2021 No. 5

Editorial
Attaching Importance to the Standardized Construction of Artificial Intelligence Database of Medical Imaging
SHI Zhenwei, LIU Zaiyi
2021, 12(5): 599-601. doi: 10.12290/xhyxzz.2021-0507
Abstract:

Medical imaging is regarded as one of the most potential domains where artificial intelligence can be applied in practice. However, artificial intelligence is facing challenges resulting from continuous growth of data, such as lack of high-quality data, lack of standardization in domain, lack of effective data management and regulation. Therefore, it is necessary to construct a standardized medical imaging database complying with the national condition of China, laws/regulations, and using habits of researchers. FAIR data principle (findable, accessible, interoperable, and reusable) may play a key role in database construction, data acquisition, and regulating descriptions of medical imaging data. Looking forward to boosting the standardized construction of artificial intelligence databases of medical imaging under the combined efforts of national researchers.

Specialist Forum
Multi-modal Deep Learning and Its Applications in Ophthalmic Artificial Intelligence
LI Xirong
2021, 12(5): 602-607. doi: 10.12290/xhyxzz.2021-0500
Abstract:
Deep learning, for its powerful learning capability and high usability, has been a prevalent algorithm of machine learning and a core technique for artificial intelligence(AI) in medicine and healthcare. Due to the importance of medical imaging in many tasks such as health screening, disease diagnosis, precise treatment, and prognosis prediction, deep learning of structural analysis and semantic understanding for medical images is becoming an important interdisciplinary research direction. In clinical scenarios, in order to achieve a more accurate diagnosis, doctors need to simultaneously refer to multiple modalities of medical imaging for a comprehensive analysis and judgment. This article introduced the basic concepts and working principles of multimodal deep learning in such scenarios, reviewed recent research progress on applying multi-modal deep learning in both generic medical fields and ophthalmology, and discussed technical challenges and also envision potential applications of multi-modal deep learning in AI-assisted ophthalmology.
Application of Human-Computer Interaction Technology in Ancillary Diagnosis of Nervous System Diseases: Current Situation and Prospect
LI Yang, WANG Liuping, HUANG Jin, FAN Xiangmin, TIAN Feng
2021, 12(5): 608-613. doi: 10.12290/xhyxzz.2021-0522
Abstract:
With the development of human-computer interaction, how to use intelligent, natural and efficient methods to promote the development of the medical field has become a hot topic of research. Nervous system diseases have a great impact on the quality of people's daily life. Using the method of human-computer interaction for early warning and ancillary diagnosis of nervous system diseases can reduce discomfort of patients during exams and work intensity of doctors, which is of great significance to both doctors and patients. This paper discusses the application status, existing problems, future development of human-computer interaction in the ancillary diagnosis of nervous system diseases, and how to use computer technology to improve traditional medical diagnosis methods from the perspective of interaction.
Big Medical Data and Medical AI Standards: Status Quo, Opportunities and Challenges
ZHANG Zhifei, YANG Zhengxin, HUANG Yunyou, ZHAN Jianfeng
2021, 12(5): 614-620. doi: 10.12290/xhyxzz.2021-0472
Abstract:
Big medical data and medical artificial intelligence (AI) not only have the great potential for improving the utilization of medical resources and the quality of medical service, but also pose challenges to privacy protection and technical risks. Standards are the consensus and norms for constructing, evaluating, and applying new technologies. The clinical application of big medical data and medical AI urgently needs regulations on data, systems, measurement standards, and codes of practice for evaluating new technologies. This paper defines big medical data and medical AI standards, including data-related standards, public datasets, benchmarks, codes of practice, and summarizes state-of-the-art and state-of-the-practice of big medical data and medical AI standards. While looking forward to the development prospect of big medical data and medical AI, we propose an innovative architecture consisting of big-data-and-AI-enhanced medical information systems and the big medical science infrastructure.
Opinion
Universal Definition and Classification of Heart Failure Triggers the Standardization of Diagnosis and Treatment of Heart Failure
XU Fen, ZHOU Zhou
2021, 12(5): 621-623. doi: 10.12290/xhyxzz.2021-0436
Abstract:
Recently, the Heart Failure Society of America, Heart Failure Association of the European Society of Cardiology and Japanese Heart Failure Society jointly issued a consensus statement on Universal Definition and Classification of Heart Failure. Natriuretic peptides were written into the universal definition of heart failure and became the second cardiac marker written into the definition of disease after troponin.The consensus proposed new four classes of heart failure according to ejection fraction(EF), including heart failure with reduced EF(HFrEF), heart failure with mildly reduced EF(HFmrEF), heart failure with preserved EF(HFpEF)and heart failure with improved EF(HFimpEF). Universal Definition and Classification of Heart Failure has wide applicability, which is helpful for clinicians, researchers and patients to understand and adopt.It is an important first step forward to the standardization of diagnosis and treatment of heart failure. While using the universal definition and classification of heart failure, clinical staff should actively conduct scientific research to promote the standardization of diagnosis and treatment of heart failure.
