Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology

Cervical cancer (CC) remains a significant health issue, especially in low- and middle-income countries (LMICs). While Pap smears are the standard screening method, they have limitations, like low sensitivity and subjective interpretation. Liquid-based cytology (LBC) offers improvements but still re...

Full description

Saved in:
Bibliographic Details
Main Authors: Mariangel Rodríguez, Claudio Córdova, Isabel Benjumeda, Sebastián San Martín
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Computation
Subjects:
Online Access:https://www.mdpi.com/2079-3197/12/12/232
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850049366702686208
author Mariangel Rodríguez
Claudio Córdova
Isabel Benjumeda
Sebastián San Martín
author_facet Mariangel Rodríguez
Claudio Córdova
Isabel Benjumeda
Sebastián San Martín
author_sort Mariangel Rodríguez
collection DOAJ
description Cervical cancer (CC) remains a significant health issue, especially in low- and middle-income countries (LMICs). While Pap smears are the standard screening method, they have limitations, like low sensitivity and subjective interpretation. Liquid-based cytology (LBC) offers improvements but still relies on manual analysis. This study explored the potential of deep learning (DL) for automated cervical cell classification using both Pap smears and LBC samples. A novel image segmentation algorithm was employed to extract single-cell patches for training a ResNet-50 model. The model trained on LBC images achieved remarkably high sensitivity (0.981), specificity (0.979), and accuracy (0.980), outperforming previous CNN models. However, the Pap smear dataset model achieved significantly lower performance (0.688 sensitivity, 0.762 specificity, 0.8735 accuracy). This suggests that noisy and poor cell definition in Pap smears pose challenges for automated classification, whereas LBC provides better classifiable cells patches. These findings demonstrate the potential of AI-powered cervical cell classification for improving CC screening, particularly with LBC. The high accuracy and efficiency of DL models combined with effective segmentation can contribute to earlier detection and more timely intervention. Future research should focus on implementing explainable AI models to increase clinician trust and facilitate the adoption of AI-assisted CC screening in LMICs.
format Article
id doaj-art-b579e3762f5a4b98bd651e66d2951d6a
institution DOAJ
issn 2079-3197
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Computation
spelling doaj-art-b579e3762f5a4b98bd651e66d2951d6a2025-08-20T02:53:43ZengMDPI AGComputation2079-31972024-11-01121223210.3390/computation12120232Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based CytologyMariangel Rodríguez0Claudio Córdova1Isabel Benjumeda2Sebastián San Martín3PhD Program in Health Sciences and Engineering, Faculty of Medicine-Engineering-Sciences, Universidad de Valparaíso, Viña del Mar 2540064, ChileCenter of Interdisciplinary Biomedical and Engineering Research for Health (MEDING), School of Medicine, Faculty of Medicine, Universidad de Valparaíso, Viña del Mar 2540064, ChileDepartment of Sciences, Faculty of Liberal Arts, Adolfo Ibañez University, Viña del Mar 2200055, ChileCenter of Interdisciplinary Biomedical and Engineering Research for Health (MEDING), School of Medicine, Faculty of Medicine, Universidad de Valparaíso, Viña del Mar 2540064, ChileCervical cancer (CC) remains a significant health issue, especially in low- and middle-income countries (LMICs). While Pap smears are the standard screening method, they have limitations, like low sensitivity and subjective interpretation. Liquid-based cytology (LBC) offers improvements but still relies on manual analysis. This study explored the potential of deep learning (DL) for automated cervical cell classification using both Pap smears and LBC samples. A novel image segmentation algorithm was employed to extract single-cell patches for training a ResNet-50 model. The model trained on LBC images achieved remarkably high sensitivity (0.981), specificity (0.979), and accuracy (0.980), outperforming previous CNN models. However, the Pap smear dataset model achieved significantly lower performance (0.688 sensitivity, 0.762 specificity, 0.8735 accuracy). This suggests that noisy and poor cell definition in Pap smears pose challenges for automated classification, whereas LBC provides better classifiable cells patches. These findings demonstrate the potential of AI-powered cervical cell classification for improving CC screening, particularly with LBC. The high accuracy and efficiency of DL models combined with effective segmentation can contribute to earlier detection and more timely intervention. Future research should focus on implementing explainable AI models to increase clinician trust and facilitate the adoption of AI-assisted CC screening in LMICs.https://www.mdpi.com/2079-3197/12/12/232cervical cancerdeep learningcell segmentationpap smearliquid-based cytology
spellingShingle Mariangel Rodríguez
Claudio Córdova
Isabel Benjumeda
Sebastián San Martín
Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology
Computation
cervical cancer
deep learning
cell segmentation
pap smear
liquid-based cytology
title Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology
title_full Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology
title_fullStr Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology
title_full_unstemmed Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology
title_short Automated Cervical Cancer Screening Using Single-Cell Segmentation and Deep Learning: Enhanced Performance with Liquid-Based Cytology
title_sort automated cervical cancer screening using single cell segmentation and deep learning enhanced performance with liquid based cytology
topic cervical cancer
deep learning
cell segmentation
pap smear
liquid-based cytology
url https://www.mdpi.com/2079-3197/12/12/232
work_keys_str_mv AT mariangelrodriguez automatedcervicalcancerscreeningusingsinglecellsegmentationanddeeplearningenhancedperformancewithliquidbasedcytology
AT claudiocordova automatedcervicalcancerscreeningusingsinglecellsegmentationanddeeplearningenhancedperformancewithliquidbasedcytology
AT isabelbenjumeda automatedcervicalcancerscreeningusingsinglecellsegmentationanddeeplearningenhancedperformancewithliquidbasedcytology
AT sebastiansanmartin automatedcervicalcancerscreeningusingsinglecellsegmentationanddeeplearningenhancedperformancewithliquidbasedcytology