Objective To observe the value of machine learning(ML)models based on ultrasound radiomics for preoperatively distinguishing atypical parathyroid tumor(APT)/parathyroid carcinoma(PC)and parathyroid adenoma(PA).Methods Totally 330 primary hyperparathyroidism patients who underwent surgical treatments were retrospectively enrolled and categorized into APT/PC group(n=78)and PA group(n=252)according to surgical pathology and clinical follow-up results,also divided into training set(n=231)and test set(n=99)at the ratio of 7∶3.Based on preoperative ultrasound,545 radiomics features were extracted,and recursive feature elimination(RFE),Kruskal-Wallis or analysis of variance methods were used to screen the features,respectively.Support vector machine(SVM),linear discriminant analysis(LDA),least absolute shrinkage and selection operator logistic regression(LRLASSO),also random forest(RF)and decision tree(DT)algorithms were adopted to construct ML models for differentiating APT/PC and PA,respectively.Then the models were trained in training set,their performance were verified in test set,and a 5-fold cross-validation was adopted to screen out the better combinations.Results Compared with Kruskal-Wallis and analysis of variance methods,the distinguishing efficacy of SVM,LDA,LRLASSO,RF and DT models constructed based on features screened out using RFE method in training set(area under the curve[AUC]=0.870,0.878,0.850,0.847,1.000)and test set(AUC=0.856,0.842,0.827,0.847 and 0.704)were all relatively higher.In test set,the AUC of SVM,LDA,LRLASSO and RF models constructed based on the features screened out using RFE method(included 25,23,17 and 23 features)were all higher than that of DT model(8 features)(all P＜0.001).No significant difference of AUC was found between SVM,LRLASSO or RF models and LDA model(all P＞0.05).The AUC of SVM and RF models were higher than that of LRLASSO model(both P＜0.05),while of SVM and RF models were not significantly different(P＞0.05),indicating that SVM,LDA and RF models were better ones.Conclusion SVM,LDA,LRLASSO,RF and DT models based on ultrasound radiomics could effectively distinguish APT/PC and PA preoperatively,among which SVM,LDA and RF models had better diagnostic efficacy.

Objective To observe the value of machine learning(ML)models based on ultrasound radiomics for preoperatively distinguishing atypical parathyroid tumor(APT)/parathyroid carcinoma(PC)and parathyroid adenoma(PA).Methods Totally 330 primary hyperparathyroidism patients who underwent surgical treatments were retrospectively enrolled and categorized into APT/PC group(n=78)and PA group(n=252)according to surgical pathology and clinical follow-up results,also divided into training set(n=231)and test set(n=99)at the ratio of 7∶3.Based on preoperative ultrasound,545 radiomics features were extracted,and recursive feature elimination(RFE),Kruskal-Wallis or analysis of variance methods were used to screen the features,respectively.Support vector machine(SVM),linear discriminant analysis(LDA),least absolute shrinkage and selection operator logistic regression(LRLASSO),also random forest(RF)and decision tree(DT)algorithms were adopted to construct ML models for differentiating APT/PC and PA,respectively.Then the models were trained in training set,their performance were verified in test set,and a 5-fold cross-validation was adopted to screen out the better combinations.Results Compared with Kruskal-Wallis and analysis of variance methods,the distinguishing efficacy of SVM,LDA,LRLASSO,RF and DT models constructed based on features screened out using RFE method in training set(area under the curve[AUC]=0.870,0.878,0.850,0.847,1.000)and test set(AUC=0.856,0.842,0.827,0.847 and 0.704)were all relatively higher.In test set,the AUC of SVM,LDA,LRLASSO and RF models constructed based on the features screened out using RFE method(included 25,23,17 and 23 features)were all higher than that of DT model(8 features)(all P＜0.001).No significant difference of AUC was found between SVM,LRLASSO or RF models and LDA model(all P＞0.05).The AUC of SVM and RF models were higher than that of LRLASSO model(both P＜0.05),while of SVM and RF models were not significantly different(P＞0.05),indicating that SVM,LDA and RF models were better ones.Conclusion SVM,LDA,LRLASSO,RF and DT models based on ultrasound radiomics could effectively distinguish APT/PC and PA preoperatively,among which SVM,LDA and RF models had better diagnostic efficacy.