Guideline and Consensus
Clinical Practice Guidelines for ERAS in China (2021)(Ⅰ)
Chinese Society of Surgery, Chinese Society of Anesthesiology
2021, 12(5): 624-631. doi: 10.12290/xhyxzz.20210001
Abstract:
Clinical Practice Guidelines for ERAS in China (2021)(Ⅱ)
Chinese Society of Surgery, Chinese Society of Anesthesiology
2021, 12(5): 632-640. doi: 10.12290/xhyxzz.20210002
Abstract:
Clinical Practice Guidelines for ERAS in China(2021)(Ⅲ)
Chinese Society of Surgery, Chinese Society of Anesthesiology
2021, 12(5): 641-649. doi: 10.12290/xhyxzz.20210003
Abstract:
Clinical Practice Guidelines for ERAS in China(2021)(Ⅳ)
Chinese Society of Surgery, Chinese Society of Anesthesiology
2021, 12(5): 650-657. doi: 10.12290/xhyxzz.20210004
Abstract:
Clinical Practice Guidelines for ERAS in China(2021)(Ⅴ)
Chinese Society of Surgery, Chinese Society of Anesthesiology
2021, 12(5): 658-665. doi: 10.12290/xhyxzz.20210005
Abstract:
Correlation Between the Dermoscopic Features and the Histopathological Features of Common Erythematosquamous Skin Diseases: the Consensus Statement of Chinese Experts (2021)
Dermatology Branch of China International Exchange and Promoting Association for Medical and Health Care, Huaxia Skin Image and Artificial Intelligence Cooperation, China International Exchange and Promotive Association for Medical and Health Care, Department of Dermatology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College
2021, 12(5): 666-673. doi: 10.12290/xhyxzz.2021-0520
Abstract:
As a non-invasive and rapid diagnostic tool, dermoscopy has been widely applied to erythematosquamous skin diseases, and has significantly improved the efficiency of clinical work and diagnostic accuracy. Most dermoscopic features of erythematosquamous skin diseases are associated with histopathological changes. Here a consensus was reached on the correlation between the dermoscopic features and the histopathological features of common erythematosquamous skin diseases, in order to improve the understanding of dermatologists on the dermoscopic features of erythematosquamous skin diseases.
Consensus on the Treatment of Adrenocortical Carcinoma with Mitotane(2021)
Multi-disciplinary Team for Rare Diseases, Peking Union Medical College Hospital
2021, 12(5): 674-683. doi: 10.12290/xhyxzz.2021-0554
Abstract:
Adrenocortical carcinoma (ACC) is a rare malignant tumor that occurs in the adrenal cortex, and surgery is the first choice for treating ACC. Medical therapy and radiotherapy are adjuvant treatments for patients with unresectable metastasized tumors or incomplete resection. Mitotane is the only adrenergic cytotoxic drug approved by the American Food and Drug Administration and European Medicines Agency for the treatment of ACC by inhibiting mitochondrial respiration, causing mitochondrial membrane dysfunction and inducing endoplasmic reticulum stress, and it reduces the secretory function of ACC cells by inhibiting the function and expression of several enzymes in the adrenocortical steroidogenesis pathway. Mitotane is recommended as the first choice for adjuvant therapy by multiple international guidelines and consensus. During treatment with mitotane, its blood concentration should be maintained between 14 to 20 mg/L, while the related adverse drug reactions and drug-drug interactions should also be monitored. To promote further standardization of the clinical application of mitotane, and to ensure the effectiveness and safety of mitotane, Peking Union Medical College Hospital relied on the multi-disciplinary team for the diagnosis and treatment of rare diseases, organized experts and scholars in related fields to discuss and modify many times, and finally formed this consensus for clinicians' reference.
Standard and Specification
Procedures of Establishing a Well-annotated Database of Color Fundus Photography of Diabetic Retinopathy for Artificial Intelligence Research
YU Weihong, ZHANG Xiao, WU Chan, CHEN Huan, YANG Zhikun, HE Feng, ZHANG Zhiqiao, ZHANG Bilei, GONG Di, WANG Yuelin, YANG Jingyuan, LI Bing, SUN Yanyuan, MA Yajing, LU Huiqin, XIA Wei, ZHOU Wei, ZHANG Donglei, PAN Qingmin, YANG Ning, WANG Shuna, SUN Xiaolei, YU Ying, SU Chang, WAN Bo, WANG Mingqi, WANG Min, CHEN Youxin
2021, 12(5): 684-688. doi: 10.12290/xhyxzz.2021-0613
Abstract:
With the development of artificial intelligence technology, explosive research on the recognition of retina diseases with artificial intelligence was published worldwide, among which, diabetic retinopathy attracted most attention. However, evaluation of different artificial intelligence software remains a problem. The Department of Ophthalmology, Peking Union Medical College Hospital published the first standard well-annotated color photo database of diabetic retinopathy for artificial intelligence research in China. We herein present the procedures for the establishment of the database and aim to provide reference for the future development of similar databases.
