Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid
BackgroundIt is a critical challenge to diagnose leptomeningeal metastasis (LM), given its technical difficulty and the lack of typical symptoms. The existing gold standard of diagnosing LM is to use positive cerebrospinal fluid (CSF) cytology, which consumes significantly more time to classify cell...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-02-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2022.821594/full |
| _version_ | 1818985096337686528 |
|---|---|
| author | Wenjin Yu Wenjin Yu Wenjin Yu Yangyang Liu Yunsong Zhao Haofan Huang Jiahao Liu Xiaofeng Yao Jingwen Li Zhen Xie Luyue Jiang Heping Wu Xinhao Cao Jiaming Zhou Yuting Guo Gaoyang Li Matthew Xinhu Ren Yi Quan Tingmin Mu Guillermo Ayuso Izquierdo Guoxun Zhang Guoxun Zhang Runze Zhao Di Zhao Jiangyun Yan Haijun Zhang Junchao Lv Qian Yao Yan Duan Huimin Zhou Tingting Liu Ying He Ting Bian Wen Dai Jiahui Huai Xiyuan Wang Qian He Yi Gao Yi Gao Yi Gao Yi Gao Wei Ren Gang Niu Gang Zhao Gang Zhao |
| author_facet | Wenjin Yu Wenjin Yu Wenjin Yu Yangyang Liu Yunsong Zhao Haofan Huang Jiahao Liu Xiaofeng Yao Jingwen Li Zhen Xie Luyue Jiang Heping Wu Xinhao Cao Jiaming Zhou Yuting Guo Gaoyang Li Matthew Xinhu Ren Yi Quan Tingmin Mu Guillermo Ayuso Izquierdo Guoxun Zhang Guoxun Zhang Runze Zhao Di Zhao Jiangyun Yan Haijun Zhang Junchao Lv Qian Yao Yan Duan Huimin Zhou Tingting Liu Ying He Ting Bian Wen Dai Jiahui Huai Xiyuan Wang Qian He Yi Gao Yi Gao Yi Gao Yi Gao Wei Ren Gang Niu Gang Zhao Gang Zhao |
| author_sort | Wenjin Yu |
| collection | DOAJ |
| description | BackgroundIt is a critical challenge to diagnose leptomeningeal metastasis (LM), given its technical difficulty and the lack of typical symptoms. The existing gold standard of diagnosing LM is to use positive cerebrospinal fluid (CSF) cytology, which consumes significantly more time to classify cells under a microscope.ObjectiveThis study aims to establish a deep learning model to classify cancer cells in CSF, thus facilitating doctors to achieve an accurate and fast diagnosis of LM in an early stage.MethodThe cerebrospinal fluid laboratory of Xijing Hospital provides 53,255 cells from 90 LM patients in the research. We used two deep convolutional neural networks (CNN) models to classify cells in the CSF. A five-way cell classification model (CNN1) consists of lymphocytes, monocytes, neutrophils, erythrocytes, and cancer cells. A four-way cancer cell classification model (CNN2) consists of lung cancer cells, gastric cancer cells, breast cancer cells, and pancreatic cancer cells. Here, the CNN models were constructed by Resnet-inception-V2. We evaluated the performance of the proposed models on two external datasets and compared them with the results from 42 doctors of various levels of experience in the human-machine tests. Furthermore, we develop a computer-aided diagnosis (CAD) software to generate cytology diagnosis reports in the research rapidly.ResultsWith respect to the validation set, the mean average precision (mAP) of CNN1 is over 95% and that of CNN2 is close to 80%. Hence, the proposed deep learning model effectively classifies cells in CSF to facilitate the screening of cancer cells. In the human-machine tests, the accuracy of CNN1 is similar to the results from experts, with higher accuracy than doctors in other levels. Moreover, the overall accuracy of CNN2 is 10% higher than that of experts, with a time consumption of only one-third of that consumed by an expert. Using the CAD software saves 90% working time of cytologists.ConclusionA deep learning method has been developed to assist the LM diagnosis with high accuracy and low time consumption effectively. Thanks to labeled data and step-by-step training, our proposed method can successfully classify cancer cells in the CSF to assist LM diagnosis early. In addition, this unique research can predict cancer’s primary source of LM, which relies on cytomorphologic features without immunohistochemistry. Our results show that deep learning can be widely used in medical images to classify cerebrospinal fluid cells. For complex cancer classification tasks, the accuracy of the proposed method is significantly higher than that of specialist doctors, and its performance is better than that of junior doctors and interns. The application of CNNs and CAD software may ultimately aid in expediting the diagnosis and overcoming the shortage of experienced cytologists, thereby facilitating earlier treatment and improving the prognosis of LM. |
| first_indexed | 2024-12-20T18:29:28Z |
| format | Article |
| id | doaj.art-e7b0a8fca6b94cddb29f45c522794095 |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-12-20T18:29:28Z |
