Advances in colorectal cancer diagnosis using optimal deep feature fusion approach on biomedical images DOI Creative Commons
Sultan Alotaibi, Manal Abdullah Alohali, Mashael Maashi

и другие.

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

Опубликована: Фев. 4, 2025

Colorectal cancer (CRC) is the second popular in females and third males, with an increased number of cases. Pathology diagnoses complemented predictive prognostic biomarker information first step for personalized treatment. Histopathological image (HI) analysis benchmark pathologists to rank colorectal various kinds. However, pathologists' are highly subjective susceptible inaccurate diagnoses. The improved diagnosis load pathology laboratory, incorporated reported intra- inter-variability assessment, has prompted quest consistent machine-based techniques be integrated into routine practice. In healthcare field, artificial intelligence (AI) achieved extraordinary achievements applications. Lately, computer-aided (CAD) based on HI progressed rapidly increase machine learning (ML) deep (DL) models. This study introduces a novel Cancer Diagnosis using Optimal Deep Feature Fusion Approach Biomedical Images (CCD-ODFFBI) method. primary objective CCD-ODFFBI technique examine biomedical images identify (CRC). technique, median filtering (MF) approach initially utilized noise elimination. utilizes fusion three DL models, MobileNet, SqueezeNet, SE-ResNet, feature extraction. Moreover, models' hyperparameter selection performed Osprey optimization algorithm (OOA). Finally, belief network (DBN) model employed classify CRC. A series simulations accomplished highlight significant results method under Warwick-QU dataset. comparison showed superior accuracy value 99.39% over existing techniques.

Язык: Английский

Advances in colorectal cancer diagnosis using optimal deep feature fusion approach on biomedical images DOI Creative Commons
Sultan Alotaibi, Manal Abdullah Alohali, Mashael Maashi

и другие.

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

Опубликована: Фев. 4, 2025

Colorectal cancer (CRC) is the second popular in females and third males, with an increased number of cases. Pathology diagnoses complemented predictive prognostic biomarker information first step for personalized treatment. Histopathological image (HI) analysis benchmark pathologists to rank colorectal various kinds. However, pathologists' are highly subjective susceptible inaccurate diagnoses. The improved diagnosis load pathology laboratory, incorporated reported intra- inter-variability assessment, has prompted quest consistent machine-based techniques be integrated into routine practice. In healthcare field, artificial intelligence (AI) achieved extraordinary achievements applications. Lately, computer-aided (CAD) based on HI progressed rapidly increase machine learning (ML) deep (DL) models. This study introduces a novel Cancer Diagnosis using Optimal Deep Feature Fusion Approach Biomedical Images (CCD-ODFFBI) method. primary objective CCD-ODFFBI technique examine biomedical images identify (CRC). technique, median filtering (MF) approach initially utilized noise elimination. utilizes fusion three DL models, MobileNet, SqueezeNet, SE-ResNet, feature extraction. Moreover, models' hyperparameter selection performed Osprey optimization algorithm (OOA). Finally, belief network (DBN) model employed classify CRC. A series simulations accomplished highlight significant results method under Warwick-QU dataset. comparison showed superior accuracy value 99.39% over existing techniques.

Язык: Английский

Процитировано

0