Cargando…

A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images

Routine Pap smears can facilitate early detection of cervical cancer and improve patient outcomes. The objective of this work is to develop an automated, clinically viable deep neural network for the multi-class Bethesda System diagnosis of multi-cell images in Liquid Pap smear samples. 8 deep learn...

Descripción completa

Detalles Bibliográficos
Autores principales: Karasu Benyes, Yasmin, Welch, E. Celeste, Singhal, Abhinav, Ou, Joyce, Tripathi, Anubhav
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9406372/
https://www.ncbi.nlm.nih.gov/pubmed/36010189
http://dx.doi.org/10.3390/diagnostics12081838
_version_ 1784774104813928448
author Karasu Benyes, Yasmin
Welch, E. Celeste
Singhal, Abhinav
Ou, Joyce
Tripathi, Anubhav
author_facet Karasu Benyes, Yasmin
Welch, E. Celeste
Singhal, Abhinav
Ou, Joyce
Tripathi, Anubhav
author_sort Karasu Benyes, Yasmin
collection PubMed
description Routine Pap smears can facilitate early detection of cervical cancer and improve patient outcomes. The objective of this work is to develop an automated, clinically viable deep neural network for the multi-class Bethesda System diagnosis of multi-cell images in Liquid Pap smear samples. 8 deep learning models were trained on a publicly available multi-class SurePath preparation dataset. This included the 5 best-performing transfer learning models, an ensemble, a novel convolutional neural network (CNN), and a CNN + autoencoder (AE). Additionally, each model was tested on a novel ThinPrep Pap dataset to determine model generalizability across different liquid Pap preparation methods with and without Deep CORAL domain adaptation. All models achieved accuracies >90% when classifying SurePath images. The AE CNN model, 99.80% smaller than the average transfer model, maintained an accuracy of 96.54%. During consecutive training attempts, individual transfer models had high variability in performance, whereas the CNN, AE CNN, and ensemble did not. ThinPrep Pap classification accuracies were notably lower but increased with domain adaptation, with ResNet101 achieving the highest accuracy at 92.65%. This indicates a potential area for future improvement: development of a globally relevant model that can function across different slide preparation methods.
format Online
Article
Text
id pubmed-9406372
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-94063722022-08-26 A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images Karasu Benyes, Yasmin Welch, E. Celeste Singhal, Abhinav Ou, Joyce Tripathi, Anubhav Diagnostics (Basel) Article Routine Pap smears can facilitate early detection of cervical cancer and improve patient outcomes. The objective of this work is to develop an automated, clinically viable deep neural network for the multi-class Bethesda System diagnosis of multi-cell images in Liquid Pap smear samples. 8 deep learning models were trained on a publicly available multi-class SurePath preparation dataset. This included the 5 best-performing transfer learning models, an ensemble, a novel convolutional neural network (CNN), and a CNN + autoencoder (AE). Additionally, each model was tested on a novel ThinPrep Pap dataset to determine model generalizability across different liquid Pap preparation methods with and without Deep CORAL domain adaptation. All models achieved accuracies >90% when classifying SurePath images. The AE CNN model, 99.80% smaller than the average transfer model, maintained an accuracy of 96.54%. During consecutive training attempts, individual transfer models had high variability in performance, whereas the CNN, AE CNN, and ensemble did not. ThinPrep Pap classification accuracies were notably lower but increased with domain adaptation, with ResNet101 achieving the highest accuracy at 92.65%. This indicates a potential area for future improvement: development of a globally relevant model that can function across different slide preparation methods. MDPI 2022-07-29 /pmc/articles/PMC9406372/ /pubmed/36010189 http://dx.doi.org/10.3390/diagnostics12081838 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Karasu Benyes, Yasmin
Welch, E. Celeste
Singhal, Abhinav
Ou, Joyce
Tripathi, Anubhav
A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images
title A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images
title_full A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images
title_fullStr A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images
title_full_unstemmed A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images
title_short A Comparative Analysis of Deep Learning Models for Automated Cross-Preparation Diagnosis of Multi-Cell Liquid Pap Smear Images
title_sort comparative analysis of deep learning models for automated cross-preparation diagnosis of multi-cell liquid pap smear images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9406372/
https://www.ncbi.nlm.nih.gov/pubmed/36010189
http://dx.doi.org/10.3390/diagnostics12081838
work_keys_str_mv AT karasubenyesyasmin acomparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT welcheceleste acomparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT singhalabhinav acomparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT oujoyce acomparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT tripathianubhav acomparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT karasubenyesyasmin comparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT welcheceleste comparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT singhalabhinav comparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT oujoyce comparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages
AT tripathianubhav comparativeanalysisofdeeplearningmodelsforautomatedcrosspreparationdiagnosisofmulticellliquidpapsmearimages