|Table of Contents|

Study on pathological diagnosis model of intraoperative frozen sections of epithelial ovarian cancer based on deep transfer learning

Journal Of Modern Oncology[ISSN:1672-4992/CN:61-1415/R]

Issue:
2024 22
Page:
4250-4254
Research Field:
Publishing date:

Info

Title:
Study on pathological diagnosis model of intraoperative frozen sections of epithelial ovarian cancer based on deep transfer learning
Author(s):
FU Xiaojuan1; ZHAO Ying1; ZHOU Zhihao2; LI Xiaorong1; HOU Mengchen1; CAI Fengmei1; WANG Huifang1
1.Department of Pathology,Xi'an People's Hospital (The Fourth Hospital of Xi'an),Shaanxi Xi'an 710004,China;2.School of Information Science and Technology,Northwest University,Shaanxi Xi'an 710127,China.
Keywords:
epithelial ovarian cancer; intraoperative frozen section; artificial intelligence; transfer learning; pathological diagnosis
PACS:
R737.31
DOI:
10.3969/j.issn.1672-4992.2024.22.006
Abstract:
Objective:To explore the pathological diagnostic model for intraoperative frozen sections of epithelial ovarian cancer using deep transfer learning and to assess its practical value.Methods:From January 2021 to December 2022,25 intraoperative frozen sections and 121 postoperative paraffin sections of ovarian epithelial cancer from department of pathology of our hospital were collected.These included five histopathological types (serous carcinoma,mucinous carcinoma,endometrioid carcinoma,clear cell carcinoma,and metastatic signet ring cell carcinoma),which were scanned into whole-slide digital pathology images.Image block datasets were extracted from these sections,with those outside the annotation frames categorized as "Other", resulting in six distinct image block datasets.The datasets were split into training,testing,and validation sets in a 3∶1∶1 ratio respectively.A model from scratch was trained using the frozen section dataset,while a pretrained model was trained using the paraffin section dataset.The pretrained model was further optimized on the frozen section dataset through transfer learning to develop a deep transfer learning model.The performance of the trained six-category datasets was compared in corresponding test sets with and without the application of the transfer learning model.Results:Compared to the model trained from scratch,the transfer learning model presented a significant improvement across all metrics,achieving a prediction accuracy of 90% for all six classes of image blocks.Conclusion:The transfer learning model exhibits a high level of accuracy and stability in the classification of pathological images from intraoperative frozen sections of epithelial ovarian cancer,indicating good potential for wider implementation.

References:

