Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 2 de 2
Add filters

Document Type
Year range
Comput Biol Med ; 144: 105340, 2022 05.
Article in English | MEDLINE | ID: covidwho-1734294


The outbreak of COVID-19 has caused a severe shortage of healthcare resources. Ground Glass Opacity (GGO) and consolidation of chest CT scans have been an essential basis for imaging diagnosis since 2020. The similarity of imaging features between COVID-19 and other pneumonia makes it challenging to distinguish between them and affects radiologists' diagnosis. Recently, deep learning in COVID-19 has been mainly divided into disease classification and lesion segmentation, yet little work has focused on the feature correlation between the two tasks. To address these issues, in this study, we propose MultiR-Net, a 3D deep learning model for combined COVID-19 classification and lesion segmentation, to achieve real-time and interpretable COVID-19 chest CT diagnosis. Precisely, the proposed network consists of two subnets: a multi-scale feature fusion UNet-like subnet for lesion segmentation and a classification subnet for disease diagnosis. The features between the two subnets are fused by the reverse attention mechanism and the iterable training strategy. Meanwhile, we proposed a loss function to enhance the interaction between the two subnets. Individual metrics can not wholly reflect network effectiveness. Thus we quantify the segmentation results with various evaluation metrics such as average surface distance, volume Dice, and test on the dataset. We employ a dataset containing 275 3D CT scans for classifying COVID-19, Community-acquired Pneumonia (CAP), and healthy people and segmented lesions in pneumonia patients. We split the dataset into 70% and 30% for training and testing. Extensive experiments showed that our multi-task model framework obtained an average recall of 93.323%, an average precision of 94.005% on the classification test set, and a 69.95% Volume Dice score on the segmentation test set of our dataset.

COVID-19 , Pneumonia , COVID-19/diagnostic imaging , Humans , Tomography, X-Ray Computed/methods
JMIR Med Inform ; 9(3): e24497, 2021 Mar 15.
Article in English | MEDLINE | ID: covidwho-1158312


BACKGROUND: Histology and Embryology and Pathology are two important basic medical morphology courses for studying human histological structures under healthy and pathological conditions, respectively. There is a natural succession between the two courses. At the beginning of 2020, the COVID-19 pandemic suddenly swept the world. During this unusual period, to ensure that medical students would understand and master basic medical knowledge and to lay a solid foundation for future medical bridge courses and professional courses, a web-based medical morphology teaching team, mainly including teachers of courses in Histology and Embryology and Pathology, was established. OBJECTIVE: This study aimed to explore a new teaching mode of Histology and Embryology and Pathology courses during the COVID-19 pandemic and to illustrate its feasibility and acceptability. METHODS: From March to July 2020, our team selected clinical medicine undergraduate students who started their studies in 2018 and 2019 as recipients of web-based teaching. Meanwhile, nursing undergraduate students who started their studies in 2019 and 2020 were selected for traditional offline teaching as the control group. For the web-based teaching, our team used the Xuexi Tong platform as the major platform to realize a new "seven-in-one" teaching method (ie, videos, materials, chapter tests, interactions, homework, live broadcasts, and case analysis/discussion). This new teaching mode involved diverse web-based teaching methods and contents, including flipped classroom, screen-to-screen experimental teaching, a drawing competition, and a writing activity on the theme of "What I Know About COVID-19." When the teaching was about to end, a questionnaire was administered to obtain feedback regarding the teaching performance. In the meantime, the final written pathology examination results of the web-based teaching and traditional offline teaching groups were compared to examine the mastery of knowledge of the students. RESULTS: Using the Xuexi Tong platform as the major platform to conduct "seven-in-one" teaching is feasible and acceptable. With regard to the teaching performance of this new web-based teaching mode, students demonstrated a high degree of satisfaction, and the questionnaire showed that 71.3% or more of the students in different groups reported a greater degree of satisfaction or being very satisfied. In fact, more students achieved high scores (90-100) in the web-based learning group than in the offline learning control group (P=.02). Especially, the number of students with objective scores >60 in the web-based learning group was greater than that in the offline learning control group (P=.045). CONCLUSIONS: This study showed that the web-based teaching mode was not inferior to the traditional offline teaching mode for medical morphology courses, proving the feasibility and acceptability of web-based teaching during the COVID-19 pandemic. Our findings lay a solid theoretical foundation for follow-up studies of medical students.