Purpose To investigate the clinicopathological features and prognosis of alpha-fetoprotein-producing colorectal carcinoma(AFPCRC).Methods 42 cases of AFPCRC from 2 012 colorectal carcinomas of preoperative serum AFP detected and surgically resected were identified.The clinicopathological data of AFPCRC and other 42 cases of conventional colorectal carcinoma exactly matched for age,gender,stage were also col-lected.Immunohistochemical EnVision method was performed to detect the expression of HER2,MMR,p53,AFP,Glypican3,and SALL4.Cases presenting HER2 2+were further analyzed by fluorescence in situ hybridization.Elastic staining was per-formed in cases with ambiguous extramural venous invasion.The clinicopathlogical features and prognosis between two groups were compared.Cases with AFPCRC were divided into high-AFP group and low-AFP group.The clinicopathological features and prognosis of the two groups were compared.Results AF-PCRC accounted for 2.1％(42/2 012)of colorectal carcinoma in the same period.The frequency of extramural vascular inva-sion and moderate/high grade of tumor budding of AFPCRC was 35.7％and 61.9％,while that of control group was 14.3％and 40.5％respectively.The 5-year survival rate of AFPCRC and control group was 66.8％and 85.1％respectively.The differ-ence of aforementioned clinicopathological features between 2 groups was significant(P＜0.05).The proportion of tumor in rectum in the high-AFP group was significantly higher than that in the low-AFP group(61.9％vs 23.8％,P＜0.05).Conclu-sion AFPCRC is a rare subset of colorectal carcinoma,which has a propensity for extramural vessel invasion,moderate-or high-grade of tumor budding and poor prognosis.

Prepubertal-type testicular neuroendocrine tumor is a rare neoplasm of low malignant potential, which is classified as germ cell tumors unrelated to germ cell neoplasia in situ, and needs to be differentiated from metastatic neuroendocrine tumor, postpubertal-type testicular neuroendocrine tumor, and testicular seminoma. The clinicopathological and molecular features of a case of prepubertal-type testicular neuroendocrine tumor were reported. The tumour cells were uniform in size and arranged in nested and insular pattern. The tumor was positive for CgA and Syn, and the Ki-67 index was less than 2% by immunostaining. Next-generation sequencing identified no variants of pathogenicity, potential pathogenicity or uncertain significance. The patient was followed without evidence of recurrence and metastasis 56 months after surgery.

In recent years,mechanical thrombectomy has been the most important research progress in the treatment of acute cerebral infarction,especially the positive results of five endovascular therapy studies in 2015 has rewritten its clinical guidelines.However,the focus of these studies was mainly on the small vessel infarction(SVI),and the inclusion criteria of these studies include the following aspects:ASPECTS ≥6 points,Alberta Stroke Program Early CT Score within 6 hours after stroke onset,the infarct volume＜70 mL within 6-24 hours after stroke onset,and the presence of image mismatch or the presence of mismatch between clinical condition and perfusion imaging.The above studies excluded patients with ASPECTS＜6 points or infarct volume ≥70 mL of large core infarction(LCI).With the continuous progress of the endovascular treatment of acute ischemic stroke(AIS),the mechanical thrombectomy therapy strategy has crossed from the"time window"to the"tissue window",meanwhile,the therapeutic goal of mechanical thrombectomy has also moved from treating SVI to a new era of treating LCI that has been a very hot topic recently.Whether endovascular treatment is beneficial for patients with LCI remains uncertain.This paper aims to make a comprehensive review concerning the relevant research progress in endovascular therapy for anterior circulation large core ischemic stroke,including the imaging determination and study inclusion criteria of LCI,the postoperative blood pressure management,and the factors influencing ineffective recanalization and prognosis.(J Intervent Radiol,2023,32:1256-1262)

