2018年 · 第52卷第10期

According to the latest data from the Mental Health Center of the Chinese Center for Disease Control and Prevention, as of the end of 2017, the number of schizophrenia patients in my country exceeded 6.4 million, but its pathogenesis is not fully understood, which seriously affects human health and brings a heavy burden to individuals, families and society^[1]。 The diagnosis of schizophrenia mainly depends on the complaint of the patient and the experience of the psychiatrist. However, some patients conceal their disease and describe their symptoms inaccurately, so there is some subjectivity in the assessment using clinical scales. Moreover, the clinical symptoms of schizophrenia patients themselves vary greatly, the genetic and environmental risk factors are different, the treatment effects are different, and the disease is prone to recurrence and spread. Clinical diagnosis and treatment at home and abroad urgently need to transform from traditional doctors' subjective judgment to accurate and quantitative analysis mode, and find objective biological markers for early diagnosis, risk assessment and treatment plan choice of schizophrenia.

MRI is one of the main imaging methods for prostate diseases. At present, MRI equipment, scanning methods, examination sequences and report contents are different in different medical institutions. It is necessary to formulate MRI examination and diagnosis standards of prostate to guide and unify the diagnosis process of prostate diseases. After discussion by many relevant experts in China, the 7th issue of this journal in 2014 published "Consensus on MR Examination and Diagnosis of Prostate Cancer"^[1]。 According to this consensus, many domestic medical institutions have improved the previous MRI examination and diagnosis process of prostate, which has played a positive role in promoting the standardized development of prostate MRI in China. In recent years, imaging technology and examination methods have developed rapidly, a large number of new research results have been published at home and abroad, and the diagnostic concept has been constantly innovated. Most of the previous ideas are supported by the latest evidence, but there are some previous ideas that remain controversial. After discussion by many relevant scholars in China, we updated the content of the previous edition of "Consensus on MR Examination and Diagnosis of Prostate Cancer" on the basis of the 2014 edition consensus, combined with the latest norms and guidelines, research results and concepts at home and abroad^[2], formed the "Consensus on MRI Examination and Diagnosis of Prostate Cancer (Second Edition)", hoping to play a better role in promoting future work.

As a cutting-edge science and technology that deeply affects the development of today's society, artificial intelligence (AI) has also been widely used in medicine, and is the core technology to promote the transformation and progress of new models of intelligent medicine in the future. Medical imaging (the scope of this article mainly refers to Radiology) occupies the first place in the current hospital data volume, and naturally becomes the main object of AI medical application^[1]。

Patients with schizophrenia are faced with the severe situation of high recurrence rate, high disability rate, high poverty rate and low treatment rate, which brings a heavy burden to individuals, families and society^[1]。 The clinical symptoms of schizophrenia are highly heterogeneous, the evolution of the course of the disease is variable, and the response to drug therapy is mixed. Therefore, revealing the characteristic biomarkers of schizophrenia and achieving early precise intervention is one of the hot spots of global medical research^[2]。 In recent years, the use of neuroimaging techniques such as MRI, electroencephalography and magnetoencephalography to explore the structure and function of the brain is one of the most important breakthroughs in the field of brain science. Compared with other brain imaging technologies, MRI technology has the advantages of non-invasiveness, no radiation, good soft tissue contrast, high spatial resolution, and multiple repeated operations. It plays a key role in the study of brain mechanisms of mental disorders, and gives birth to psychiatric imaging (psychoradiology), a new interdisciplinary subject^[3]It provides an opportunity for the final realization of auxiliary diagnosis and guidance treatment of mental disorders through medical imaging.

prostate cancer (PCa) is the most common malignant tumor in men, accounting for the second largest cancer death among men in the United States^[1]。 Although the prevalence of PCa in China is much lower than that in European and American countries, it has shown a significant upward trend in recent years, and its age-standardized mortality rate has also increased year by year^[2]。 MRI is currently recognized as the best method of prostate examination, and it is of great value in early diagnosis, localization, staging, invasiveness assessment and follow-up of PCa^[3,4,5]。 Because the MRI examination methods of prostate in different medical institutions in China are different, and the diagnostic level is uneven, relevant domestic experts have summarized and compiled relevant consensus^[6,7]And has played a standardizing and guiding role in MRI examination and diagnosis of PCa in clinical work. The consensus has been published for more than 3 years. During this period, after the improvement of the prostate imaging reporting and data system (PI-RADS) version and the release of many new ideas and research results at home and abroad, the editorial board of Chinese Journal of Radiology specially organized experts to reorganize the examination requirements and diagnostic standards of prostate MRI. After many discussions by relevant experts in China, the original consensus was revised, and striving to achieve standardized examination and diagnostic standards in this field.