Original Contributions
Early Diagnosis Model of Mycosis Fungoides Based on Intelligent Analysis of Dermoscopic Images
LIU Zhaorui, ZHANG Yilan, XIE Fengying, LIU Jie
2021, 12(5): 689-697. doi: 10.12290/xhyxzz.2021-0496
Abstract:
  Objective  To compare the application value of the binary classification model based on dermoscopic images of convolutional neural network (CNN) in the diagnosis of mycosis fungoides (MF) and inflammatory dermatosis.  Methods  Patients diagnosed with early MF or inflammatory dermatosis with similar clinical manifestations in the dermatology clinic of Peking Union Medical College Hospital from January 2016 to December 2020 were retrospectively included. The patients were divided into the training set and the test set at a ratio of 4∶1. Six classical network structures were trained by using the dermoscopic images of patients in the training set, and the CNN binary classification model was constructed by using transfer learning. At the same time, in the test set, 1 image of each patient that was randomly selected, together with clinical images of the skin lesions, was interpreted by 13 dermatologists. Compare the CNN binary classification model with dermatologists in the differential diagnosis of early MF and inflammatory dermatosis in the test set. The results were expressed in terms of area under the curve (AUC), sensitivity, specificity, Kappa coefficient, etc., and receiver operating characteristic (ROC) curve was used for visual analysis.  Results  A total of 48 patients with early MF (402 dermoscopic images) and 96 patients with inflammatory dermatosis (557 dermoscopic images) were included. Among them, there were 117 cases in the training set (772 dermoscopic images), and 27 cases in the test set (187 dermoscopic images). In the test set, the sensitivity and specificity of dermatologists in the differential diagnosis of early MF and inflammatory dermatosis were 70.19% (95% CI: 59.68%-80.70%) and 94.74% (95% CI: 91.77%-97.71%) respectively, and the Kappa coefficient is 0.677(95% CI: 0.566-0.789). When classified by the single image, the AUC of the CNN binary classification model for the differential diagnosis of early MF and inflammatory dermatosis was 0.87 (95% CI: 0.84-0.89); the sensitivity and specificity were 75.02% (95% CI: 70.19%-79.85%) and 82.02% (95% CI: 79.30%-84.87%), respectively; the Kappa coefficient was 0.563(95% CI: 0.507-0.620). When classified by cases, the AUC of the CNN binary classification model for the differential diagnosis of early MF and inflammatory dermatosis was 0.97 (95% CI: 0.95-0.99); the sensitivity and specificity were 87.50% (95% CI: 78.55%-96.45%) and 93.85% (95% CI: 88.93%-98.77%), respectively; the Kappa coefficient was 0.920(95% CI: 0.884-0.954). The ROC curve showed that the AUC of the CNN binary classification model with EfficientNet-B0 for diagnosing MF was 0.99 when classified by cases, the sensitivity and specifity were 88.9% and 100%, and the corresponding point of the average diagnostic sensitivity and specificity of 13 dermatologists were at the lower right of the curve.  Conclusions  The CNN binary classification model based on the intelligent analysis of dermoscopic images can accurately classify early MF and inflammatory dermatosis, and its ability of differential diagnosis is better than the average level of dermatologists.
CT-based Radiomics to Predict Recurrence of Bladder Cancer after Resection in One Year: A Preliminary Study
ZHANG Gumuyang, XU Lili, MAO Li, LI Xiuli, JIN Zhengyu, SUN Hao
2021, 12(5): 698-704. doi: 10.12290/xhyxzz.2021-0511
Abstract:
  Objective  To investigate the feasibility of the CT-based radiomics model to predict the recurrence of bladder cancer in one year postoperatively.  Methods  Patients with bladder cancer that received surgical treatment in Peking Union Medical College Hospital from May 2014 to July 2018 were retrospectively enrolled and followed up the recurrence of the disease. Nephrographic phase images of preoperative CT urography(CTU) performed in our hospital were collected. The images were filtered before radiomic feature extraction, and JMIM was used to identify the best radiomic features related to recurrence of bladder cancer. Random forest, AdaBoost, gradient boosting, and their combined model were used to build the model for predicting recurrence of bladder cancer after resection in one year. We applied 10-fold cross validation to validate each model and performed receiver operator characteristic curves to analyze the performance of each model.  Results  A total of 228 cases were included in this study according to inclusion and exclusion criteria. Fifty-one patients had recurrence and the rest 177 patients had no recurrence in one year during postoperative follow-up. In the cross validation, the random forest model, AdaBoost model, gradient boosting model and combined model predicted the recurrence of bladder cancer with AUC of 0.729(95% CI: 0.649-0.809), 0.710(95% CI: 0.627-0.793), 0.709(95% CI: 0.624-0.793)and 0.732(95% CI: 0.651-0.812), accuracy of 76.8%(95% CI: 70.6%-82.0%), 73.7%(95% CI: 67.4%-79.2%), 61.8%(95% CI: 54.7%-67.7%)and 75.0%(95% CI: 68.8%-80.4%), sensitivity of 52.9%(95% CI: 38.6%-66.8%), 62.7%(95% CI: 48.1%-75.5%), 80.4%(95% CI: 64.3%-88.2%)and 58.8%(95% CI: 44.2%-72.1%), specificity of 83.6%(95% CI: 77.1%-88.6%), 76.8%(95% CI: 69.8%-82.7%), 56.5%(95% CI: 48.9%-63.9%)and 79.7%(95% CI: 72.8%-85.2%), respectively.  Conclusion  Integration of CT-based radiomics prediction models can predict the recurrence risk of bladder cancer in one year postoperatively.