Improper control of depth of anesthesia is not only detrimental to the rapid and stable recovery of anesthesia, but also affects the postoperative outcome of patients. Therefore, accurate control of anesthesia depth is an urgent clinical and scientific problem in the field of anesthesiology. At present, different algorithm models derived from electroencephalogram (EEG) signals are used to monitor the depth of anesthesia, but they cannot meet the requirements of anesthesiologists to accurately evaluate the depth of anesthesia. In recent years, the research on the mechanism and modulation of anesthesia-related neural network suggests that it has potential value as a method to monitor depth of anesthesia. Anesthesia-related neural networks mainly include sleep-wake circuit, thalamic-cortical circuit and corticocortical network. A thorough understanding of the neural network involved in the loss of consciousness caused by anesthesia will guide the depth of anesthesia monitoring more accurately and provide possibility for improving the quality of clinical anesthesia resuscitation.

Objective:To analyze the clinical characteristics and the risk factors of acute mechanical small bowel obstruction in adults.Methods:The clinical data of 487 adult patients with acute mechanical small bowel obstruction treated in Drum Tower Hospital Clinical College of Nanjing Medical University from June 2010 to December 2020 were retrospectively analyzed. There were 259 cases of strangulated obstruction (strangulation group) and 228 cases of non-strangulated obstruction (simple group). The cases in the strangulation group were confirmed by operation or pathology, the cases in the simple group were confirmed by non-operative therapy ( n=167) or operation ( n=61). The clinical data, including age, abdominal pain, vomiting, tenderness, rebound pain and muscle defense; the CT signs, including transitional zone, dilatation of intestinal loop, high density of intestinal wall; laboratory test results, including white blood cell count, lactate dehydrogenase (LDH) and C-reactive protein (CRP) were analyzed in both groups. Univariate and multivariate regression were used to identify risk factors of strangulated small bowel obstruction. Results:There were significant significances in gender (male vs.female: χ2=4.35, OR=0.67), rebound pain (χ2=170.98, OR=45.12), muscle tension(χ2=113.79, OR=29.32), American Society of Anesthesiologists score (ASA score)≥3 (χ2=12.94, OR=51.58), white blood cell count ( t=6.19, OR=1.14), LDH ( t=2.37, OR=1.00), CRP ( t=2.23, OR=1.01), albumin ( t=2.04, OR=0.97), mesenteric fluid sign (χ2=49.44, OR=5.40), increased bowel wall density (χ2=222.78, OR=62.66), bowel wall thickening sign (χ2=43.81, OR=3.49), ascites (χ2=237.29, OR=43.01), beak sign (χ2=231.50, OR=56.62), mesenteric fat stranding (χ2=242.65, OR=34.90), and stranding sign(χ2=224.79, OR=53.48) between strangulation group and simple group ( P<0.01). The multivariate regression analysis showed that mesenteric fluid sign ( OR=12.94), muscle tension ( OR=7.28), ascites ( OR=6.42), increased bowel wall density ( OR=4.30), bowel wall thickening sign ( OR=1.85), white blood cell count ( OR=1.14) and gender (male vs. female: OR=0.50) were risk factors of strangulated small bowel obstruction. Conclusion:In acute mechanical small bowel obstruction,for female patients presenting mesenteric fluid sign, muscle tension, ascites, increased bowel wall density, bowel wall thickening sign and increased white blood cell count, the strangulated obstruction is likely to occur.