| publishDate | 2022-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-e7b0a8fca6b94cddb29f45c5227940952022-12-21T19:30:06ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2022-02-011210.3389/fonc.2022.821594821594Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal FluidWenjin Yu0Wenjin Yu1Wenjin Yu2Yangyang Liu3Yunsong Zhao4Haofan Huang5Jiahao Liu6Xiaofeng Yao7Jingwen Li8Zhen Xie9Luyue Jiang10Heping Wu11Xinhao Cao12Jiaming Zhou13Yuting Guo14Gaoyang Li15Matthew Xinhu Ren16Yi Quan17Tingmin Mu18Guillermo Ayuso Izquierdo19Guoxun Zhang20Guoxun Zhang21Runze Zhao22Di Zhao23Jiangyun Yan24Haijun Zhang25Junchao Lv26Qian Yao27Yan Duan28Huimin Zhou29Tingting Liu30Ying He31Ting Bian32Wen Dai33Jiahui Huai34Xiyuan Wang35Qian He36Yi Gao37Yi Gao38Yi Gao39Yi Gao40Wei Ren41Gang Niu42Gang Zhao43Gang Zhao44Department of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, ChinaThe College of Life Sciences and Medicine, Northwest University, Xi’an, ChinaElectronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering & The International Joint Laboratory for Micro/Nano Manufacturing and Measurement Technology, Xi’an Jiaotong University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaSchool of Biomedical Engineering, Health Science Center, Shenzhen University, Shenzhen, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, ChinaThe College of Medicine, Xiamen University, Xiamen, ChinaThe College of Life Sciences and Medicine, Northwest University, Xi’an, ChinaElectronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering & The International Joint Laboratory for Micro/Nano Manufacturing and Measurement Technology, Xi’an Jiaotong University, Xi’an, ChinaElectronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering & The International Joint Laboratory for Micro/Nano Manufacturing and Measurement Technology, Xi’an Jiaotong University, Xi’an, ChinaElectronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering & The International Joint Laboratory for Micro/Nano Manufacturing and Measurement Technology, Xi’an Jiaotong University, Xi’an, ChinaOphthalmology, Department of Clinical Science, Lund University, Lund, SwedenInstitute of Fluid Science, Tohoku University, Sendai, JapanInstitute of Fluid Science, Tohoku University, Sendai, JapanBiology Program, Faculty of Science, The University of British Columbia, Vancouver, BC, Canada0School of Microelectronics, Xidian University, Xi’an, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, China1Multiple Sclerosis Unit, Neurology Service, Vithas Nisa Hospital, Seville, SpainDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, China1Multiple Sclerosis Unit, Neurology Service, Vithas Nisa Hospital, Seville, Spain2Department of Ophthalmology, Eye Institute of PLA, Xijing Hospital, Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, China3Department of Neurology, Xiji Country People’s Hospital, Ningxia, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaThe College of Life Sciences and Medicine, Northwest University, Xi’an, ChinaThe College of Life Sciences and Medicine, Northwest University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, ChinaDepartment of Neurology, Yan’an University Medical College No. 3 Affiliated Hospital, Xianyang, ChinaSchool of Biomedical Engineering, Health Science Center, Shenzhen University, Shenzhen, China4Shenzhen Key Laboratory of Precision Medicine for Hematological Malignancies, Shenzhen, Guangzhou, China5Marshall Laboratory of Biomedical Engineering, Shenzhen, China6Peng Cheng Laboratory, Shenzhen, ChinaElectronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering & The International Joint Laboratory for Micro/Nano Manufacturing and Measurement Technology, Xi’an Jiaotong University, Xi’an, ChinaElectronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering & The International Joint Laboratory for Micro/Nano Manufacturing and Measurement Technology, Xi’an Jiaotong University, Xi’an, ChinaDepartment of Neurology, Xijing Hospital, the Fourth Military Medical University, Xi’an, ChinaThe College of Life Sciences and Medicine, Northwest University, Xi’an, ChinaBackgroundIt is a critical challenge to diagnose leptomeningeal metastasis (LM), given its technical difficulty and the lack of typical symptoms. The existing gold standard of diagnosing LM is to use positive cerebrospinal fluid (CSF) cytology, which consumes significantly more time to classify cells under a microscope.ObjectiveThis study aims to establish a deep learning model to classify cancer cells in CSF, thus facilitating doctors to achieve an accurate and fast diagnosis of LM in an early stage.MethodThe cerebrospinal fluid laboratory of Xijing Hospital provides 53,255 cells from 90 LM patients in the research. We used two deep convolutional neural networks (CNN) models to classify cells in the CSF. A five-way cell classification model (CNN1) consists of lymphocytes, monocytes, neutrophils, erythrocytes, and cancer cells. A four-way cancer cell classification model (CNN2) consists of lung cancer cells, gastric cancer cells, breast cancer cells, and pancreatic cancer cells. Here, the CNN models were constructed by Resnet-inception-V2. We evaluated the performance of the proposed models on two external datasets and compared them with the results from 42 doctors of various levels of experience in the human-machine tests. Furthermore, we develop a computer-aided diagnosis (CAD) software to generate cytology diagnosis reports in the research rapidly.ResultsWith respect to the validation set, the mean average precision (mAP) of CNN1 is over 95% and that of CNN2 is close to 80%. Hence, the proposed deep learning model effectively classifies cells in CSF to facilitate the screening of cancer cells. In the human-machine tests, the accuracy of CNN1 is similar to the results from experts, with higher accuracy than doctors in other levels. Moreover, the overall accuracy of CNN2 is 10% higher than that of experts, with a time consumption of only one-third of that consumed by an expert. Using the CAD software saves 90% working time of cytologists.ConclusionA deep