Abstract
The frozen section (FS) diagnoses of pathology experts are used in China to determine whether sentinel lymph nodes of breast cancer have metastasis during operation. Direct implementation of ...a deep neural network (DNN) in clinical practice may be hindered by misdiagnosis of the algorithm, which affects a patient's treatment decision. In this study, we first obtained the prediction result of the commonly used patch-DNN, then we present a relative risk classification and regression tree (RRCART) to identify the misdiagnosed whole-slide images (WSIs) and recommend them to be reviewed by pathologists. Applying this framework to 2362 WSIs of breast cancer lymph node metastasis, test on frozen section results in the mean area under the curve (AUC) reached 0.9851. However, the mean misdiagnosis rate (0.0248), was significantly higher than the pathologists’ misdiagnosis rate (
p
< 0.01). The RRCART distinguished more than 80% of the WSIs as a high-accuracy group with an average accuracy reached to 0.995, but the difference with the pathologists’ performance was not significant (
p
> 0.01). However, the other low-accuracy group included most of the misdiagnoses of DNN models. Our research shows that the misdiagnosis from deep learning model can be further enriched by our method, and that the low-accuracy WSIs must be selected for pathologists to review and the high-accuracy ones may be ready for pathologists to give diagnostic reports.
Breast cancer is the most common malignant tumor in the world. Intraoperative frozen section of sentinel lymph nodes is an important basis for determining whether axillary lymph node dissection is ...required for breast cancer surgery. We propose an RRCART model based on a deep-learning network to identify metastases in 2362 frozen sections and count the wrongly identified sections and the associated reasons. The purpose is to summarize the factors that affect the accuracy of the artificial intelligence model and propose corresponding solutions.
We took the pathological diagnosis of senior pathologists as the gold standard and identified errors. The pathologists and artificial intelligence engineers jointly read the images and heatmaps to determine the locations of the identified errors on sections, and the pathologists found the reasons (false reasons) for the errors. Through NVivo 12 Plus, qualitative analysis of word frequency analysis and nodal analysis was performed on the error reasons, and the top-down error reason framework of "artificial intelligence RRCART model to identify frozen sections of breast cancer lymph nodes" was constructed based on the importance of false reasons.
There were 101 incorrectly identified sections in 2362 slides, including 42 false negatives and 59 false positives. Through NVivo 12 Plus software, the error causes were node-coded, and finally, 2 parent nodes (high-frequency error, low-frequency error) and 5 child nodes (section quality, normal lymph node structure, secondary reaction of lymph nodes, micrometastasis, and special growth pattern of tumor) were obtained; among them, the error of highest frequency was that caused by normal lymph node structure, with a total of 45 cases (44.55%), followed by micrometastasis, which occurred in 30 cases (29.70%).
The causes of identification errors in examination of sentinel lymph node frozen sections by artificial intelligence are, in descending order of influence, normal lymph node structure, micrometastases, section quality, special tumor growth patterns and secondary lymph node reactions. In this study, by constructing an artificial intelligence model to identify the error causes of frozen sections of lymph nodes in breast cancer and by analyzing the model in detail, we found that poor quality of slices was the preproblem of many identification errors, which can lead to other errors, such as unclear recognition of lymph node structure by computer. Therefore, we believe that the process of artificial intelligence pathological diagnosis should be optimized, and the quality control of the pathological sections included in the artificial intelligence reading should be carried out first to exclude the influence of poor section quality on the computer model. For cases of micrometastasis, we suggest that by differentiating slices into high- and low-confidence groups, low-confidence micrometastatic slices can be separated for manual identification. The normal lymph node structure can be improved by adding samples and training the model in a targeted manner.
Both anxiety and depression in family caregivers (FCs) of advanced cancer patients are common, and they have a negative influence on both the FCs and the patients. Some studies suggested that a ...variety of interventions could alleviate the psychological symptoms of FCs. However, there is no consensus on much more effective methods for intervention, and relatively high-quality research is blank in psychological problems of these population in China. The validity of mindfulness-based stress reduction (MBSR) and psychological consultation guided by the needs assessment tool (NST) in the psychological status of caregivers will be compared in this study to select a more suitable intervention for the FCs of advanced cancer patients in China.