The Value of Model Based on Radiomics Features of T2-weighted Imaging and Clinical Feature in Diagnosing the Depth of Stromal Invasion of Cervical Squamous Cell Carcinoma
REN Jing, HE Yonglan, LI Yuan, CAO Ying, XIA Chen, XIANG Yang, XUE Huadan, JIN Zhengyu
2021, 12(5): 705-712. doi: 10.12290/xhyxzz.2021-0437
Abstract:
  Objective  To investigate the prediction value of a clinical-radiomics model based on T2- weighted imaging (T2WI) and clinical features for diagnosing deep stromal invasion (DSI) in patients with early-stage cervical squamous cell carcinoma.  Methods  Patients with early-stage cervical squamous cell carcinoma that underwent radical hysterectomy in Peking Union Medical College Hospital from January 2017 to February 2021 were retrospectively included and randomly divided into the training set and the validation set with the the ratio of 8∶2. The preoperative clinical features and the radiomics features of sagittal T2WI images were obtained. After selection of key features, a radiomics model, a clinical model, and a clinical-radiomics model for diagnosing DSI in early-stage cervical squamous cell carcinoma were developed by Logistic regression based on the training set. The performance of different models was compared by the receiver operating characteristic (ROC) curve in the validation set.  Results  A total of 168 patients with early-stage cervical squamous cell carcinoma that met the inclusion and exclusion criteria were included in this study. They were randomly divided into the training set (n=135) and the validation set (n=33), in which 72 cases had histopathologically confirmed superficial stromal invasion and 96 cases had DSI. Four radiomics features and three clinical parameters (age, Federation International of Gynecology and Obstetrics stage, and preoperative squamous cell carcinoma antigen levels) were selected and used to develop models. In the validation set, the clinical-radiomics model showed better diagnostic performance with the area under the curve (AUC) of 0.820 (95% CI: 0.665-0.974) than the clinical model[AUC: 0.797(95% CI: 0.623-0.971)] and the radiomics model[AUC: 0.793(95% CI: 0.633-0.954)].The sensitivity, specificity, and accuracy of the clinical-radiomics model were 85.7%(95% CI: 49.8%-100%), 73.7%(95% CI: 57.9%-100%), and 78.8%(95% CI: 69.7%-93.9%), respectively.  Conclusion  Radiomics features based on T2WI images combined with clinical features can be used as a noninvasive preoperative method to determine the depth of stromal invasion in early-stage cervical squamous cell carcinoma.
Texture Analysis of Sequential Images of T2-weighted Imaging and Diffusion-weighted Imaging for Predicting the Efficacy of Chemoradiotherapy in Cervical Squamous Cell Carcinoma
SU Baiyan, QI Yafei, GUAN Hui, HE Yonglan, XUE Huadan, JIN Zhengyu
2021, 12(5): 713-720. doi: 10.12290/xhyxzz.2021-0380
Abstract:
  Objective  To investigate the correlation of the texture parameters of T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) with the efficacy of chemoradiotherapy in cervical squamous cell carcinoma.  Methods  Patients with squamous cell carcinoma of the cervix that underwent chemoradiotherapy from February 2015 to January 2016 in Peking Union Medical College Hospital were included retrospectively, and were divided into the disease-progressive group and the disease-stable group according to their prognosis. Texture analysis of baseline T2WI and DWI images before chemoradiotherapy was carried out with Texrad software, and the texture parameters of spatial scale filter (SSF) with radius values of 2, 4 and 6 were obtained. The differences of texture parameters between the two groups were compared, and the correlation between the texture parameters and the curative of chemoradiotherapy in patients with cervical squamous cell carcinoma was analyzed by multivariate Cox regression. Receiver operating characteristic (ROC) curve was used to analyze the performance of texture parameters in predicting disease progression after chemoradiotherapy in patients with cervical squamous cell carcinoma.  Results  A total of 121 patients with squamous cell carcinoma of the cervix that met the inclusion and exclusion criteria were enrolled in this study. There were 46 cases in the disease-progressive group and 75 cases in the disease-stable group. In T2WI sequential images, there were significant differences in the texture parameters of means (SSF2, SSF4, SSF6), skewness (SSF2, SSF4), and entropy (SSF4, SSF6) between disease-progressive group and disease-stable group (all P < 0.05). In DWI sequential images, there were significant differences in the texture parameters of means (SSF2, SSF4, SSF6), skewness (SSF4, SSF6), and kurtosis (SSF2, SSF4) between the two groups (all P < 0.05). Multivariate Cox regression analysis showed that the texture parameter of means (SSF2, SSF4, SSF6) of T2WI and the texture parameters of means (SSF2, SSF6), entropy (SSF2, SSF4, SSF6) and skewness (SSF4, SSF6) of DWI were correlated with the efficacy of chemoradiotherapy in patients with cervical squamous cell carcinoma (P < 0.05). The Results of ROC analysis showed that the texture parameter of means (T2WI-SSF2, T2WI-SSF4, T2WI-SSF6, DWI-SSF2, DWI-SSF6) and skewness (DWI-SSF6) could predict the progression of cervical squamous cell carcinoma after chemoradiotherapy in patients with cervical squamous cell carcinoma. The area under the curve (AUC) was 0.625-0.746. Among them, the mean of T2WI-SSF4 was the most effective (AUC: 0.746), followed by the mean of T2WI-SSF2 (AUC: 0.725) and the mean of T2WI-SSF6 (AUC: 0.703).  Conclusions  The texture parameters of baseline T2WI and DWI sequences were correlated with the curative effect of chemoradiotherapy in patients with cervical squamous cell carcinoma. The parameters of means and skewness can predict the progression of cervical squamous carcinoma after chemoradiotherapy, and the mean has a higher predictive power.
