormal cell evolution in cervix and a vital functional structure uterus. The importance of timely recognition cannot be overstated, has led to various screening methods such as colposcopy, Human Papillomavirus (HPV) testing, Pap smears identify potential threats enable early intervention. Early detection during precancerous phase crucial, it provides an opportunity for effective treatment. diagnosis CC depend on colposcopy cytology models. Deep learning (DL) appropriate technique computer vision, developed latent solution increase efficiency accuracy when equated conventional clinical inspection models that are vulnerable human error. This study presents Leveraging Swin Transformer with Ensemble Learning Model Cancer Screening (LSTEDL-CCS) images. presented LSTEDL-CCS aims detect classify Initially, wiener filtering (WF) model used image pre-processing. Next, swin transformer (ST) network utilized feature extraction. For process, ensemble method performed by employing three models, namely autoencoder (AE), bidirectional gated recurrent unit (BiGRU), deep belief (DBN). Finally, hyperparameter tuning DL techniques using Pelican Optimization Algorithm (POA). A comprehensive experimental analysis conducted, results evaluated under diverse metrics. performance validation methodology portrayed superior value 99.44% over existing

Язык: Английский

Leveraging swin transformer with ensemble of deep learning model for cervical cancer screening using colposcopy images DOI Creative Commons
D. Dakshayani Himabindu, E. Laxmi Lydia, M. Rajesh

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 6, 2025

Cervical cancer (CC) is the leading cancer, which mainly affects women worldwide. It generally occurs from abnormal cell evolution in cervix and a vital functional structure uterus. The importance of timely recognition cannot be overstated, has led to various screening methods such as colposcopy, Human Papillomavirus (HPV) testing, Pap smears identify potential threats enable early intervention. Early detection during precancerous phase crucial, it provides an opportunity for effective treatment. diagnosis CC depend on colposcopy cytology models. Deep learning (DL) appropriate technique computer vision, developed latent solution increase efficiency accuracy when equated conventional clinical inspection models that are vulnerable human error. This study presents Leveraging Swin Transformer with Ensemble Learning Model Cancer Screening (LSTEDL-CCS) images. presented LSTEDL-CCS aims detect classify Initially, wiener filtering (WF) model used image pre-processing. Next, swin transformer (ST) network utilized feature extraction. For process, ensemble method performed by employing three models, namely autoencoder (AE), bidirectional gated recurrent unit (BiGRU), deep belief (DBN). Finally, hyperparameter tuning DL techniques using Pelican Optimization Algorithm (POA). A comprehensive experimental analysis conducted, results evaluated under diverse metrics. performance validation methodology portrayed superior value 99.44% over existing

Язык: Английский

Процитировано

0