Objective To compare the MRI manifestations and characteristics of ≤ 3 cm nonepithelial hepatic angiomyolipomas (HAML)and small hepatocellular carcinoma (SHCC),then improve the preoperative diagnostic accuracy.Methods A retrospective analysis of 20 patients ≤3 cm nonepithelial HAML and 26 cases of SHCC,confirmed by clinical pathology,with both in clinical data and MRI characteristics.Results ≤3 cm nonepithelial HAML commonly occurs in women;The enhancement patterns“wash in and wash out”in 1 1 cases,6 cases “wash in but slow out”,3 cases with delayed enhancement;Mature adipose tissue found in 5 cases, 3 cases has pseudocapsule enhancemen in delayed phase,13 cases can see central vessels;While SHCC often occurs in men,with“wash in and wash out”enhancement pattern in 23 cases,1 7 cases with pseudocapsule enhancemen in delayed phase,12 cases can see central vessels in lesions.ADC values for SHCC was significantly lower than that for ≤3 cm nonepithelial HAML,the ADC values of SHCC significantly lower than the surrounding liver parenchyma;The ADC values between ≤3 cm nonepithelial and liver parenchyma around has no significant differences.ADC values of liver parenchyma with liver cirrhosis was lower than that without cirrhosis.Conclusion The existence of mature adipose tissue,the MRI enhancement pattern and the value of ADC can help to distinguish between ≤3 cm nonepithelial HAML and SHCC,then improve the preoperative diagnostic accuracy.

Objective:To explore the clinicopathological characteristics and prognostic factors of bladder squamous cell carcinoma (SqCC)and bladder adenocarcinoma.Methods:A retrospective analysis of the clinical data of 107 patients with nonurothelial carcinoma of the bladder admitted to the First Affiliated Hospital of Zhengzhou University from October 2011 to January 2019 was performed. Among the 107 patients, 78 were males and 29 were females(ratio 2.69∶1), and the median age of onset was 62.0 years. According to histological types, patients were divided into SqCC group, urachal adenocarcinoma group and non-urachal adenocarcinoma group. There were 55 cases in the SqCC group, including 40 males and 15 females. Their mean age was 69.0(58.0, 75.0) years. 14 cases had the history of smoking. The clinic manifestation included hematuria in 35 cases, bladder irritation in 13 cases, dysuria in 2 cases and pain in 5 cases .Tumors located at the anterior and posterior walls in 18 cases, at the lateral wall in 27 cases, at the triangular area in 8 cases and at the apical wall in 2 cases. The average diameter of tumor was 4.5(3.0, 6.0) cm. 37 cases suffered with single tumor and 18 cases suffered with multiple tumors. The surgical method was radical cystectomy in 38 cases, partial cystectomy in 4 cases, TURBT in 9 cases, interventional surgery in 2 cases, and no operation in 2 cases. There were 20 cases in the urachal adenocarcinoma group, including 14 males and 6 females; age 53.5(43.5, 57.8) years; 6 cases with a history of smoking. The clinic manifestation included hematuria in 16 cases, bladder irritation in 1 case, pain in 2 cases and asymptomatic in 1 case. Tumors located at anterior and posterior walls in 4 cases and at apical wall in 16. The average diameter of tumor was 3.0(2.0, 4.8) cm. Single tumor was present in 18 cases, multiple tumors were present in 2 cases. The surgical method was partial cystectomy in 16 cases, radical cystectomy in 1 case, TURBT in 1 case and no operation in 2 cases. There were 32 cases in the non-urachal adenocarcinoma group, including 24 males and 8 females. Their mean age was 55.0(46.3, 70.8) years.11 cases had a history of smoking. The clinic manifestation included hematuria in 16 cases, bladder irritation in 3 cases, dysuria in one case and pain in 7 cases. Tumor located at anterior and posterior walls in 17 cases, at lateral wall in 7 cases, at triangular area in 5 cases and at apical wall in 3 cases. The average diameter of tumor was 3.6(2.0, 4.5) cm. 23 cases suffered with single tumor, 9 cases suffered with multiple tumors.The surgical method was radical cystectomy in 11 cases, partial cystectomy in 9 cases, TURBT in 9 cases , and no operation in 3 cases. The preoperative data of the three groups of tissue types were compared, the differences of age of onset, tumor diameter, tumor location, reason for