1. Home
  2. Vol 29, No 4 (2025)
  3. Zhao

Detection of precancerous lesions in cervical images of perimenopausal women using U-net deep learning

Na Zhao(1), Yan Gao(2), Fang Li(3), Jingtian Shi(4), Yanni Huang(5), Hongyun Ma(6),


(1) Department of Gynecology, Peking University First Hospital Ningxia Women and Children’s Hospital (Ningxia Hui Autonomous Region Maternal and Child Health Hospital), Yinchuan 750001, Ningxia Hui Autonomous Region, China
(2) Department of Obstetrics and Gynecology, Yinchuan Second People's Hospital, Yinchuan, 750011, Ningxia Hui Autonomous Region, China
(3) Department of Pathology, Peking University First Hospital Ningxia Women and Children’s Hospital (Ningxia Hui Autonomous Region Maternal and Child Health Hospital), Yinchuan 750001, Ningxia Hui Autonomous Region, China
(4) Department of Pharmacy, Peking University First Hospital Ningxia Women and Children’s Hospital (Ningxia Hui Autonomous Region Maternal and Child Health Hospital), Yinchuan 750001, Ningxia Hui Autonomous Region, China
(5) Department of Gynecology, Peking University First Hospital Ningxia Women and Children’s Hospital (Ningxia Hui Autonomous Region Maternal and Child Health Hospital), Yinchuan 750001, Ningxia Hui Autonomous Region, China
(6) Department of Obstetrics and Gynecology, People's Hospital of Ningxia Hui Autonomous Region, Ningxia Medical University, Yinchuan 750000, Ningxia Hui Autonomous Region, China
Corresponding Author

Abstract


Due to physiological changes during the perimenopausal period, the morphology of cervical cells undergoes certain alterations. Accurate cell image segmentation and lesion identification are of great significance for the early detection of precancerous lesions. Traditional detection methods may have certain limitations, thereby creating an urgent need for the development of more effective models. This study aimed to develop a highly efficient and accurate cervical cell image segmentation and recognition model to enhance the detection of precancerous lesions in perimenopausal women. based on U-shaped Network(U-Net) and Residual Network (ResNet). The model integrates U-Net with Segmentation Network (SegNet) and incorporates the Squeeze-and-Excitation (SE) attention mechanism to create the 2Se/U-Net segmentation model. Additionally, ResNet is optimized with the local discriminant loss function (LD-loss) and deep residual learning (DRL) blocks to develop the LD/ResNet lesion recognition model. The performance of the models is evaluated using data from 103 cytology images of perimenopausal women, focusing on segmentation metrics like mean pixel accuracy (MPA) and mean intersection over union (mIoU), as well as lesion detection metrics such as accuracy (Acc), precision (Pre), recall (Re), and F1-score (F1). Results show that the 2Se/U-Net model achieves an MPA of 92.63% and mIoU of 96.93%, outperforming U-Net by 12.48% and 9.47%, respectively. The LD/ResNet model demonstrates over 97.09% accuracy in recognizing cervical cells and achieves high detection performance for precancerous lesions, with Acc, Pre, and Re at 98.95%, 99.36%, and 98.89%, respectively. The model shows great potential for enhancing cervical cancer screening in clinical settings. (

References


Mansuri SM, Shrivastawa N and Parveen, N. Study on

histopathological analycis at endometrium pattern in

peri-menopause patient with abnormal uterine

bleeding. Journal of Cardiovascular Disease

Research 2024; 15: 06.

Liu X, Liu H, Xiong Y, Yang L, Wang C, Zhang R and Zhu

X. Postmenopausal osteoporosis is associated with the

regulation of SP, CGRP, VIP, and NPY. Biomedicine

& Pharmacotherapy 2018; 104: 742-750.

AL-Qarawy AKM, AL-Obaidi ZME and Mohammad AK.

Role of dehydroepiandrosterone hormone in relieving

of vasomotor symptoms in post and peri menopause

women. In: Obstetrics and Gynaecology Forum.

; 34(3): 105-108.

An H, Ding L, Ma M, Huang A, Gan Y, Sheng D and Zhang

X. Deep Learning-Based Recognition of Cervical

Squamous Interepithelial Lesions. Diagnostics 2023;

(10): 1720.

Liu X, Yang R, Xiong T, Yang X, Li W, Song L and Geng L.

CBCT-to-CT Synthesis for Cervical Cancer Adaptive

Radiotherapy via U-Net-Based Model Hierarchically

Trained with Hybrid Dataset. Cancers 2023; 15(22):

Shinohara T, Murakami K and Matsumura N. Diagnosis

assistance in colposcopy by segmenting acetowhite

epithelium using U-Net with images before and after

acetic acid solution application. Diagnostics 2023;

(9): 1596.

Huang Y, Zhu H, Wang P and Dong D. Segmentation of

overlapping cervical smear cells based on U-Net and

improved level set. IEEE 2019;3031-3035.

Wang J, Zhang B, Wang Y, Zhou C, Vonsky M S,

Mitrofanova L B and Li Q. CrossU-Net: Dual- modality cross-attention U-Net for segmentation of

precancerous lesions in gastric cancer. Computerized

Medical Imaging and Graphics 2024;112, 102339.