Difference in Gut Microbiota between the First and the Second Trimester of Pregnancy and the Association of Gut Microbiota with Gestational Diabetes Mellitus: A Prospective Cohort Study
WANG Pei, MA Liangkun, LIU Juntao
2021, 12(5): 721-728. doi: 10.12290/xhyxzz.20200122
Abstract:
  Objective  To characterize the characteristics of gut microbiota and its function in the first (T1) to second trimester(T2) of pregnancy and to evaluate its association with gestational mellitus diabetes (GDM).  Methods  A prospective cohort study was conducted in Peking Union Medical College Hospital from May to December 2017. The pregnancies were divided into GDM group and non-GDM group (control group) according to the Results of 75 g oral glucose tolerance test at 24 to 48 weeks of gestation. Stool samples of all participants were collected in the first and the second trimester. The V4 region of the 16S rRNA gene was sequenced and analyzed. Multivariate Logistic regression analysis was used to investigate the relationship between Alpha diversity, relative abundance of intestinal flora and GDM.  Results  A total of 145 pregnancies, of whom 34 diagnosed with GDM (GDM group) and 111 healthy (control group) were analyzed. The Alpha diversity of the GDM group (Shannon index and Simpson index) was significantly lower than that of the control group (P < 0.05). LEfSe analysis revealed that the relative abundance of several genera was different between the 2 groups in T1 or T2. Multivariate Logistic analysis showed that Shannon index ≤6.51 (OR=3.15, 95% CI: 1.32-7.52), Simpson index ≤0.96 (OR=2.54, 95% CI: 1.09-5.89), the lower relative abundance of Alloprevotella(OR=2.65, 95% CI: 1.09-6.44) and Lachnospira(OR=3.17, 95% CI: 1.33-7.55) in the first trimester were risk factors for GDM. The pathways of LPS biosynthesis, energy metabolism, glucose metabolism and amino acid metabolism of gut microbiome revealed through the Tax4Fun analysis were significantly enriched in the GDM group in T2.  Conclusions  Compared with healthy controls, the functional characteristics of intestinal microflora in GDM patients during the second trimester were significantly enriched in functional pathways related to LPS synthesis, energy metabolism, glucose metabolism and amino acid metabolism. The decreases of the diversity as well as the relative abundance of some genus in the early pregnancy are the risk factors for GDM.
Clinical Efficacy, Safety and Pharmacokinetics of Tracing Injection of Mitoxantrone Hydrochloride for Tracing Sentinel Lymph Nodes in Thyroid Carcinoma: A Phase Ⅰ Clinical Trial
CHEN Shaobo, LIU Weinan, ZHANG Shengjie, SUN Jian, YANG Rui, WANG Shujun, CHEN Ge
2021, 12(5): 729-735. doi: 10.12290/xhyxzz.2021-0281
Abstract:
  Objective  To investigate the staining and tracing effect of mitoxantrone hydrochloride injection for tracing lymph nodes in radical thyroidectomy, and its safety and tolerability.  Methods  In this study, a single-centered, open-labeled, blank control trial was designed. According to the principle of dose escalation, the human tolerance test and kinetic test were performed in dose groups one by one. Patients scheduled for radical thyroidectomy in Peking Union Medical College Hospital from December 2016 to March 2017 were prospectively enrolled as subjects. The subjects were divided into the blank control group and the test groups. The blank control group routinely underwent surgery without tracer injection; the test groups were injected with 0.2 mL, 0.4 mL, 0.6 mL, 0.8 mL, or 1.2 mL of mitoxantrone hydrochloride for tracing in the thyroid gland after exposing the surgical field, and an escalation test was performed in the five dose groups from low to high. The primary efficacy measures were lymph node tracing rate and lymph node staining rate; the secondary efficacy measures were staining-degree scores of lymph nodes and the success rate of tracing persistence. Changes in signs of subjects were recorded for safety assessment. Venous blood was collected from the subjects before administration and at 10 min, 20 min, 30 min, 60 min, 90 min and 120 min after administration to calculate the kinetic parameters.  Results  A total of 27 subjects were included, including 3 subjects in the blank control group, and 3, 3, 6, 6 and 6 subjects in the 0.2 mL, 0.4 mL, 0.6 mL, 0.8 mL and 1.2 mL dose group, respectively. When the test proceeded to the third dose group (0.6 mL), the injection site of the subjects was saturated with the injection dose, and the test was terminated. Therefore, a total of 15 subjects completed the test. No lymph node was stained in blank control group.The lymph nodes of 12 subjects showed blue staining, and the parathyroid glands were not stained and negatively visualized. There was significant difference in the staining-degree scores of lymph nodes before and after administration in the 0.2 mL, 0.4 mL and 0.6 mL dose groups (P < 0.05). The average staining rates of lymph nodes were 90.47%, 91.67% and 91.36%, respectively. And the average tracing rates of lymph nodes were 79.17%, 100% and 98.67%, respectively. In the test groups, the duration of lymph node dissection was 5-20 min; no stained lymph nodes faded, and the success rate of tracing was 100%. In the safety evaluation, no serious adverse events occurred in the blank control group and test groups, and the cause of adverse events was certainly not related to this tracer. Kinetic tests showed that the tracer was rapidly absorbed after glandular injection; the blood concentration peaked at 10 min after injection (the maximum plasma concentration was 13.1 μg/L) and it was rapidly eliminated.  Conclusions  The use of 0.2-0.6 mL mitoxantrone hydrochloride injection in radical thyroidectomy can play a better role in tracing lymph nodes, with high safety and good tolerance.