treatment, operation method ( P<0.05)among the 3 groups were statistically significant. The clinicopathological characteristics and prognosis of the 3 groups of tissue types were compared, and the Cox proportional regression risk model was used to analyze the clinical factors affecting the prognosis. Results:91 patients were followed up, the overall follow-up rate was 85.1%, the median follow-up time was 26(7, 48) months. The 3-year and 5-year overall survival rates were 54.1% and 42.2%, respectively. In the SqCC group, 11 cases received chemotherapy; 3 cases received postoperative radiotherapy; 12 cases received postoperative perfusion.10 cases had recurrence; 17 cases had lymph node metastasis; 19 cases had distant metastasis; 5 cases were pT x in pT stage, 36 cases were pT 1-pT 2, 14 cases were pT 3-pT 4. 19 cases had unknown tumor differentiation, 4 cases had well differentiated, 24 cases had moderately differentiated and 8 cases had poorly differentiated. In the urachal adenocarcinoma group, 7 cases received chemotherapy, 3 cases had recurrence, 2 cases had lymph node metastasis, 2 cases had distant metastasis; 1 case was pT x in pT stage, 16 cases were pT 1-pT 2, 3 cases were pT 3-pT 4. 9 cases had unknown tumor differentiation, 3 cases had well differentiated, and 5 cases had moderately differentiated, 3 cases had poorly differentiated. In the urachal adenocarcinoma group, 3 cases received chemotherapy, 1 case received postradiotherapy, 11 cases received postoperative perfusion; 10 cases had recurrence. 5 cases had lymph node metastasis, 4 cases had distant metastasis, 6 cases were pT x, 21 cases were pT 1-pT 2 and 5 cases were pT 3-pT 4. 14 cases had unknown tumor differentiation, 8 cases had moderately differentiated and 10 cases had poorly differentiated.The postoperative general data of the three groups of tissue types were compared. There was statistically significant difference between whether there was postoperative perfusion and whether there was distant metastasis (all P<0.05). The univariate analysis results showed gender, age, surgical method, lymph node metastasis, distant metastasis, pT staging, tumor differentiation and histological type were risk factors that affect the prognosis(all P<0.05). Cox multivariate analysis showed that women ( HR=2.604, P=0.032) and distant metastases ( HR=2.571, P=0.026) were independent risk factors affecting the prognosis of patients. Conclusions:SqCC and adenocarcinoma are clinically rare and have poor prognosis. They often present with hematuria. Surgical treatment is the first choice. Radical cystectomy is the first choice for SqCC and non-urachal adenocarcinoma, and extended partial cystectomy is the first choice for urachal adenocarcinoma. Female and distant metastasis are independent risk factors affecting the prognosis of patients.

Objective To summarize the MRI characteristics of fat-containing hepatocellular carcinoma (HCC).Methods The retrospective cross-sectional study was conducted.The clinicopathological data of 18 patients with fat-containing HCC who were admitted to the Eastern Hepatobiliary Surgery Hospital of the Second Military Medical University between January 2015 and April 2016 were collected.All patients underwent MRI examination.Observation indicators:(1) imaging examination;(2) pathological features;(3) treatment and follow-up.Follow-up using outpatient imaging examination was performed to detect tumor recurrence once at 1 month postoperatively and once every 3 months postoperatively up to December 2016.Measurement data with normal distribution were represented as-x±s.Results (1) Imaging examination:plain scans of MRI:of 18 patients,lesions of 9,7 and 2 patients were respectively located in left,right and caudate lobes of liver,and lesions were round-like,with a diameter of 1.6-9.0 cm.Tumors were single,and tumor thrombosis in the hepatic vein was seen in 1 patient.T2WI on fat-suppression of 15 and 3 patients showed respectively high signals and iso-signal;MRI examination showed that out-phase signal on T1WI was variously decreased than in-phase signal,showing focal or diffuse signal decreasing;lipid phase on T1WI showed diffuse distribution in 10 patients,patchy fat-containing regions were scattered within the peritumoral zone in 7 patients,and lipid of 1 patient was located in the central