Recently, deep learning has achieved impressive results in medical image tasks. However, this method usually requires large-scale annotated data, and medical images are expensive to annotate, so it is a challenge to learn efficiently from the limited annotated data. Currently, the two commonly used methods are transfer learning and self-supervised learning. However, these two methods have been little studied in multimodal medical images, so this study proposes a contrastive learning method for multimodal medical images. The method takes images of different modalities of the same patient as positive samples, which effectively increases the number of positive samples in the training process and helps the model to fully learn the similarities and differences of lesions on images of different modalities, thus improving the model's understanding of medical images and diagnostic accuracy. The commonly used data augmentation methods are not suitable for multimodal images, so this paper proposes a domain adaptive denormalization method to transform the source domain images with the help of statistical information of the target domain. In this study, the method is validated with two different multimodal medical image classification tasks: in the microvascular infiltration recognition task, the method achieves an accuracy of (74.79 ± 0.74)% and an F1 score of (78.37 ± 1.94)%, which are improved as compared with other conventional learning methods; for the brain tumor pathology grading task, the method also achieves significant improvements. The results show that the method achieves good results on multimodal medical images and can provide a reference solution for pre-training multimodal medical images.
Traditional Chinese medicine equipment plays an indispensable role in the prevention, diagnosis, treatment and rehabilitation of traditional Chinese medicine from the needs of people's life and health, and provides technical support for the simple, convenient, cheap and effective clinical practice of traditional Chinese medicine. The traditional Chinese medicine equipment industry has the development advantages of large demand gap, strong policy support and emerging technology empowerment. At the same time, there are also bottlenecks such as lagging standardization construction, weak industrial foundation, insufficient characteristics of traditional Chinese medicine and immature evidence-based evaluation research. The coming of the era of digital intelligence has brought new opportunities for the development and reform of the traditional Chinese medicine equipment industry. This paper provides development ideas for the transformation of traditional Chinese medicine equipment from traditional to modern from the aspects of standardization construction, digital intelligence industry upgrading, improvement of evidence-based evaluation system and in-depth international exchanges and cooperation.
ObjectiveTo report the short-term outcomes of a standardized, simplified and reproducible strategy of mitral valvuloplasty (MVP), which was focused on leaflet foldoplasty and anatomic anomalies of congenital mitral regurgitation (MR).MethodsConsecutive 74 patients who underwent MVP by our standardized strategy in our institution from 2016 to 2018 were included retrospectively. There were 30 males and 44 females with a median age of 18.5 (6-146) months and weight of 15.4 (7-51) kg.ResultsAnatomic anomalies of MR included: (1) subvalvular apparatus: 72 (97.3%) patients with mal-connected chordae tendineae, 31 (41.9%) with absent chordae tendineae and 14 (18.9%) with fused or dysplastic papillary muscle; (2) leaflet: 10 (13.5%) patients with cleft of anterior leaflet, 61 (82.4%) with leaflet prolapse including 56 (91.8%) with anterior leaflet prolapse; (3) annulus: 71 (95.9%) patients with annular dilatation. Leaflet foldoplasty was performed in 61 (82.4%) patients with leaflet prolapse. All patients were successfully discharged and 4 (5.4%) patients were with moderate MR. The follow-up time was 22.0 (9.1-41.8) months. During the follow-up period, 3 patients had moderate MR and 1 patient had reoperation for severe MR. All patients were in normal cardiac function with a mean left ventricular ejection fraction of 66.0%±6.1%. In addition, the mean left ventricular end-diastolic dimension was 31.8±6.0 mm, which was significant smaller than that before the operation (t=6.090, P<0.000 1).ConclusionThe standardized leaflet foldoplasty with resection of mal-connected chordae tendineae and posterior annuloplasty technique is safe and feasible with favorable short-term outcomes in MR patients.
Traditional classifiers, such as support vector machine and Bayesian classifier, require data normalization for removing experimental batch effects, which limit their applications at the individual level. In this paper, we aim to build a classifier to distinguish lung cancer and non-cancer lung tissues (pneumonia and normal lung tissues). We identified gene pairs as signatures to build a classifier based on the within-sample relative expression orderings of gene pairs in a particular type of tissues (cancer or non-cancer). Using multiple independent datasets as the training data, including a total of 197 lung cancer cases and 189 non-cancer cases, we identified three gene pairs. Classifying a sample by the majority voting rule, the average accuracy reached 95.34% in the training data. Using multiple independent validation datasets, including a total of 251 lung cancer samples and 141 non-cancer samples without data normalization, the average accuracy was as high as 96.78%. The rank-based signature is robust against experimental batch effects and can be used to diagnose lung cancer using samples measured by different laboratories at the individual level.