A randomized N-of-1 trial would be conducted at the Cancer Hospital, Chinese Academy of Medical Sciences. Fifty eligible FCs of advanced cancer patients will be recruited, and all will receive three cycles of psychological intervention treatment, with each cycle including both of MBSR and psychological consultation guided by the NST. MBSR and psychological consultation guided by the NST will be compared with each other in each cycle, and the intervention sequence will be based on the random number table generated after the informed consent has been completed. Each treatment period is 2 weeks, and the interval between different treatment cycles or treatment periods is 1 week. The self-reported scales are measured at the beginning and end of each treatment period, including the Self-Rating Anxiety Scale (SAS), the Self-Rating Depression Scale (SDS), Distress Thermometer (DT), Zarit Burden Interview (ZBI), Chinese version of the Medical Outcomes Study 12-item Short Form (C-SF-12), and Family Carer Satisfaction with Palliative Care scale (FAMCARE-2).
The protocol of the study was approved by the Institutional Review Board of the Ethical Committee of the Cancer Hospital, Chinese Academic of Medical Science. The results will be published in a peer-reviewed medical journal. The study is registered at Chinese Clinical Trials Registry with the trial registration number chiCTR2000033707. This study employs an innovative methodological approach on the effectiveness of MBSR and psychological consultation guided by the NST for psychological status of FCs of advanced cancer patients. The findings of the study will be helpful to provide high-quality evidence-based medical data for psychological intervention of FCs of advanced cancer patients, and guide clinicians on best quality treatment recommendations.
Display omitted
•A novel information model is designed to represent the whole-course patient information of breast cancer.•The first study to extract comprehensive clinical information from multiple ...types of breast cancer notes in Chinese.•Extraction models generated by fine-tuning BERT outperform current state-of-the-art methods on breast cancer notes.
Breast cancer is the most common malignant tumor among women. The diagnosis and treatment information of breast cancer patients is abundant in multiple types of clinical fields, including clinicopathological data, genotype and phenotype information, treatment information, and prognosis information. However, current studies are mainly focused on extracting information from one specific type of clinical field. This study defines a comprehensive information model to represent the whole-course clinical information of patients. Furthermore, deep learning approaches are used to extract the concepts and their attributes from clinical breast cancer documents by fine-tuning pretrained Bidirectional Encoder Representations from Transformers (BERT) language models.
The clinical corpus that was used in this study was from one 3A cancer hospital in China, consisting of the encounter notes, operation records, pathology notes, radiology notes, progress notes and discharge summaries of 100 breast cancer patients. Our system consists of two components: a named entity recognition (NER) component and a relation recognition component. For each component, we implemented deep learning-based approaches by fine-tuning BERT, which outperformed other state-of-the-art methods on multiple natural language processing (NLP) tasks. A clinical language model is first pretrained using BERT on a large-scale unlabeled corpus of Chinese clinical text. For NER, the context embeddings that were pretrained using BERT were used as the input features of the Bi-LSTM-CRF (Bidirectional long-short-memory-conditional random fields) model and were fine-tuned using the annotated breast cancer notes. Furthermore, we proposed an approach to fine-tune BERT for relation extraction. It was considered to be a classification problem in which the two entities that were mentioned in the input sentence were replaced with their semantic types.
Our best-performing system achieved F1 scores of 93.53% for the NER and 96.73% for the relation extraction. Additional evaluations showed that the deep learning-based approaches that fine-tuned BERT did outperform the traditional Bi-LSTM-CRF and CRF machine learning algorithms in NER and the attention-Bi-LSTM and SVM (support vector machines) algorithms in relation recognition.
In this study, we developed a deep learning approach that fine-tuned BERT to extract the breast cancer concepts and their attributes. It demonstrated its superior performance compared to traditional machine learning algorithms, thus supporting its uses in broader NER and relation extraction tasks in the medical domain.
Triaxial cyclic loading tests were conducted to examine the damage and failure of gas-bearing coal under stepped cyclic loading by considering the frequency and amplitude of the cyclic load. The ...stress-strain curves were obtained, and the effects of load frequency and amplitude on parameters such as strength, dynamic tangent modulus, and acoustic emission characteristics were analyzed. The stress-strain curve exhibits stepped hysteresis loops, which can be divided into the compaction stage, the linear elastic stage, and the yield failure stage. As the frequency of cyclic loading increases from 2 to 5 Hz, the strength of gas-bearing coal increases from 32.55 to 40.36 MPa, the maximum dynamic tangent modulus and peak AE counts increase, while the cumulative AE counts decrease. When the amplitude of cyclic loading is increased from 2 to 5 MPa, the strength of gas-bearing coal decreases from 37.94 to 31 MPa, which has negligible effect on the maximum dynamic tangent modulus. The cumulative AE counts increase and the maximum AE counts decrease. Finally, through theoretical derivation and fitting of experimental results, a constitutive model for coal damage and deterioration under the combined effects of gas adsorption and stepped cyclic loading was established. This study provides an important reference for the reasonable selection of mining parameters for gas-bearing coal mines.
