An automatic cervical cell classification model based on improved DenseNet121 DOI Creative Commons
Yue Zhang,

Chunyu Ning,

Wenjing Yang

и другие.

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

Опубликована: Янв. 25, 2025

The cervical cell classification technique can determine the degree of cellular abnormality and pathological condition, which help doctors to detect risk cancer at an early stage improve cure survival rates patients. Addressing issue low accuracy in classification, a deep convolutional neural network A2SDNet121 is proposed. takes DenseNet121 as backbone network. Firstly, SE module embedded increase model's focus on nucleus region, contains important diagnostic information, reduce redundant information. Secondly, sizes kernel pooling window Stem layer are adjusted adapt characteristics images, so that model extract local detailed information more effectively. Finally, Atrous Dense Block (ADB) constructed, four ADB modules integrated into enable acquire global salient feature for two seven-classification tasks Herlev dataset 99.75% 99.14%, respectively. two, three, five-classification SIPaKMeD reaches 99.55%, 99.22%, Compared with other state-of-the-art algorithms, performs better multi-classification task cells, significantly efficiency screening.

Язык: Английский

Novelty Classification Model Use in Reinforcement Learning for Cervical Cancer DOI Open Access
Shakhnoza Muksimova, Sabina Umirzakova,

Khusanboy Shoraimov

и другие.

Cancers, Год журнала: 2024, Номер 16(22), С. 3782 - 3782

Опубликована: Ноя. 10, 2024

Cervical cancer significantly impacts global health, where early detection is piv- otal for improving patient outcomes. This study aims to enhance the accuracy of cervical diagnosis by addressing class imbalance through a novel hybrid deep learning model.

Язык: Английский

Процитировано

5
An automatic cervical cell classification model based on improved DenseNet121 DOI Creative Commons
Yue Zhang,

Chunyu Ning,

Wenjing Yang

и другие.

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

Опубликована: Янв. 25, 2025

The cervical cell classification technique can determine the degree of cellular abnormality and pathological condition, which help doctors to detect risk cancer at an early stage improve cure survival rates patients. Addressing issue low accuracy in classification, a deep convolutional neural network A2SDNet121 is proposed. takes DenseNet121 as backbone network. Firstly, SE module embedded increase model's focus on nucleus region, contains important diagnostic information, reduce redundant information. Secondly, sizes kernel pooling window Stem layer are adjusted adapt characteristics images, so that model extract local detailed information more effectively. Finally, Atrous Dense Block (ADB) constructed, four ADB modules integrated into enable acquire global salient feature for two seven-classification tasks Herlev dataset 99.75% 99.14%, respectively. two, three, five-classification SIPaKMeD reaches 99.55%, 99.22%, Compared with other state-of-the-art algorithms, performs better multi-classification task cells, significantly efficiency screening.

Язык: Английский

Процитировано

0