ObjectiveTo analyze the relationship between maximum standardized uptake value (SUVmax) of primary tumor detected by 18F-FDG positron emission tomography/computed tomography (PET/CT) and clinicopathologic factors in stageⅠnon-small cell lung cancer (NSCLC), and investigate the prognostic value of PET/CT on pathological feature. MethodsWe retrospectively analyzed clinical data of 182 patients with stageⅠNSCLC who underwent 18F-FDG PET/CT scan before lobectomy or segmentectomy in China-Japan Friendship Hospital from April 2013 to June 2014. There were 121 male and 61 female patients with their ages of 34-85 (68.1±9.8) years. Clinicopathologic factors including sex, age, smoking history, histology, TNM stage, T stage, tumor size, lymphatic vessel invasion, blood vessel invasion (BVI) and visceral pleural invasion were evaluated to identify the independent factors affecting SUVmax by univariate and multivariate regression analysis. The diagnostic efficiency and best cut-off point of SUVmax were calculated by the receiver operating characteristic curve. ResultsThe univariate analysis identified that sex (P=0.015), smoking history (P=0.001), histology (P < 0.001), TNM stage (P=0.004), T stage (P=0.001), tumor size (P < 0.001), BVI (P=0.001) were factors affecting SUVmax. Only histology (P=0.001), tumor size (P=0.006), BVI (P=0.009) were found to be significant independent factors according to multivariate regression analysis. The SUVmax of primary tumor was a predictor for BVI with the highest diagnostic accuracy at a cut-off value of 4.85, the sensitivity and specificity were 65.5% and 71.7%. ConclusionThe SUVmax is correlated with histology, tumor size and BVI in stageⅠNSCLC, higher in patients with non-adenocarcinoma, lager tumor and positive BVI. Furthermore, the probability of BVI could be predicted by SUVmax of the primary tumor.
ObjectiveTo explore the feasibility of introducing student-standardized patients in the teaching reform of medical nursing course. MethodsWe chose four classes of nursing students from grade 2012 between September and December 2014 as the research subjects.Cluster sampling was used to choose two classes of 84 nursing students randomly as trial group, who received student-standardized patients in their practical learning; while the rest 2 classes of 83 students were chosen as control group, who received traditional teaching method.The course scores and the effect evaluation were compared between the two groups. ResultsThe basic knowledge test score of the trial group 31.28±4.81 was not significantly different from that of the control group 32.10±2.15(P > 0.05).The case analysis test score of the trial group 54.36±3.45 was significantly higher than that of the control group 43.12±1.37(P < 0.05).The communication ability, health education ability, skill operation ability and professional quality score of nursing students in the trial group were also significantly higher than those in the control group (P < 0.05). ConclusionIntroducing student-standardized patients in practical teaching of medical nursing can improve the teaching effect and students' comprehensive ability.
ObjectiveTo develop a standardized dataset for adverse drug reactions (ADR) of Chinese herbal formula granules (CHFG) to regulate the collection content of ADR, promote the standardization and normalization of ADR data collection for CHFG, and facilitate the sharing, integration, and analysis of adverse reaction data. MethodsWe used a combination of literature research, Delphi survey and consensus meeting. ResultsA Delphi survey questionnaire was constructed based on the results of literature research, including 6 domains and 76 items. After the Delphi survey and consensus meeting, a final CHFG adverse reaction dataset was developed, including 6 domains and 75 items. The six domains were patient details, suspected drugs, other treatments/concomitant medications, detailed information on the suspected adverse reaction, possible influencing factors (causes of the suspected adverse reaction), and details of the person reporting the suspected adverse reaction. Compared with the data collected by the National Adverse Drug Reaction Monitoring Center, this dataset introduced a new domain called "Possible influencing factors", which included several items such as irrational use of CHFG, toxic varieties of Chinese herbal medicine, storage and usage conditions, physical characteristics, processing methods, and patient diet. It also contained the information on Chinese medicine syndromes and other herbs in the prescription, and modified multiple items based on the particularities of formula granules. ConclusionThe development and application of this standardized dataset of ADR for CHFG can facilitate data collection, integration, and analysis, furthermore improve doctors' awareness of prescribing safely and enhance patient medication safety.