Diagnostic performance for pulmonary adenocarcinoma on CT: comparison of radiologists with and without three-dimensional convolutional neural network

Yanagawa, Masahiro; Niioka, Hirohiko; Kusumoto, Masahiko; Awai, Kazuo; Tsubamoto, Mitsuko; Satoh, Yukihisa; Miyata, Tomo; Yoshida, Yuki; Kikuchi, Noriko; Hata, Akinori; Yamasaki, Shohei; Shoji, K.; Nagahara, Hajime; Miyake, Jun; Tomiyama, Noriyuki

doi:10.1007/s00330-020-07339-x

Cited by 21 publications

(16 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CNN model can monitor the mental state in the construction of an educational network. The latest research of Yanagawa et al (2021) on the performance of the CNN model shows that CNN has a superior prediction ability ( Yanagawa et al, 2021 ). Figure 2 is based on the mental state early warning module established by CNN, which shows that the mental health education network system is feasible.…”

Section: Discussionmentioning

confidence: 99%

Exploration and Strategy Analysis of Mental Health Education for Students in Sports Majors in the Era of Artificial Intelligence

Liang

Yong

Ge³

et al. 2022

Front. Psychol.

View full text Add to dashboard Cite

This study aims to explore new educational strategies suitable for the mental health education of college students. Big data and artificial intelligence (AI) are combined to evaluate the mental health education of college students in sports majors. First, the research status on the mental health education of college students is introduced. The internet of things (IoT) on mental health education, a structure based on big data and convolutional neural network (CNN), is constructed. Next, the survey design and questionnaire survey are carried out. Finally, the questionnaire data are analyzed and compared with the mental health status under traditional education. The results show that the CNN model has good accuracy and ability to distinguish symptoms, so it can be applied to the existing psychological work in colleges. In the symptom comparison survey, under the traditional education and big data network, the number of college students with mild mental health problems is found to be 158 (84.9%) and 170 (91.4%), respectively. It indicates that the number of college students with moderate mental health problems decreases significantly. In the comparative investigation of the severity of mental problems, the number of students with normal mental health, subhealth, and serious mental health problems under the background of traditional mental health education is 125 (67.2%), 56 (30.1%), and 5 (2.7%), respectively. The mental health status of college students under the influence of big data networks on mental health education is better than that of traditional mental health education. There are 140 students with normal mental health, a year-on-year increase of 16.7%. In the comparative survey of specific mental disorders, students with obsessive-compulsive symptoms under traditional mental health education account for 22.0% of the total sample, having the largest proportion. In the subhealth psychological group under the big data network on mental health education, the number of hostile students decreases by 7, which is the psychological factor with the most obvious improvement. Hence, the proposed path of mental health education is feasible.

show abstract

Section: Discussionmentioning

confidence: 99%

Exploration and Strategy Analysis of Mental Health Education for Students in Sports Majors in the Era of Artificial Intelligence

Liang

Yong

Ge³

et al. 2022

Front. Psychol.

View full text Add to dashboard Cite

show abstract

“…3 According to the suggestion of Fleischner Society, 34 take the center of the window level range on lung window, cover it with the maximum window width, and refer to the CT value of the air, the CT value of image is limited to À1000 to 400 HU to eliminate the interference of irrelevant components. 4 Perform min-max normalization to map the data to [À1, 1]. 5 Determine the real resolution label of the sample according to the diameter of the nodule.…”

Section: Data Preprocessingmentioning

confidence: 99%

3D multi‐resolution attention capsule network for diagnosing multi‐pathological types of pulmonary nodules

Xiao

Gao

et al. 2022

Int J Imaging Syst Tech

View full text Add to dashboard Cite

To achieve effective diagnosis of multi‐pathological types on pulmonary nodules, a three‐dimensional (3D) multi‐resolution attention (MRA) capsule network (3D MRA‐CapsNet‐DL) was proposed by using a self‐constructed lung CT image dataset. In the construction of 3D MRA‐CapsNet‐DL, an improved dynamic routing algorithm (DRA) with limiting the update increment (called DL) was first designed to solve the inactivation problem of a large number of vector neurons. Then, MRA was used to build three kinds of multi‐path 3D MRA‐CapsNet‐DL for multi‐pathological classification. Besides, we also investigated whether the interpolation perturbation introduced by multi‐resolution methods will affect the classification performance. Experiments showed that the proposed 3D MRA‐CapsNet‐DL can achieve a better classification (79.27% ACC and 0.9056 AUC) effect than classical convolutional neural networks (CNNs), and the interpolation disturbance has favorable and unfavorable effects on the classification performance of multi‐path and single‐path models, respectively.

show abstract

“…Another critical machine learning method is radiomics, which refers to the manual extraction and analysis of advanced quantitative imaging features and provides predictive models relating image features to tumor phenotypes [ 18 , 19 ]. In addition, previous studies have demonstrated that deep learning or radiomics system could distinguish AAH-AIS, MIA and invasive adenocarcinoma (IAC), and achieved better classification performance than the radiologists [20] , [21] , [22] . However, there was very little literature to predict the specific pathological subtype categories of lung adenocarcinoma using deep learning or radiomics methods and further to help with the precision of survival estimations.…”

Section: Introductionmentioning

confidence: 99%

Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography

Wang

Shao

et al. 2021

Translational Oncology

View full text Add to dashboard Cite

Highlights A CT imaging dataset of 1222 lung adenocarcinoma patients from 3 medical centers was used to construct the classification model. The proposed algorithms achieved encouraging performance in the differentiation of lung adenocarcinoma subtypes. The C-index of prognosis model based on deep radiomics combined classifier was 0.892(95% confidence Intervals: 0.846–0.937) in internal validation set.

show abstract

Diagnostic performance for pulmonary adenocarcinoma on CT: comparison of radiologists with and without three-dimensional convolutional neural network

Cited by 21 publications

References 30 publications

Exploration and Strategy Analysis of Mental Health Education for Students in Sports Majors in the Era of Artificial Intelligence

Exploration and Strategy Analysis of Mental Health Education for Students in Sports Majors in the Era of Artificial Intelligence

3D multi‐resolution attention capsule network for diagnosing multi‐pathological types of pulmonary nodules

Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography

Contact Info

Product

Resources

About