Reviews
Artificial Intelligence Analysis of Nerve Fibers Based on Corneal Confocal Microscopy
WU Jun, FEI Sijia, SHEN Bo, ZHANG Hanwen, HUANG Jianfeng, PAN Qi, ZHAO Jianchun, DING Dayong
2021, 12(5): 736-741. doi: 10.12290/xhyxzz.2021-0510
Abstract:
Diabetic peripheral neuropathy (DPN) is one of the most common chronic complications of diabetes. Traditional DPN diagnostic methods are based on clinical symptoms and signs as well as electrophysiological examination, which are mainly used to detect the lesions of large nerve fibers. However, the small nerve fibers are the earliest ones damaged in DPN. Corneal confocal microscopy (CCM) can analyze the changes of corneal nerve fibers under a high power microscope. It is a rapid, repeatable and quantitative noninvasive technique to measure small nerve fibers. It can diagnose DPN early and evaluate neuromorphological changes prospectively. It has a good application expectation. During this article, we summarized the role and limitations of DPN's most reliable parameters of corneal nerve in evaluating diabetic autonomic neuropathy and diabetic micro-vascular complications. Further, we reviewed the clinical application of CCM in evaluating diabetic neuropathy and analysis methods of CCM related artificial intelligence, in order to provide references for clinical diagnosis and treatment.
Research Progress of Cancer Classification Based on Deep Learning and Histopathological Images
YAN Rui, CHEN Limeng, LI Jintao, REN Fei
2021, 12(5): 742-748. doi: 10.12290/xhyxzz.2021-0452
Abstract:
Accurate classification of cancer is directly related to the choice of treatment options and prognosis. Pathological diagnosis is the gold standard for cancer diagnosis. The digitalization of pathological images and breakthroughs in deep learning have made computer-aided diagnosis and prediction about prognosis possible. In this paper, we first briefly describe four deep learning methods commonly used in this field, and then review the latest research progress in cancer classification based on deep learning and histopathological images. Finally, the general problems in this field are summarized, and the possible development direction in the future is suggested.
Application of Artificial Intelligence in Neuroimaging of Stroke
HAN Xiaowei, LI Ming, ZHANG Bing
2021, 12(5): 749-754. doi: 10.12290/xhyxzz.2021-0491
Abstract:
Artificial intelligence (AI) has risen rapidly in the research field of computer science. The amount of data generated in the process of medical imaging is huge, so it is very suitable to use artificial intelligence technology for related data processing. Neuroimaging of patients with stroke plays a key role in clinical diagnosis, treatment and follow-up. AI technology plays an increasingly important role in the processing and analysis of imaging data of stroke. This paper mainly reviews the research progress of AI technology in neuroimaging of ischemic and hemorrhagic stroke, focusing on the detection of ischemic stroke, judgment of ischemic state of responsible brain area and treatment evaluation, as well as the application of AI technology in the diagnosis, quantitative analysis and treatment evaluation of hemorrhagic stroke. At the same time, the current situation of its clinical transformation application was analyzed. Furthermore, it discusses the main limitations of the current application of AI in stroke neuroimaging, and prospects for future development.
Current Situation and Prospect of Teleophthalmology
ZHANG Xiao, CHEN Youxin
2021, 12(5): 755-760. doi: 10.12290/xhyxzz.2021-0198
Abstract:
With the development of technology, doctors can diagnose and treat many diseases through telemedicine. At present, teleophthalmology is mainly used in screening and diagnosing some ocular diseases, monitor-ing chronic ocular diseases, as well as teleconsultation. With the increasing demand and application of teleophthalmology, some problems will gradually become prominent, such as insufficient equipment and staff, medical risks, patient acceptance and satisfaction, network security, privacy, and covering of medical insurance. The global pandemic of COVID-19 has unexpectedly brought telemedicine to the forefront of ophthalmic services, and may continue to change the way of ophthalmic diagnosis and treatment. With the development of artificial intelligence technology, the expansion of 5G communication network coverage, the standardized training of primary medical staff, and the introduction of relevant laws and regulations, teleophthalmology will become more improved, universal, and widely applied, so as to provide patients with sustainable medical services of higher quality.
The Application and Research Progress of Artificial Intelligence in Corneal Related Diseases
DI Yu, LI Ying
2021, 12(5): 761-767. doi: 10.12290/xhyxzz.2020-0098
Abstract:
Artificial intelligence(AI)is the frontier of computer science. In recent years, AI has developed rapidly in many fields, and its research in ophthalmology is also increasing. The research of AI in corneal related diseases mainly includes the early diagnosis and grading of keratoconus, preoperative evaluation of corneal refractive surgery, prediction of surgical parameters, judgment of the classification and degree of infectious keratitis, evaluation of reintervention after corneal transplantation, auxiliary detection of corneal nerve endings in diabetic peripheral neuropathy, and screening of pterygium. Through the neural network, the support vector machine, and the decision tree, the sensitivity and specificity of the model can reach more than 90%. AI can provide objective clinical decision-making for clinicians and precise clinical treatments for patients. This article reviews the research of AI in corneal diseases in recent years.
