1
|
Shi L, Xiong CR, Liu MM, Wei XS, Wang XY, Wang T, Huang YX, Hong QB, Li W, Yang HT, Zhang JF, Yang K. [Establishment of a deep learning-based visual model for intelligent recognition of Oncomelania hupensis]. Zhongguo Xue Xi Chong Bing Fang Zhi Za Zhi 2021; 33:445-451. [PMID: 34791840 DOI: 10.16250/j.32.1374.2021033] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Subscribe] [Scholar Register] [Indexed: 11/27/2022]
Abstract
OBJECTIVE To establish a deep learning-based visual model for intelligent recognition of Oncomelania hupensis, the intermediate host of Schistosoma japonicum, and evaluate the effects of different training strategies for O. hupensis image recognition. METHODS A total of 2 614 datasets of O. hupensis snails and 4 similar snails were generated through field sampling and internet capture, and were divided into training sets and test sets. An intelligent recognition model was created based on deep learning, and was trained and tested. The precision, sensitivity, specificity, accuracy, F1 score and Youden index were calculated. In addition, the receiver operating characteristic (ROC) curve of the model for snail recognition was plotted to evaluate the effects of "new learning", "transfer learning" and "transfer learning + data enhancement" training strategies on the accuracy of the model for snail recognition. RESULTS Under the "transfer learning + data enhancement" strategy, the precision, sensitivity, specificity, accuracy, Youden index and F1 score of the model were 90.10%, 91.00%, 97.50%, 96.20%, 88.50% and 90.51% for snail recognition, which were all higher than those under both "new learning" and "transfer learning" strategies. There were significant differences in the sensitivity, specificity and accuracy of the model for snail recognition under "new learning", "transfer learning" and "transfer learning + data enhancement" training strategies (all P values < 0.001). In addition, the area under the ROC curve of the model was highest (0.94) under the "transfer learning + dataenhancement" training strategy. CONCLUSIONS This is the first visual model for intelligent recognition of O. hupensis based on deep learning, which shows a high accuracy for snail image recognition. The "transfer learning + data enhancement" training strategy is helpful to improve the accuracy of the model for snail recognition.
Collapse
Affiliation(s)
- L Shi
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - C R Xiong
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - M M Liu
- School of Public Health, Nanjing Medical University, China
| | - X S Wei
- School of Computer Science and Engineering, Nanjing University of Science and Technology, Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education, Jiangsu Provincial Key Laboratory of Image and Video Understanding for Social Safety, China
| | - X Y Wang
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - T Wang
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - Y X Huang
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - Q B Hong
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - W Li
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - H T Yang
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - J F Zhang
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China
| | - K Yang
- Key Laboratory of National Health Commission on Parasitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Parasite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases, Public Health Research Center of Jiangnan University, Wuxi 214064, China.,School of Public Health, Nanjing Medical University, China
| |
Collapse
|
2
|
Peng WB, Sun X, Zhou M, Wei XS, Wu XZ, Ruan WW, Shi HZ, Lan XL, Zhou Q. [Value of PET/MRI in the diagnosis of malignant pleural effusion in comparison with PET/CT]. Zhonghua Yi Xue Za Zhi 2021; 101:2363-2369. [PMID: 34404128 DOI: 10.3760/cma.j.cn112137-20210516-01142] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
Abstract
Objective: To investigate the diagnostic value of PET/MRI for malignant pleural effusion (MPE), and compare its diagnostic difference with PET/CT. Methods: The data of 57 patients with suspected MPE admitted into Union Hospital of Tongji Medical College of Huazhong University of Science and Technology from October 2017 to January 2020 was analyzed. A total of 53 patients were included in the prospective study, and the whole body PET/CT and thoracic PET/MRI were performed on them respectively. Two physicians used a blind method to evaluate the morphological features of PET/CT and PET/MRI images, delineate the region of interest (ROI), obtain the maximum standard uptake value (SUVmax) of the ROI in the PET/CT and PET/MRI images. The target-to-background ratio (TBR) of the lesion was calculated. The diffusion-weighted imaging (DWI) characteristics of the pleura in PET/MRI images were analyzed. Taking pathological diagnosis as the gold standard, the diagnostic effect of PET/CT and PET/MRI on MPE were evaluated. Results: The 53 patients who were finally included were (62.8±1.7) years old, consisting of 31 males. Pathological results showed that 41 cases were MPE and 12 cases were benign pleural effusion (BPE). There were no statistical differences in age, gender and smoking history between the two groups (P>0.05). Bland-Altman analysis showed that the SUVmax of pleural lesions by PET/MRI was higher than that by PET/CT (6.4±0.6 vs 5.3±0.5, P<0.001). The TBR of PET/MRI was higher than that of PET/CT (2.2±0.2 vs 1.8±0.2, P<0.001). The sensitivity, specificity, and accuracy of PET/MRI in the diagnosis of MPE by combining imaging features such as SUVmax and DWI of pleural lesions were 75.6%, 100%, and 81.1%, respectively. The sensitivity, specificity, and accuracy of PET/CT combined with SUVmax and imaging features of pleural lesions in the diagnosis of MPE were 85.4%, 83.3%, and 77.4%, respectively. There was no statistically significant difference between PET/MRI and PET/CT in the area under the curve (AUC) for diagnosing MPE (0.934 vs 0.873, P>0.05). Conclusions: PET/MRI and PET/CT have the equivalent diagnostic efficiency for MPE. However, PET/MRI shows higher SUVmax and TBR for pleural lesions, and has specific pleural DWI imaging characteristics, which is worthy of further clinical research.
Collapse
Affiliation(s)
- W B Peng
- Department of Respiratory and Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China
| | - X Sun
- Department of Nuclear Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology/Hubei Province Key Laboratory of Molecular Imaging, Wuhan 430022, China
| | - M Zhou
- Department of Respiratory and Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China
| | - X S Wei
- Department of Respiratory and Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China
| | - X Z Wu
- Department of Respiratory and Critical Care Medicine, Beijing Institute of Respiratory Medicine and Beijing Chaoyang Hospital, Capital Medical University, Beijing 100020, China
| | - W W Ruan
- Department of Nuclear Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology/Hubei Province Key Laboratory of Molecular Imaging, Wuhan 430022, China
| | - H Z Shi
- Department of Respiratory and Critical Care Medicine, Beijing Institute of Respiratory Medicine and Beijing Chaoyang Hospital, Capital Medical University, Beijing 100020, China
| | - X L Lan
- Department of Nuclear Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology/Hubei Province Key Laboratory of Molecular Imaging, Wuhan 430022, China
| | - Q Zhou
- Department of Respiratory and Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China
| |
Collapse
|