papillary renal cell carcinoma (PRCC) is a common subtype of renal cancer, accounting for 14% to 17%^[1]Conventional PRCC is mainly divided into type I and type II. In 2005, Lefevre et al.^[2]PRCC with eosinophilic characteristics was first reported. With the deepening of pathological research, it was gradually classified as the third subtype of PRCC and called oncocytic papillary renal cell carcinoma (OPRCC)^[3]。 The author found only one case report to discuss its performance on CT^[4]。 The MRI manifestations of OPRCC were reported in order to improve the understanding of the imaging manifestations of OPRCC.

The patient was a 56-year-old male. He was admitted to hospital on October 18, 2017 due to choking and coughing with hiccups for half a month after eating. The patient had decreased appetite. Physical examination showed no obvious abnormalities. Coagulation indexes and liver function were mildly abnormal in laboratory.

A 50-year-old male suffered from neck and shoulder pain for 5 years, muscle atrophy and weakness of both upper limbs gradually aggravated with instability of holding objects for 9 months, and weakness and numbness of limbs for 2 months. The above symptoms were significantly aggravated in the past week, and he was admitted to hospital on June 16, 2014. The patient had a history of gout for 20 years, and was hospitalized with gout in the left wrist joint diagnosed in a foreign hospital 1 year ago. Physical examination: shallow hypoesthesia below the bilateral mastoid plane, obvious below the bilateral knee joint. Laboratory tests: uric acid 542 μ mol/L (normal reference value 208~428 μ mol/L), C-reactive protein 125 mg/L (normal reference value 0~8 mg/L).

A 50-year-old female developed binocular vision loss without obvious trigger 2 years ago, which was not paid attention to; I developed dizziness 3 months ago and visited our hospital on November 16, 2017. Physical examination: The mind is clear and the eyes are clear, the pupils on both sides are equal size and round, about 3.0 mm in diameter, sensitive to light reflection, and the visual acuity of both eyes is reduced in rough measurement. Physical examination of the remaining nervous system showed no obvious abnormalities. Growth hormone (GH) 10.595 ng/ml (normal 0.010~3.607 ng/ml). MR plain scan and enhancement of sellar area: sellar enlargement, sellar floor sinking, nodular mass-like abnormal signal shadow on the left side of the sellar, the boundary with the pituitary gland is not clear, the size is about 1.8 cm ×2.0 cm ×1.4 cm, T₁WI showed a slightly lower signal (Figure 1), T₂WI showed a heterogeneous slightly high signal (Figure 2), the enhancement scan was not uniformly enhanced and the degree of enhancement was lower than that of the normal pituitary (Figure 3), the lesion involved the left cavernous sinus, immediately adjacent to the left internal carotid artery, the pituitary stem was slightly deviated to the left, the optic chiasma ran well, and the impression diagnosis was pituitary adenoma.

In 2015, the proposal of the Precision Medicine Plan in the United States kicked off the global precision medicine. Precision medicine is essentially personalized medicine, which is a precise diagnosis and treatment system with precise diagnosis as the premise and precise treatment as the purpose. Imaging medicine is an indispensable technical means in clinical diagnosis and treatment, which provides important information for clinical diagnosis and treatment. Precision medicine, imaging first. To achieve accurate imaging, it is necessary to break through the traditional imaging medicine mode based on morphology and semi-quantitative analysis. However, imagomics adopts high-throughput feature extraction algorithm to quantitatively analyze images, deeply mine and analyze the additional information hidden in images, and make the most efficient use of imaging examination results, thus laying a solid foundation for accurate diagnosis and treatment of clinical diseases. The author will review the research progress of radiomics in accurate diagnosis and treatment of clinical diseases.

CNS diseases are complex and diverse, and the prognosis is often poor, so early diagnosis is very important, especially CNS tumors, which are malignant nervous system tumors with the highest mortality and disability rate. Correct diagnosis can avoid unnecessary surgery^[1]。 Traditional medical imaging diagnosis still stays on the change of anatomical shape, and there is still a large part of effective information to be extracted and utilized. radiomics is dedicated to improving image analysis and can extract quantitative features of a large number (more than 200 kinds) of medical images. It is a new method of medical imaging diagnosis^[2]。 The application and research of radiomics in various systems of the whole body are gradually being carried out. The author will review the current research of radiomics in central nervous system, discuss the progress and challenges of radiomics in central nervous system research, in order to improve the diagnosis and differentiation of central nervous system diseases and provide evidence-based support for precision medicine.