Detection Technology and Prospects of Clinical Application of Circulating Tumor Cells in Pancreatic Cancer
YANG Xiaoying, QIN Cheng, ZHAO Bangbo, LI Tianhao, CAO Hongtao, WANG Weibin
2021, 12(5): 768-773. doi: 10.12290/xhyxzz.2020-0005
Abstract:
Pancreatic cancer is one of the highly malignant tumors of the digestive system, and early metastasis is an important reason for its short survival time. Circulating tumor cells (CTCs) are a type of tumor cells that are released into the bloodstream from the primary tumor and/or metastatic sites, which have properties similar to primary tumor cells and are closely related to distant metastasis of tumors. In vitro quantitative detection of CTCs is of great significance in the early diagnosis, auxiliary staging, treatment efficacy, and prognosis evaluation of pancreatic cancer. Here we generally reviewed current research on the CTCs enrichment methods of pancreatic cancer and its application in screening pancreatic cancer and metastasis, and evaluating the prognosis.
New Advances in Anesthesia Management of Enhanced Recovery After Surgery in Craniotomy
LI Yi, LIU Zijia, LABA Ciren, SHEN Le
2021, 12(5): 774-779. doi: 10.12290/xhyxzz.20200265
Abstract:
Enhanced recovery after surgery (ERAS) is a new perioperative management concept with a series of evidence-based clinical optimization protocols to minimize surgical stress response, reduce postoperative complications and enhance patients' recovery. Anesthesia management is an important content of ERAS for the approach of a multidisciplinary team. Craniotomy is one of the most high-risk operations with a long duration and a high rate of complication. In recent years, the clinical research and implementation of ERAS management in craniotomy have gradually been carried out and rapidly developed. This article reviews the new progress in anesthesia management of ERAS in craniotomy.
Complicated and Rare Disease
A Case Report of Gorham-Stout Syndrome with Chyloperitoneum
WANG Qiang, ZHOU Xi, HUO Li, WU Xi
2021, 12(5): 780-783. doi: 10.3969/j.issn.1674-9081.2020.00.009
Abstract:
Gorham-Stout syndrome is a rare disease characterized by osteolysis, in which idiopathic intraosseous angiomatous proliferation leads to progressive resorption of bone. Chylothorax can be associated with Gorham-Stout syndrome in some cases, but chyloperitoneum is seldom seen. This paper reports a case of Gorham-stout syndrome that was admitted to the gastroenterology department because of ascites. The way of chyloperitoneum formation was confirmed by lymphangiography.
Clinical Practice Guidelines
Investigation and Evaluation of Chinese Clinical Practice Guidelines Published in Medical Journals in 2019:Funding and Conflicts of Interest
SHI Qianling, WANG Zijun, MA Yanfang, WANG Jianjian, LIU Hui, LUO Xufei, YANG Nan, LYU Meng, LIU Yunlan, XUN Yangqin, GUAN Quanlin, CHEN Yaolong
2021, 12(5): 784-790. doi: 10.12290/xhyxzz.2021-0541
Abstract:
  Objective  To analyze the situation of funding and conflicts of interest of Chinese clinical practice guidelines published in medical journals in 2019, so as to provide reference for the standardized management of funding and conflicts of interest in Chinese guidelines.  Methods  We systematically searched four electronic databases and additional sources for Chinese guidelines published in medical journals from January 1 to December 31, 2019. After screening, two researchers independently extracted data on funding and conflicts of interest and cross checked. We analyzed data by using EndNote X9, Microsoft Excel 2019 and SPSS 25.0 software.  Results  We finally included 226 guidelines, of which, 187 were published in Chinese and 39 in English. There were 68 (30.1%) and 120 (53.1%) guidelines that reported funding and conflicts of interest, respectively, and only 41 (18.1%) that reported on both. There were 64 (28.3%) guidelines reported to have received funding, involving 170 projects, which were mainly at the national level (82, 48.2%) and ministerial and provincial-level (52, 30.6%). Only 5 (2.2%) guidelines reported the role of the funder in the stages of the development process. Among the 120 guidelines that reported conflicts of interest, 102 (85.0%) simply described "no conflicts of interest", while there were less than 10% of guidelines that reported the methods of collection, assessment and management for conflicts of interest, and there was no public access available to such information.  Conclusions  There is a lack of effective management of funding and conflicts of interest in the Chinese guidelines published in medical journals in 2019, which is reflected in the low reporting rate, insufficient reporting contents and unavailable materials provided by the guideline developers. We suggest that guideline developers in the future should pay more attention to the field of guideline funding and conflicts of interest. Guidelines managers and researchers should actively improve management policies, and develop relevant reporting standards.
