2018 Vol. 9, No. 1

2018, 9(1): 1-1.
HTML (84) PDF(37)
Abstract:
2018, 9(1): 93-96. doi: 10.3969/j.issn.1674-9081.2018.01.017
HTML (12) PDF(166)
Abstract:
Editorials
Abstract:
With the rapid development of artificial intelligence, there is a general consensus of opinion that radiologists' workload can be dramatically reduced with the aid of intelligent image recognition. On the issues of comprehensive diagnosis and treatment, however, there is no certain answer whether or not artificial intelligence can provide better suggestions and comments. Currently, in China, the artificially intelligent imaging technique is mainly focused on simple image recognition, but there is a lack of experience in the accumulation of medical data and the analysis of radiological reports. The mode of integrating artificial intelligence with medical imaging science has just begun. We believe that the progress of science and technology will continue to be the engine of human civilization.
Abstract:
Since rare diseases contribute to a heavy socio-economic burden globally, research into rare diseases will provide enormous social and scientific values. China has a large population, which provides abundant resources, as well as immense challenges, for studying rare diseases. Although advances in specific diseases have been made, lacking of research collaboration and imbalance in allocation of healthcare resources remain as significant challenges in China. With the growing attention towards rare diseases from the government, academic institutions, research societies, and individuals, it is anticipated that research of rare diseases in China will be under swift development with collaboration and innovation. Efforts in various aspects have been made nationally and globally to better meet the challenges of rare diseases.
Specialist Forum
Abstract:
Deep learning, as the most popular research field in artificial intelligence, has been developing rapidly in recent years and become the focus of global attention. Deep learning has demonstrated a powerful role in many application areas. In some visual and auditory recognition tasks, deep learning even shows better performance than human beings. In medical domain, deep learning has become the top choice for researchers to analyze big data, especially medical imaging. This review briefly introduces the history and development of deep learning, and elaborates on the progress of research on deep learning in medical imaging by reviewing the latest and most influential research results. In addition, this paper briefly discusses application of deep learning in medical imaging analysis, as well as the future prospect and challenges of deep learning.
Abstract:
With the advent of the big data era, deep learning has made remarkable breakthroughs compared with traditional pattern-recognition methods in many tasks, such as image classification and detection. In January 2017, the artificial intelligence laboratory of Stanford University applied deep learning to automatic classification of clinical skin images and dermoscopy images, and published the research results in Nature, which represents the latest research progress in the field of automatic analysis of skin images. Herein, we interpret this research from the aspects of database establishment, design of research methods and analysis of the experimental results; we also elaborate the current research status of computer-aided diagnosis of skin images in China, as well as the future developmental prospect of multi-source big data analysis of skin images and intelligent auxiliary diag-nosis, in order to promote the medical diagnostic level of skin diseases in China.
Abstract:
The development of multi-modal magnetic resonance imaging (MRI) provides a new method for early diagnosis of brain diseases. Currently, diagnosis of most neuropsychiatric diseases is only based on clinical symptoms, which lacks objective neuroimaging biomarkers. As univariate analysis that is used in traditional research can only reveal disease-related structural and functional alterations at group level, which limits the clinical application. Recent attention has turned toward integrating multi-modal neuroimaging and machine learning to assist clinical disease diagnosis. Machine learning can obtain rules via automatically analyzing neuro-imaging data, and apply these rules to predict unknown data, to identify the brain regions highly correlated to the brain disease, to provide individual levels of diagnosis, and to detect pathophysiological mechanisms. This paper reviews the concrete steps of neuroimaging analysis based on machine learning and the application of machine learning in intelligent diagnosis and prediction of neuropsychiatric diseases. Finally, some new directions for future research are forecasted.
Abstract:
In recent years, the industry of wearable devices for pregnant women has been developing rapidly. Devices, ranging from fetal monitors to multi-functional health examination instruments, help in monitoring and managing maternal health indicators, such as fetal heart rate, blood glucose, and blood pressure, in and outside the hospital. Hospitals, maternal women and fetuses are bounded together by wearable devices for pregnant women in an unprecedented way. This article introduces wearable devices and their key technologies, and also investigates classical applications of such devices for pregnant women and the monitoring and management modes for the obstetrics department in the hospital. Finally, current major issues have also been analysed, with which we propose the way to deal.
