Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology
Abstract Objective Rapid on-site evaluation (ROSE) of respiratory cytology specimens is a critical technique for accurate and timely diagnosis of lung cancer. However, in China, limited familiarity with the Diff-Quik staining method and a shortage of trained cytopathologists hamper utilization of RO...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BMC Cancer |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12885-024-13402-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559396796071936 |
|---|---|
| author | Wei Gong Deep K. Vaishnani Xuan-Chen Jin Jing Zeng Wei Chen Huixia Huang Yu-Qing Zhou Khaing Wut Yi Hla Chen Geng Jun Ma |
| author_facet | Wei Gong Deep K. Vaishnani Xuan-Chen Jin Jing Zeng Wei Chen Huixia Huang Yu-Qing Zhou Khaing Wut Yi Hla Chen Geng Jun Ma |
| author_sort | Wei Gong |
| collection | DOAJ |
| description | Abstract Objective Rapid on-site evaluation (ROSE) of respiratory cytology specimens is a critical technique for accurate and timely diagnosis of lung cancer. However, in China, limited familiarity with the Diff-Quik staining method and a shortage of trained cytopathologists hamper utilization of ROSE. Therefore, developing an improved deep learning model to assist clinicians in promptly and accurately evaluating Diff-Quik stained cytology samples during ROSE has important clinical value. Methods Retrospectively, 116 digital images of Diff-Quik stained cytology samples were obtained from whole slide scans. These included 6 diagnostic categories - carcinoid, normal cells, adenocarcinoma, squamous cell carcinoma, non-small cell carcinoma, and small cell carcinoma. All malignant diagnoses were confirmed by histopathology and immunohistochemistry. The test image set was presented to 3 cytopathologists from different hospitals with varying levels of experience, as well as an artificial intelligence system, as single-choice questions. Results The diagnostic accuracy of the cytopathologists correlated with their years of practice and hospital setting. The AI model demonstrated proficiency comparable to the humans. Importantly, all combinations of AI assistance and human cytopathologist increased diagnostic efficiency to varying degrees. Conclusions This deep learning model shows promising capability as an aid for on-site diagnosis of respiratory cytology samples. However, human expertise remains essential to the diagnostic process. |
| format | Article |
| id | doaj-art-d5ce6bfa7b6b4c339149c24ce8fff170 |
| institution | Kabale University |
| issn | 1471-2407 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Cancer |
| spelling | doaj-art-d5ce6bfa7b6b4c339149c24ce8fff1702025-01-05T12:33:05ZengBMCBMC Cancer1471-24072025-01-0125111010.1186/s12885-024-13402-3Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytologyWei Gong0Deep K. Vaishnani1Xuan-Chen Jin2Jing Zeng3Wei Chen4Huixia Huang5Yu-Qing Zhou6Khaing Wut Yi Hla7Chen Geng8Jun Ma9Department of Pathology, Lishui Municipal Central HospitalSchool of International Studies, Wenzhou Medical UniversitySchool of Clinical Medicine, Wenzhou Medical UniversitySchool of Clinical Medicine, Wenzhou Medical UniversityRenji College, Wenzhou Medical UniversityDepartment of Archives, Lishui Second People’s HospitalSchool of International Studies, Wenzhou Medical UniversitySchool of International Studies, Wenzhou Medical UniversitySuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of SciencesDepartment of Pathology, The First Affiliated Hospital of Wenzhou Medical UniversityAbstract Objective Rapid on-site evaluation (ROSE) of respiratory cytology specimens is a critical technique for accurate and timely diagnosis of lung cancer. However, in China, limited familiarity with the Diff-Quik staining method and a shortage of trained cytopathologists hamper utilization of ROSE. Therefore, developing an improved deep learning model to assist clinicians in promptly and accurately evaluating Diff-Quik stained cytology samples during ROSE has important clinical value. Methods Retrospectively, 116 digital images of Diff-Quik stained cytology samples were obtained from whole slide scans. These included 6 diagnostic categories - carcinoid, normal cells, adenocarcinoma, squamous cell carcinoma, non-small cell carcinoma, and small cell carcinoma. All malignant diagnoses were confirmed by histopathology and immunohistochemistry. The test image set was presented to 3 cytopathologists from different hospitals with varying levels of experience, as well as an artificial intelligence system, as single-choice questions. Results The diagnostic accuracy of the cytopathologists correlated with their years of practice and hospital setting. The AI model demonstrated proficiency comparable to the humans. Importantly, all combinations of AI assistance and human cytopathologist increased diagnostic efficiency to varying degrees. Conclusions This deep learning model shows promising capability as an aid for on-site diagnosis of respiratory cytology samples. However, human expertise remains essential to the diagnostic process.https://doi.org/10.1186/s12885-024-13402-3Artificial Intelligence (AI)Lung CancerComputer-aided diagnosisResNet-18Rapid on-site evaluation (ROSE) |
| spellingShingle | Wei Gong Deep K. Vaishnani Xuan-Chen Jin Jing Zeng Wei Chen Huixia Huang Yu-Qing Zhou Khaing Wut Yi Hla Chen Geng Jun Ma Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology BMC Cancer Artificial Intelligence (AI) Lung Cancer Computer-aided diagnosis ResNet-18 Rapid on-site evaluation (ROSE) |
| title | Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology |
| title_full | Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology |
| title_fullStr | Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology |
| title_full_unstemmed | Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology |
| title_short | Evaluation of an enhanced ResNet-18 classification model for rapid On-site diagnosis in respiratory cytology |
| title_sort | evaluation of an enhanced resnet 18 classification model for rapid on site diagnosis in respiratory cytology |
| topic | Artificial Intelligence (AI) Lung Cancer Computer-aided diagnosis ResNet-18 Rapid on-site evaluation (ROSE) |
| url | https://doi.org/10.1186/s12885-024-13402-3 |
| work_keys_str_mv | AT weigong evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT deepkvaishnani evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT xuanchenjin evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT jingzeng evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT weichen evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT huixiahuang evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT yuqingzhou evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT khaingwutyihla evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT chengeng evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology AT junma evaluationofanenhancedresnet18classificationmodelforrapidonsitediagnosisinrespiratorycytology |