［1］ SIEGEL RL,GIAQUINTO AN,JEMAL A.Cancer statistics 2024［J］.CA Cancer J Clin,2024,74(1):12-49.
［2］ LAN A,YANG G.Clinicopathological parameters and survival of invasive epithelial ovarian cancer by histotype and disease stage［J］.Future Oncol,2019,15(17):2029-2039.
［3］ SHIN HC,ROTH HR,GAO M,et al.Deep convolutional neural networks for computer-aided detection:cnn architectures,dataset characteristics and transfer learning［J］.IEEE Trans Med Imaging,2016,35(5):1285-1298.
［4］ JAHN SW,PLASS M,MOINFAR F.Digital pathology:advantages,limitations and emerging perspectives［J］.J Clin Med,2020,9(11):3697.
［5］ ZHOU Y,GAO M,FANG D,et al.Unsupervised background-constrained tank segmentation of infrared images in complex background based on the otsu method［J］.Springerplus,2016,5(1):1409.
［6］ QIN F,GAO N,PENG Y,et al.Fine-grained leukocyte classification with deep residual learning for microscopic images［J］.Comput Methods Programs Biomed,2018,162:243-252.
［7］ LIN Y,DING W,QIANG S,et al.ES-Imagenet:a million event-stream classification dataset for spiking neural networks［J］.Front Neurosci,2021,15:726582.
［8］ SHIN J,CHUNG W.Multi-band cnn with band-dependent kernels and amalgamated cross entropy loss for motor imagery classification［J］.IEEE J Biomed Health Inform,2023,27(9):4466-4477.
［9］ CHIANG PJ.Adaptive penalty method with adam optimizer for enhanced convergence in optical waveguide mode solvers［J］.Opt Express,2023,31(17):28065-28077.
［10］ NOMURA N,SUGIIYAMA T,TAJIRI T,et al.Accuracy of imprint cytology and frozen section histology for intraoperative diagnosis of ovarian epithelial tumors:a comparative study and proposed algorithm［J］.Diagn Cytopathol,2021,49(6):682-690.
［11］ KUNG FY,TSANG AK,YU EL.Intraoperative frozen section analysis of ovarian tumors:a 11-year review of accuracy with clinicopathological correlation in a Hong Kong regional hospital［J］.Int J Gynecol Cancer,2019,29(4):772-778.
［12］ CHAN HP,SAMALA RK,HADJIISKI LM,et al.Deep learning in medical image analysis［J］.Adv Exp Med Biol,2020,1213:3-21.
［13］ KHOSRAVI P,KAZEMI E,IMIELINSKI M,et al.Deep convolutional neural networks enable discrimination of heterogeneous digital pathology images［J］.EBioMedicine,2018,27:317-328.
［14］ ABDOLAHI M,SALEHI M,SHOKATIAN I,et al.Artificial intelligence in automatic classification of invasive ductal carcinoma breast cancer in digital pathology images［J］.Med J Islam Repub Iran,2020,34:140.
［15］ ORSULIC S,JOHN J,WALTS AE,et al.Computational pathology in ovarian cancer［J］.Front Oncol,2022,12:924945.
［16］ WU M,YAN C,LIU H,et al.Automatic classification of ovarian cancer types from cytological images using deep convolutional neural networks［J］.Biosci Rep,2018,38(3):180289.
［17］ GUO Y,ZHANG J,SUN B,et al.Adversarial deep transfer learning in fault diagnosis:progress,challenges,and future prospects［J］.Sensors (Basel),2023,23(16):7263.
［18］ HOSSEINZADEH KS,HOSSEINZADEH KP,WESOLOWSKI MJ,et al.Deep transfer learning based model for colorectal cancer histopathology segmentation:a comparative study of deep pre-trained models［J］.Int J Med Inform,2022,159:104669.
［19］ 严丹丹,饶洁,尹修恒,等.基于深度迁移学习的甲状腺癌术中冷冻切片病理诊断模型研究［J］.临床与实验病理学杂志,2023,39(12):1448-1452. YAN DD,RAO J,YIN XH,et al.Pathological diagnosis of thyroid cancer histopathological image from intraoperative frozen sections based on deep transfer learning ［J］ .J Clin Exp Pathol,2023,39(12):1448-1452.
［20］ 刘东,薛祺,宋刚,等.人工智能在听神经瘤诊疗中应用的研究进展［J］.现代肿瘤医学，2024,32(01):146-150. LIU D,XUE Q,SONG G,et al.Research progress of artificial intelligence in the diagnosis and treatment of acoustic neuroma［J］.Modern Oncology,2024,32(01):146-150.
［21］ 董国辉,高波,潘云.人工智能在鼻咽癌诊断与预后领域的研究进展［J］.现代肿瘤医学，2023,31(09):1765-1768. DONG HJ.GAO B,PAN Y.Research progress of artificial intelligence in diagnosis and prognosis of nasopharyngeal carcinoma［J］.Modern Oncology,2023,31(09):1765-1768.
［22］ MARTIN-GINZALEZ P,CRISPIN-ORTUZAR M,RUNDO L,et al.Integrative radiogenomics for virtual biopsy and treatment monitoring in ovarian cancer［J］.Insights Imaging,2020,11(1):94.

Memo

Memo:
陕西省2022年重点研发计划项目（编号：2022SF-504）
Last Update: 1900-01-01