Abstract:
Epithelioid trophoblastic tumor(ETT) is a rare type of gestational trophoblastic neoplasm(GTN), which originates from intermediate trophoblastic cells of chorionic-type. Its presentation is complex and diverse, making the diagnosis challenging. The diagnosis depends mainly on pathological diagnosis. Generally, ETT is a slow-growing malignant tumor, though some cases may have aggressive and fatal clinical courses involving multiple metastases, and demonstrate a greater malignant behavior compared to the placental site trophoblastic tumor(PSTT). Surgical resection remains the cornerstone of treatment. There is no standardized postoperative adjuvant chemotherapy regimen. Multifocal disease within the uterus, the presence of extra-uterine disease, an interval greater than 4 years from antecedent pregnancy, and high Ki-67 index may be markers of poor prognosis. At present, the behavior and clinical course of ETT are yet not fully understood. More cases and data of ETT should be collected to help guide diagnosis and management.
2018, 9(1): 36-41. doi: 10.3969/j.issn.1674-9081.2018.01.008
HTML (206) PDF(1116)
Abstract:
Original Contributions
Abstract:
  Objective   This study aimed to explore the feasibility and effectiveness of Smart Health Internet of Things System in managing therapeutic lifestyle changes(TLC)for the treatment of dyslipidemia.   Methods   Fifty-two patients with dyslipidemia were recruited in a three-month TLC program. Smart Health Internet of Things System (a wearable home-monitoring smart APP) was adopted to interfere the management of lifestyle. Parameters that were observed and investigated included behavioral habits (diet, exercise) and physical examination.   Results   With the intervention of Smart Health Internet of Things System, the parameters of the subjects' behavioral habits achieved total beneficial rates ranging from 5.5% to 49.9%, and the parameters of physiological examination achieved total beneficial rates ranging from -6.9% to 63.6%. The body mass index, waistline, hipline, waist-to-hip ratio, blood pressure, triglyceride, total cholesterol, and fasting plasma glucose all showed significant decreases after the TLC program(all P < 0.05).   Conclusions   Smart Health Internet of Things System has fairlygood feasibility and effectiveness on managing the TLC of dyslipidemic patients.
Abstract:
  Objective   This study was to investigate the antibiotic resistance profile of clinically important bacteria that World Health Organization(WHO) is concerned about, in Peking Union Medical College Hospital(PUMCH) from January 1, 2007 to December 31, 2016.   Methods   Disc diffusion test (Kirby-Bauer method) and automated systems were employed to detect antibiotic resistance. Data were analyzed by WHONET 5.6 software according to the 2016 edition of antbioticl susceptibility testing standards issued by the Clinical and Laboratory Standards Institute (CLSI) of the United States.   Results   A total of 46 168 non-duplicated main clinical isolates were collected in PUMCH from 2007 to 2016, including 6679 strains of Pseudomonas aeruginosa, 6422 strains of Acinetobacter baumannii, 24 001 strains of Enterobacteriaceae (11 046 strains of Escherichia coli and 6034 strains of Klebsiella pneumonia), 2358 strains of Enterococcus faecium, 6056 strains of Staphylococcus aureus, 652 strains of Streptococcus pneumonia, and 999 strains of Haemophilus influenza. Based on the surveillance data during the 10 years, we found that, the detection rate of the carbapenem-resistant P. aeruginosa decreased from 38.3% in 2007 to 22.4% in 2016; the detection rate of carbapenem-resistant A. baumannii increased from 52.8% in 2007 to 71.9% in 2016; the imipenem-resistance rate of K.pneumonia increased from 1.3% in 2007 to 14.4% in 2016. The vancomycin-resistance rate of E. faecium was 3.3%-5.8% in recent 5 years. The detection rate of Methicillin-resistant S. aureus (MRSA) decreased from 56.5% in 2007 to 27.0% in 2016. The detection rate of penicillin-insensitive S.pneumonia was 0.9%-6.4% in recent 2 years.   Conclusions   The prevalence of carbapenem-resistant K.pneumoniae and A.baumannii is still increasing. Carbapenem-resistant strains pose a huge challenge for anti-infection therapy.