Objective: To investigate the clinicopathological characteristics of cardial mixed adenoneuroendocrine carcinoma (MANEC) and analyze its prognostic factors. Methods: The retrospective and descriptive study was conducted. The clinicopathological data of 34 patients with primary cardial MANEC who were admitted to the Changzhou No.2 People′s Hospital of Nangjing Medical University from January 2008 to January 2018 were collected. There were 20 males and 14 females, aged from 39 to 81 years, with an average age of 66 years. All the 34 patients underwent resection of cardia cancer and postoperative pathological examination. Observation indicators: (1) surgery and treatment; (2) clinical manifestations and pathological conditions; (3) pathological examination of surgical resection specimens; (4) follow-up and survival; (5) analysis of prognostic factors. Follow-up using outpatient examination and telephone interview was conducted to detect the survival of patients and tumor recurrence and metastasis up to December 2018. Measurement data with normal distribution were represented as Mean±SD. Measurement data with skewed distribution were expressed as M (range). Count data were expressed as absolute numbers or percentages. Kaplan-Merier method was used to calculate the survival time and rate and draw the survival curve, and log-rank test was used for the survival analysis. Univariate and multivariate analyses were performed using the COX proportional risk model. Results: (1) Surgery and treatment: all the patients underwent surgery successfully and postoperative systemic chemotherapy based on cisplatin + etoposide. (2) Clinical manifestations and pathological conditions: patients had epigastric discomfort, abdominal pain and abdominal distension as the first symptoms. Of 34 patients, number of males, cases with age ≥60 years, cases with esophageal involvement, cases with stable microsatellite, cases with higher CA19-9, cases with elevated cancer embryonic antigen, cases of tumor pathologic TNM stage Ⅲ-Ⅳ, cases with tumor diameter ≥5 cm, cases with vascular tumor emboli, cases with positive lymph node metastasis, cases with nerve invasion were 20, 29, 31, 28, 4, 3, 29, 30, 27, 30, 29, respectively. (3) Pathological examination of surgical excision specimens: the masses of patients were mainly ulcer-type, with the diameter of 3.0-8.4 cm. Of the 34 patients, 1 had tumor infiltrated into submucosa, 5 infiltrated into muscle layer, 18 infiltrated into serosal layer, and 10 infiltrated into extra-serous fibrous adipose tissues. Microscopy examination showed that all tumors were composed of two components including adenocarcinoma and neuroendocrine carcinoma, and the two components accounted for more than 30%. Among adenocarcinoma components of the 34 patients, 14 were poorly differentiated tubular adenocarcinoma, 6 were mucinous adenocarcinoma, 6 were moderately differentiated tubular adenocarcinoma, 5 were low-adhesion carcinoma, 1 was highly differentiated tubular adenocarcinoma, and 2 were papillary adenocarcinoma. Among the neuroendocrine carcinoma components of the 34 patients, 10 were small cell type and 24 were large cell type. Of the 34 patients, 10 had adenocarcinoma and neuroendocrine carcinoma closely adjacent but not confused, and 24 had adenocarcinoma and neuroendocrine carcinoma cross-mixed. Immunohistochemistry examination of 34 patients showed that the components of neuroendocrine carcinoma were positive for synaptophysin, pheochromoin A and nerve cell adhesion molecule. The components of adenocarcinoma were positive for broad-spectrum cytokeratin, cytokeratin 8/18 and cytokeratin 7. (4) Follow-up and survival: 34 patients were followed up for 8.0-68.0 months, with a median time of 53.7 months. The 34 patients had survived for 21-49 months, with a median time of 35 months. The 1-, 3-, 5-year survival rates were 93.31%, 53.60%, and 20.62%. (5) Ananlysis of prognostic factors: results of univariate analysis showed that CA19-9, tumor diameter, intravascular tumor thrombus, tumor pathological TNM stage, lymph node metastasis, microsatellite detection, and histological classification were the related factors affecting the prognosis of patients with cardial MANEC (risk ratio =1.724, 0.327, 1.401, 1.612, 1.542, 1.876, 0.945, 95% confidence interval: 1.226-3.467, 0.218-0.776, 1.171-4.432, 0.694-4.054, 0.987-3.776, 1.217-4.341, 0.614-2.115, P<0.05). Results of multivariate analysis showed that the tumor pathological TNM stage Ⅲ-Ⅳ, positive lymph node metastasis, stable microsatellite, neuroendocrine carcinoma as the main histological classification were independent risk factors affecting the prognosis of patients with cardial MANEC (odds ratio=1.667, 1.441, 1.306, 3.501, 95% confidence interval: 1.013-4.915, 1.035-5.746, 1.006-6.213, 2.076-8.528, P<0.05). Conclusions Cardial MANEC is composed of two components including adenocarcinoma and neuroendocrine carcinoma, and the two components account for more than 30%. The tumors in the neuroendocrine cacinoma area present as solid nest-like pattern, rosettes-shaped or organ-like pattern, with high nuclear-to-plasma ratio and fine chromatin, and it is easy to see mitotic figures. Adenocarcinoma components are tubular adenocarcinoma, mucinous adenocarcinoma, papillary adenocarcinoma with various differentiation. The adenocarcinoma and neuroendocrine carcinoma components can be cross-mixed, and also can be closely adjacent but not confused. Tumor pathological TNM stage Ⅲ-Ⅳ, positive lymph node metastasis, stable microsatellite, neuroendocrine carcinoma as the main histological classification are independent risk factors affecting the prognosis of patients with cardial MANEC.