area.Dynamic contrast enhancement MRI showed that fast-in and fast-out,fast-in and slow-out,and slow-in and slow-out types were detected in 16,1 and 1 patients,respectively.Sixteen patients had capsule-like enhancement in delayed phase.Diffusion weighted imaging (DWI) showed high signals in 10 patients,slightly high signals in 6 patients and iso-signal in 2 patients.Apparent diffusion coefficient (ADC) value was (1.19± 0.20)×10-3 mm2/s,including (1.37±0.20) × 10-3 mm2/s in lipid-rich type,(1.17±0.13) × 10-3 mm2/s in clear cell type and (1.11±0.17)× 10-3 mm2/s in thick beam type.Of 18 patients,2 were preoperatively misdiagnosed as hepatic angiomyolipoma and 1 as hepatic adenoma.(2) Pathological features:pathological types of 18 patients with fat-containing HCC:5 and 13 patients were respectively in lipid-rich and non-lipid-rich types (4 in clear cell type and 9 in thick beam type);steatosis tumor cells,rich and translucent cytoplasms and large vacuolated lipid droplets were seen under the microscope;steatosis tumor cells were diffused or scattered within the lesions,some of which showed plaque-like performance of different sizes.Degree of microvascular invasion:grade M0,M1 and M2 were found in 12,3 (2 in thick beam type and 1 in clear cell type) and 3 (2 in thick beam type and 1 in lipid-rich type) patients,respectively.(3) Treatment and follow-up:18 patients underwent surgery and were followed up for 6-13 months,with a median time of 8 months.During the follow-up,5 patients in lipid-rich type and 4 in clear cell type had no recurrence;among 9 patients in thick beam type,1 of 4 with microvascular invasion had no recurrence,2 and 1 of 4 with microvascular invasion had respectively recurrence at 3 months postoperatively and 4 months postoperatively,and 5 without microvascular invasion had no recurrence.The lesions of 3 patients with recurrence is single,3 patients received local minimally invasive therapy and continuous follow-up.Conclusion MRI characteristics of fat-containing HCC include that decreased out-phase signal on T1WI compared with in-phase signal,high signal on DWI,low ADC value,fast-in and fast-out enhancement and capsule-like enhancement in delayed phase.

Objective To investigate the diagnostic value of diffusion weighted imaging(DWI)-based radiomics model to identify small hepatocellular carcinoma(＜2 cm)(SHCC)and dysplastic nodule(DN)in the background of hepatitis cirrhosis.Methods A total of 93 cases SHCC and 25 cases with DN with complete enhanced MRI images and surgically pathologically confirmed were collected retrospectively.Chi-square test was performed to analyze the signal characteristics of DWI and enhanced triphasic MRI images between the two groups.ITK-SNAP was used to draw the region of interest(ROI)on DWI,and FAE software was applied for extraction,selec-tion,and construction of support vector machine(SVM)models(dividing into training set and test set according to the ratio of 7∶3).The diagnostic performance of model was evaluated by receiver operating characteristic(ROC)curve.Results There were statisti-cally significant differences in enhanced triphasic MRI and DWI between SHCC and DN(P＜0.05).The area under the curve(AUC)of the DWI-SVM model training set was 0.936,and sensitivity,specificity and accuracy was 95.4％,88.2％and 93.9％,respec-tively,and the AUC of the test set was 0.911,and sensitivity,specificity and accuracy was 85.7％,87.5％and 86.1％,respectively,which were all significantly better than the diagnostic efficacy of DWI(AUC=0.720).Conclusion DWI-SVM model with signifi-cantly higher AUC and specificity can effectively differentiate SHCC from DN in the background of hepatitis cirrhosis.

The integration of artificial intelligence into the medical field is developing rapidly and has achieved ground-breaking advances in the diagnosis, treatment, and efficacy evaluation of imaging medicine. This article reviews the research advances in artificial intelligence in imaging diagnosis of hepatocellular carcinoma and its performance in evaluating treatment outcome and predicting prognosis in combination with clinical features and looks forward to how artificial intelligence can be better used in the practice of hepatocellular carcinoma imaging in the era of growing clinical needs and rapid advances in diagnosis and treatment techniques.