Medical Supports to Tibet & Xinjiang
Analysis of the Situation and Influencing Factors of Fungal Detection in Adult Patients with Severe Diseases in Tibet Region
FU Jianlei, SU Longxiang, PAN Wenjun, LIN Guoying, CHAI Wenzhao, WANG Xiaoting
2021, 12(5): 791-797. doi: 10.12290/xhyxzz.2021-0531
Abstract:
  Objective  To explore the situation of fungal detection in adult patients with severe diseases in Tibet region, and further analyze the influencing factors of fungal detection.  Methods  All patients admitted to the Department of Critical Care Medicine Tibet Autonomous Region People's Hospital from January 1, 2018 to December 31, 2019 were retrospectively analyzed. According to the results of fungal detection, the patients were then divided into the positive group and the negative group. The distribution of fungal strains detected and clinical data of the two groups were collected by the electronic medical record system. Multivariate Logistic regression was used to analyze the influencing factors of fungal detection.  Results  A total of 755 severe patients (4917 specimens were submitted) were enrolled, including 142 patients with positive fungal detection and 613 patients with negative fungal detection.Of which, 192 fungal strains were detected, including 183(95.3%) Candida strains, 7(3.7%) Aspergillus strains and 2(1.0%) other fungi strains. Among Candida genus, there were 164 strains (89.6%) of Candida albicans, 9 strains (4.9%) of Candida glabrata, 7 strains (3.8%) of Candida parapsilosis, 2 strains (1.1%) of Candida krusei, and 1 strain (0.6%) of Candida tropicalis. Among Aspergillus, there were 3 strains (42.8%) of Aspergillus fumigatus, 2 strains (28.6%) of Aspergillus niger, and 2 strains (28.6%) of Aspergillus flavus. Multivariate Logistic regression analysis showed that, high sequential organ failure assessment score (OR=1.402, 95% CI: 1.277-1.538, P < 0.001), severe digestive diseases (OR=2.671, 95% CI: 1.465-4.872, P=0.001), the duration of tracheal intubation ≥48 h(OR=2.661, 95% CI: 1.611-4.397, P=0.000), the use of carbapenemes ≥24 h (OR=2.825, 95% CI: 1.522-5.245, P=0.001), the use of cephalosporins plus beta lactamase inhibitor ≥24 h (OR=2.678, 95% CI: 1.679-4.272, P=0.000), and the long ICU stay (OR=1.043, 95% CI: 1.011-1.076, P=0.008) were independent risk factors for fungal detection in adult patients with severe diseases. High altitude of residence (OR=0.999, 95% CI: 0.999-1.000, P=0.040) and the high hemoglobin level (OR=0.994, 95% CI: 0.988-0.999, P=0.020) were protective factors.  Conclusions  Positive fungal detection in adult patients with severe diseases in the Tibet area is not uncommon, and Candida albicans is the main strain detected. The factors affecting the detection of fungi involve many aspects. Corresponding preventive measures should be taken according to the characteristics of Tibet.
Drug Therapy for Tumors
Diagnosis and Management of Immunotherapy-related Liver Toxicity
HUA Yuwei, ZHAO Lin
2021, 12(5): 798-806. doi: 10.12290/xhyxzz.2021-0138
Abstract:
Immune checkpoint inhibitors have revolutionized the treatment of cancer. With the increasing use of immunotherapy, clinicians have observed that immunotherapy is often related to a spectrum of adverse events. Liver toxicity is one of the most common and severe immune-related adverse events. It is very important to increase awareness and outline a clear and detailed strategy of managing immune-related liver toxicity. Some guidelines have published systematic advice on the management of immune-related liver toxicity. However, there are controversial problems remaining, including the levels of liver toxicity, the minimal effective dosage of hormone, treatments for hormone-refractory patients, and immunotherapy re-challenge. Additional, finding predictors with high specificity and sensitivity is a pressing concern.
Teaching and Research
Investigation on Cognition of Thyroid Carcinoma Among Medical Students
NIE Yongdu, LIU Penghao, LIN Jinkai, CAO Zenghan, XU Xiequn
2021, 12(5): 807-812. doi: 10.12290/xhyxzz.2021-0359
Abstract:
  Objective  To investigate how well medical students master knowledge about thyroid carcinoma (TC) and thyroid self-examination, and to provide guidance for pre-clinical education and clinical teaching.  Methods  The survey was based on the eight-year program of clinical medicine in Peking Union Medical College Medical College. Students of the third and fourth grade were defined as preclinical medical students (PMS), and students of the fifth and sixth grade were clinical medical students (CMS). The survey was conducted in March 2020 through online questionnaires.  Results  A total of 337 questionnaires were distributed and 274 valid responses were collected with 129 from PMS and 145 from CMS. Generally, CMS had better comprehension of TC, including prognosis (97.2% vs. 64.5%, P < 0.001), diagnosis (95.6% vs. 33.1%, P < 0.001) and surgery indications (82.1% vs. 58.1%, P=0.001). There was no significant difference between PMS and CMS on the acquaintance of the risk factors. However, more CMS stated that ≤5% of thyroid nodules might turn malignant (45.5% vs. 6.5%, P < 0.001), and more CMS suggested that people without nodules should receive TC screening tests (62.1% vs. 41.9%, P=0.001). The percentage of thyroid self-examination in CMS was higher than that of PMS (55.8% vs. 11.6%, P < 0.001), but the rate of regular self-examination was lower (19.8%, 16/81).  Conclusions  Medical education on TC is effective, but CMS tend to be over optimistic about TC, ignore the importance of self-examination, and lack proper awareness of cost-effectiveness in thyroid screening. It is suggested that medical schools should focus more on education of preclinical general health and clinical practices in the future to engage medical students in popularization of health science.
History of Medicine
Academic Influence of the Pharmacy Department of Peking Union Medical College Hospital from 1921 to 1951
ZHU Zhu, ZHANG Bo, LI Naishi, TANG Yan, MEI Dan, LI Dakui
2021, 12(5): 813-818. doi: 10.12290/xhyxzz.20200232
Abstract:
The publications written by the employees of the Pharmacy Department of Peking Union Medical College Hospital from 1921 to 1951 were searched and reviewed systematically. A total of 61 publications, 5 versions of formularies and 6 books have been found so far. The Chinese Pharmacopeia published in 1930 was also involved. Above finding demonstrated a profound and great influence of the early generation of our department on the development of Chinese medical science and pharmaceutical science.