Research evidence continues to reveal findings important for health professionals' clinical practices, yet it is not consistently disseminated to those who can use it. The resulting deficits in ...knowledge and service provision may be especially pronounced in low- and middle-income countries that have greater resource constraints. Tuberculosis treatment is an important area for assessing professionals' knowledge and practices because of the effectiveness of existing treatments and recognized gaps in professionals' knowledge about treatment. This study surveyed 384 health professionals in China, India, Iran, and Mexico on their knowledge and practices related to tuberculosis treatment. Few respondents correctly answered all five knowledge questions (12%) or self-reported performing all five recommended clinical practices "often or very often" (3%). Factors associated with higher knowledge scores included clinical specialization and working with researchers. Factors associated with better practices included training in the care of tuberculosis patients, being based in a hospital, trusting systematic reviews of randomized controlled double-blind trials, and reading summaries of articles, reports, and reviews. This study highlights several strategies that may prove effective in improving health professionals' knowledge and practices related to tuberculosis treatment. Facilitating interactions with researchers and training in acquiring systematic reviews may be especially helpful.
It is widely agreed that the practices of clinicians should be based on the best available research evidence, but too often this evidence is not reliably disseminated to people who can make use of ...it. This "know-do" gap leads to ineffective resource use and suboptimal provision of services, which is especially problematic in low- and middle-income countries (LMICs) which face greater resource limitations. Family planning, including intrauterine device (IUD) use, represents an important area to evaluate clinicians' knowledge and practices in order to make improvements.
A questionnaire was developed, tested and administered to 438 individuals in China (n = 115), Kazakhstan (n = 110), Laos (n = 105), and Mexico (n = 108). The participants responded to ten questions assessing knowledge and practices relating to contraception and IUDs, and a series of questions used to determine their individual characteristics and working context. Ordinal logistic regressions were conducted with knowledge and practices as dependent variables.
Overall, a 96 % response rate was achieved (n = 438/458). Only 2.8 % of respondents were able to correctly answer all five knowledge-testing questions, and only 0.9 % self-reported "often" undertaking all four recommended clinical practices and "never" performing the one practice that was contrary to recommendation. Statistically significant factors associated with knowledge scores included: 1) having a masters or doctorate degree; and 2) often reading scientific journals from high-income countries. Significant factors associated with recommended practices included: 1) training in critically appraising systematic reviews; 2) training in the care of patients with IUDs; 3) believing that research performed in their own country is above average or excellent in quality; 4) being based in a facility operated by an NGO; and 5) having the view that higher quality available research is important to improving their work.
This analysis supports previous work emphasizing the need for improved knowledge and practices among clinicians concerning the use of IUDs for family planning. It also identifies areas in which targeted interventions may prove effective. Assessing opportunities for increasing education and training programs for clinicians in research and IUD provision could prove to be particularly effective.
•The indigenous aerobic denitrifiers were enhanced in situ reservoir system.•The enhanced system performed very well in terms of nitrogen and carbon removal.•The carbon utilization diversity of the ...microbial communities varied significantly over the spatial and temporal variation.
Indigenous oligotrophic aerobic denitrifiers nitrogen removal characteristics, community metabolic activity and functional genes were analyzed in a micro-polluted reservoir. The results showed that the nitrate in the enhanced system decreased from 1.71±0.01 to 0.80±0.06mg/L, while the control system did little to remove and there was no nitrite accumulation. The total nitrogen (TN) removal rate of the enhanced system reached 38.33±1.50% and the TN removal rate of surface sediment in the enhanced system reached 23.85±2.52%. TN removal in the control system experienced an 85.48±2.37% increase. The densities of aerobic denitrifiers in the enhanced system ranged from 2.24×105 to 8.13×107cfu/mL. The abundance of nirS and nirK genes in the enhanced system were higher than those of in the control system. These results suggest that the enhanced in situ indigenous aerobic denitrifiers have potential applications for the bioremediation of micro-polluted reservoir system.