Objective: To investigate the clinical pathological feature of primary pulmonary sarcomatold carcinoma and to make a further understandine of the disease. Methods: Data including clinical manifesation, pathological findings, molecular detection and immunophenotyping with pathologically confirmed primary pulmonary sarcomatold carcinoma was retrospectively analyzed. Results: 15 patients with PPSC were identified(13 men and 2 women, age ranged 56-76 years, median age 66 years). The tumor were located in the left lobus superior(8 cases), lobus inferior(6 cases), and the right lobus medius(1 case). The main clinical symptoms was cough, sputum, bloody sputum, chest pain. Among the 13 males, 10 had smoking history of more than 30 years, and 2 females had no smoking history. All cases presented with a spheroid solid lung mass. All tumor showed mild enhancement similar to that of the surrounding musculature after contrast enhancement, and inhomogeneous central low-attenuation areas were seen in 15 patients. Pathological pattern: 6 cases spindle cell carcinoma, 4 cases pleomorphic carcinoma, 2 cases giant cells carcinoma, 2 cases carcinosarcoma, 1 case pulmonary blastoma. The tumors were composed of both carcinomatous and sarcomatous elements. Immunohistochemistry showed that CK was all positive, EMA was positive in 7, VIM was positive in 10 of 15 cases.10 patients were tested for common related genes of lung cancer, 4 patients had MET14 jump mutation, EGFR L858R gene mutation occurred in 3 cases, KARS G13D gene mutation in 2 cases, and BRAF V600E mutation in 1 case. All 15 patients underwent lobectomy, 13 underwent adjuvant chemotherapy, and 6 underwent local radiotherapy. Postoperative follow-up was 8 to 50 months, 3 cases were lost, and 4 cases were survival 3 years after the surgery. Conclusion Pulmonary sarcomatold carcinoma is a rare histologic subtype of non-small cell lung cancer. Compared with other NSCLC, there is no special clinical and imageing characteristics. Its definite diagnosis relies on postoperative pathological analysis and immunohistochemical staining, and PSC needs to be diatinguished from a variely of disease. PPSC is more aggressive and poor prognosis.

Objective To investigate the clinicopathological features of early gastric cancer ( EGC) with micropapillary pattern. Methods In 447 consecutive EGC patients at Changzhou No. 2 People′s Hospital admitted from January 2006 to December 2016, 8 ( 1. 8％) with micropapillary adenocarcinoma (≥5％) were enrolled in the observation group and the remaining 439 were included in the control group. Clinicopathologic features including age, gender, gross pattern, tumor location, size, invasion depth, lymphovascular invasion, lymphatic metastasis and pathological stage were analyzed and compared between the two groups. Results There were no significant differences between the two groups in age, gender, gross pattern, tumor location or size (all P>0. 05). The observation group showed significantly more frequent submucosal invasion ( 100. 0％, 8/8 ) , lymphovascular invasion ( 62. 5％, 5/8 ) , lymphatic metastasis (62. 5％, 5/8), and higher pathological stage (Ⅱ+Ⅲ 25. 0％, 2/8) compared with those in the control group, which were 52. 4％ ( 230/439 ) , 9. 1％ ( 40/439 ) , 13. 9％ ( 61/439 ) , and 5. 9％ ( 26/439 ) , respectively ( all P<0. 05 ) . Conclusion EGC with micropapillary pattern is rare with a high liability to lymphovascular invasion, lymphatic metastasis, and higher pathological stage, compared to EGC without micropapillary pattern.