Desiani A, Utama Y, Arhami M, Affandi AK, Sasongko MA

and Ramayanti I. Simple Data Augmentation and U

Net CNN for Neclui Binary Segmentation on Pap

Smear Images. Journal of Electronics, Electromedical

Engineering, and Medical Informatics 2024; 6(3):

-275.

Teixeira JBA, Rezende MT, Diniz DN, Carneiro CM, Luz

EJDS, Souza, MJ and Bianchi AGC. Segmentation of

cervical nuclei using convolutional neural network for

conventional cytology. Computer Methods in

Biomechanics and Biomedical Engineering: Imaging

& Visualization 2023; 11(5): 1876-1888.

Liu H and Tang T. MAPK signaling pathway-based glioma

subtypes, machine-learning risk model, and key hub

proteins identification. Scientific Reports 2023; 13(1):

Zhang C, Lu W, Wu J, Ni C and Wang H. SegNet Network

Architecture for Deep Learning Image Segmentation

and Its Integrated Applications and Prospects.

Academic Journal of Science and Technology 2024;

(2): 224-229.

Zhou S, Li X, Fu Y, Ouyang X, Yang J, Grzegorzek M and

Li C. High Order Conditional Random Field Based

Cervical Cancer Histopathological Image

Classification. In: International Conference on Image,

Vision and Intelligent Systems. Singapore: Springer

Nature Singapore 2023; 3-16.

Fauzi H, Ansori RB, Siadari T, Harsono AB and Rahmah

QN. Classification of Cervical Cancer Images Using

Deep Residual Network Architecture. International

Journal of Artificial Intelligence Research 2023; 7(1):

-63.

Liu Y, Chen C, Xie X, Lv X and Chen C. For cervical cancer

diagnosis: Tissue Raman spectroscopy and multi

level feature fusion with SENet attention mechanism.

Spectrochimica Acta Part A: Molecular and

Biomolecular Spectroscopy 2023; 303: 123147.

Zhao Y, Fu C, Xu S, Cao L and Ma HF. LFANet:

Lightweight feature attention network for abnormal

cell segmentation in cervical cytology images.

Computers in Biology and Medicine 2022; 145:

Alquran H, Mustafa WA, Qasmieh IA, Yacob YM, Alsalatie

M, Al-Issa Y and Alqudah AM. Cervical cancer

classification using combined machine learning and

deep learning approach. Comput Mater Contin 2022;

(3): 5117-5134.

Nazir N, Sarwar A, Saini BS and Shams R. A robust deep

learning approach for accurate segmentation of

cytoplasm and nucleus in noisy pap smear images.

Computation 2023; 11(10): 195.

Li G, Sun C, Xu C, Zheng Y and Wang K. Cervical cell

segmentation method based on global dependency and

local attention. Applied Sciences 2022; 12(15): 7742.

Fekri-Ershad S and Ramakrishnan, S. Cervical cancer

diagnosis based on modified uniform local ternary

patterns and feed forward multilayer network

optimized by genetic algorithm. Computers in Biology

and Medicine 2022; 144: 105392.

Shiny TL and Parasuraman K. ROI Extraction and Nuclei

Classification of Pap Smear Images for Cervical

Cancer Detection. In: 2023 2nd International

Conference on Vision Towards Emerging Trends in

Communication and Networking Technologies

(ViTECoN). IEEE 2023; 1-7.

Chitra B and Kumar SS. Early cervical cancer diagnosis

using Sooty tern-optimized CNN-LSTM classifier.

International Journal of Imaging Systems and

Technology 2022; 32(6): 1846-1860.

Li G, Shi G and Jiao J. YOLOv5-KCB: A new method for

individual pig detection using optimized K-means,

CA attention mechanism and a bi-directional feature

pyramid network. Sensors 2023; 23(11): 5242.

Zhang Q, Liang Y, Zhang Y, Tao Z, Li R and Bi H. A

comparative study of attention mechanism based deep

learning methods for bladder tumor segmentation.

International Journal of Medical Informatics 2023;

: 104984.

Rose MM, Dhamodharan S, Revathidevi S, Chakkarappan

SR, Jagadeesan MG, Subbiah S and Munirajan, AK.

High incidence of PI3K pathway gene mutations in

South Indian cervical cancers. Cancer Genet. 2022;

-265: 100-108.

Özbay E and Özbay FA. Interpretable pap-smear image

retrieval for cervical cancer detection with rotation

invariance mask generation deep hashing. Computers

in Biology and Medicine 2023; 154: 106574.

Rudiansyah R, Iryani, L, Kesuma LI, Sari P and Alamsyah

A. Combination of Image Enhancement and U-Net

Architecture for Cervical Cell Semantic

Segmentation. JOURNAL OF INFORMATICS AND

TELECOMMUNICATION ENGINEERING 2024;

(2): 575-586.

Rasheed A, Shirazi SH, Umar AI, Shahzad M, Yousaf W

and Khan Z. Cervical cell’s nucleus segmentation

through an improved UNet architecture. Plos one

; 18(10): e0283568.

Lavanya Devi N and Thirumurugan P. Cervical cancer

classification from pap smear images using modified

fuzzy C means, PCA, and KNN. IETE Journal of

Research 2022; 68(3): 1591-1598.


Full Text: PDF

Article Metrics

Abstract View : 846 times
PDF Download : 258 times

Refbacks

  • There are currently no refbacks.