Abstract:
  Objective   In this retrospective study, we investigated whether visceral obesity is associated with Fuhrman grade in middle-aged and elderly patients with renal cell carcinoma.   Methods   Medical records of 278 middle-aged and elderly patients who underwent radical or partial nephrectomy at Peking University People's Hospital from January 2009 to September 2014 were retrospectively reviewed. The quantities of visceral, subcutaneous and total adipose tissue were measured with pre-operative computed tomography scans at the level of umbilicus. Visceral obesity was indicated by the percentage of visceral adipose tissue. The logistic regression analysis was used to determine the risk factor of high grade disease(Fuhrman grade Ⅲ or Ⅳ).   Results   A total of 29(10.43%) tumors were classified as high-grade disease. Patients in high-grade group were found to have a higher percentage of visceral adipose tissue(P=0.022) and a larger tumor size(P=0.021). However, body mass index, total adipose tissue, visceral adipose tissue, and subcutaneous adipose tissue were comparable between low-grade and high-grade groups. The result of logistic analysis showed that visceral obesity was associated with high-grade tumors(OR=1.045, 95% CI:1.002-1.090, P=0.042). In the subgroup analysis, the percentage of visceral adipose tissue was associated with Fuhrman grade in advanced patients(OR=1.131, 95% CI:1.017-1.256, P=0.023) or patients with a larger tumor(OR=1.061, 95% CI:1.005-1.121, P=0.032), but not in patients with an organ-confined disease or a smaller tumor.   Conclusions   Visceral obesity was associated with higher Furhman grade in middle-aged and elderly patients with renal cell carcinoma, especially in patients with an advanced disease or a larger tumor. Vesceral obesity may be the risk factor of high-grade renal cell carcinoma.
Review
Abstract:
Melanoma is one of the most aggressive cutaneous malignancies with an increasing incidence in recent decades, especially in western countries. It is considered to be an incurable disease, and patients with metastatic melanoma survive no more than 5 years. Despite rapid improvement in chemotherapy and immunotherapy, such as anti-PD-1/PD-L1 treatment, the high frequency of drug resistance remains a difficult problem. Thus, recent research has shifted slightly to the field of biomarkers to achieve the more urgent goal of aiding in the diagnosis and predicting response and resistance to therapy. With the development of fascinating technologies in laboratory testing, numerous novel biomarkers have been identified, and some of them exhibit potential as therapeutic targets. In this review, we summarize the latest genetic and epigenetic biomarkers, discuss their role in the prediction of disease progression and response to therapies, and provide insights into potential targets for future therapies.
Continued Medical Education
Abstract:
Amyotrophic lateral sclerosis(ALS) is a progressive neurodegenerative disease, characterized by insidious onset, slow progression, and death due to respiratory failure. There is no cure for it. Recent studies show that many measures can prolong the survival of patients with ALS, and improve their quality of life. In order to treat ALS scientifically and rationally, this paper introduces its clinical classification, staging, treatment mode, and methods for evaluating and following up the progress of ALS disease.
Complicated and Rare Disease
Abstract:
The prevalence of autosomal dominant polycystic kidney disease(ADPKD) is 1‰ to 2‰, belonging to rare diseases. ADPKD is mainly manifested as gradually developing bilateral renal cysts, progressively increasing renal size, and gradually reduced renal function. PKD1 mutation accounts for about 81%; PKD2 mutation accounts for about 10.5% to 22%. Vasopressin and cyclic adenosine monophosphate(cAMP) signaling pathways play an important role in the development of ADPKD cysts. In recent years, the Mayo risk assessment model and predicting renal outcome in polycystic kidney disease(PROPKD) score, which are good prognostic models for ADPKD, have become important evidences for clinical decision-making. Tolvaptan, which inhibits cAMP pathway by antagonizing vasopressin receptor, has become the first specific treatment for ADPKD. Tolvaptan could effectively inhibit the growth of total renal size, and protect renal function. Long-term safety of the drug still needs further study.
Clinical Biobank
Abstract:
Due to the rapid development in research of translational medicine, the construction and application of clinical biobank has got increasing attention.The information system is the core of a biobank, which plays an important role in the construction and management of the biobank.The development of information system should be based on supporting the overall process of building, operation, management and service of the clinical biobank.Through establishing the data collecting process of sample information, data exchange system and security systems, all databases are integrated, and public network platforms are established.The Biobank information system could realize efficiently managing and comprehensively sharing all biospecimen, clinical medical records and molecular data.
Clinical Research and Evidence Based Medicine
Abstract:
The selection of outcome measures should be based on the study design, anticipated results, and available resources. Biological parameters, analysis of economic benefits, and life-quality assessment are commonly employed outcome measures. Outcome measures, observation method, and evaluation method correlate with the study design and affect the sample size estimation. The sample size should be determined according to various factors, including power, significance level, expected effect sizes and standard deviation in the case of continuous variables. In clinical research, rationally selecting outcome measures and scientifically estimating sample sizes might markedly improve the reliability and feasibility of the study results.