learning method has been developed to assist the LM diagnosis with high accuracy and low time consumption effectively. Thanks to labeled data and step-by-step training, our proposed method can successfully classify cancer cells in the CSF to assist LM diagnosis early. In addition, this unique research can predict cancer’s primary source of LM, which relies on cytomorphologic features without immunohistochemistry. Our results show that deep learning can be widely used in medical images to classify cerebrospinal fluid cells. For complex cancer classification tasks, the accuracy of the proposed method is significantly higher than that of specialist doctors, and its performance is better than that of junior doctors and interns. The application of CNNs and CAD software may ultimately aid in expediting the diagnosis and overcoming the shortage of experienced cytologists, thereby facilitating earlier treatment and improving the prognosis of LM.https://www.frontiersin.org/articles/10.3389/fonc.2022.821594/fullleptomeningeal metastasis (LM)deep learningcytologyCSFcancer cell |
| spellingShingle | Wenjin Yu Wenjin Yu Wenjin Yu Yangyang Liu Yunsong Zhao Haofan Huang Jiahao Liu Xiaofeng Yao Jingwen Li Zhen Xie Luyue Jiang Heping Wu Xinhao Cao Jiaming Zhou Yuting Guo Gaoyang Li Matthew Xinhu Ren Yi Quan Tingmin Mu Guillermo Ayuso Izquierdo Guoxun Zhang Guoxun Zhang Runze Zhao Di Zhao Jiangyun Yan Haijun Zhang Junchao Lv Qian Yao Yan Duan Huimin Zhou Tingting Liu Ying He Ting Bian Wen Dai Jiahui Huai Xiyuan Wang Qian He Yi Gao Yi Gao Yi Gao Yi Gao Wei Ren Gang Niu Gang Zhao Gang Zhao Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid Frontiers in Oncology leptomeningeal metastasis (LM) deep learning cytology CSF cancer cell |
| title | Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid |
| title_full | Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid |
| title_fullStr | Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid |
| title_full_unstemmed | Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid |
| title_short | Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid |
| title_sort | deep learning based classification of cancer cell in leptomeningeal metastasis on cytomorphologic features of cerebrospinal fluid |
| topic | leptomeningeal metastasis (LM) deep learning cytology CSF cancer cell |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2022.821594/full |
| work_keys_str_mv | AT wenjinyu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT wenjinyu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT wenjinyu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yangyangliu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yunsongzhao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT haofanhuang deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT jiahaoliu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT xiaofengyao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT jingwenli deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT zhenxie deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT luyuejiang deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT hepingwu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT xinhaocao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT jiamingzhou deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yutingguo deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT gaoyangli deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT matthewxinhuren deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yiquan deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT tingminmu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT guillermoayusoizquierdo deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT guoxunzhang deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT guoxunzhang deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT runzezhao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT dizhao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT jiangyunyan deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT haijunzhang deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT junchaolv deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT qianyao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yanduan deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT huiminzhou deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT tingtingliu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yinghe deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT tingbian deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT wendai deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT jiahuihuai deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT xiyuanwang deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT qianhe deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yigao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yigao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yigao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT yigao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT weiren deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT gangniu deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT gangzhao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid AT gangzhao deeplearningbasedclassificationofcancercellinleptomeningealmetastasisoncytomorphologicfeaturesofcerebrospinalfluid |