RESUMEN
Objective:To establish a machine learning model for the diagnosis of clinically significant prostate cancer based on transrectal contrast-enhanced ultrasound parameters and clinically relevant data.Methods:A retrospective analysis was performed on 151 patients in Chongqing University Cancer Hospital who underwent transrectal contrast-enhanced ultrasonography and transrectal ultrasound-guided needle biopsy from November 2018 to September 2021. The time intensity curve was drawn using VueBox software and 12 parameters such as rise time, peak time, average transit time, peak intensity, and rising slope were quantitatively analyzed. Age, total prostate-specific antigen, free prostate-specific antigen, free prostate-specific antigen ratio, volume, prostate-specific antigen density, and transrectal contrast-enhanced ultrasonography parameters, a total of 18 characteristic parameters, were analyzed and screened through relevant attribute values and information gain attribute values. The screening features were trained and tested by the machine learning single algorithm and integrated algorithm, and then the model was evaluated by the F1 value and the area under the ROC curve(AUC).Results:Using the related attribute value and the information gain attribute value, 12 variables and 5 variables were screened out respectively to establish a machine learning model. The model established by the ensemble algorithm was better than the single algorithm. For the two variable selection methods, the AUC (0.810 vs 0.789) and F1 values (0.748 vs 0.742) of the Bagging ensemble algorithm model, which basic algorithm was decision tree, were the highest, followed by Logistic regression and support vector machine(SVM) in order of AUC and F1 values.Conclusions:Based on transrectal contrast-enhanced ultrasound parameters and clinical data, the Bagging ensemble model based on decision tree has the best performance in diagnosing clinically significant prostate cancer.
RESUMEN
Objective: To compare the detection rate of prostate cancer (PCa) of transrectal and transperineal biopsy. Methods: Totally 128 patients with first time diagnosed PCa were enrolled, including 62 patients in transrectal approach group and 66 patients in transperineal approach group. The detection rate of PCa and clinically significant prostate cancer (CsPCa) of biopsy in patients with different total prostate specific antigen (TPSA) levels were compared between two approaches. Among them, 104 cases (42 cases in transrectal approach group and 62 cases in transperineal approach group) underwent CEUS on the basis of routine ultrasound examination, while 83 cases (28 cases in transrectal approach group and 55 cases in transperineal approach group) underwent targeted puncture. The detection rate of PCa and CsPCa of systematic puncture and targeted puncture were compared between the two groups. Results: The detection rate of PCa and CsPCa was 35.48% (22/62) and 25.81% (16/62) in transrectal approach group, 42.42% (28/66) and 28.79% (19/66) in transperineal approach group (P=0.471, 0.676). There was no significant difference in detection rates of PCa and CsPCa between the two approaches at different TPSA levels (all P>0.05). There was no significant difference in PCa detection rates (35.48% [22/62] vs 40.91% [27/66]; P=0.587), positive needle number/total times (14.25% [106/744] vs 14.52% [115/792]; P=0.879) nor CsPCa detection rates (25.81% [16/62] vs 28.79% [19/66]; P=0.676) between the two approaches. There were significant differences in PCa detection rates (35.71% [10/28] vs 14.55% [8/55]; P=0.002) and positive needle number/total times (30.77% [24/78] vs 6.76% [10/148]; P<0.001) between the two approaches. Conclusion: There is no significant difference in the detection rate of PCa and CsPCa between transrectal and transperineal approaches in ultrasound-guided prostate biopsy. CEUS can guide targeted prostate biopsy. Choosing the same patient position and anatomical section as CEUS can improve the detection rate of prostate cancer of biopsy.
RESUMEN
Objective To evaluate the clinical value of high-resolution ultrasound for predicting the extracapsular extension (ETE) of papillary thyroid cancer (PTC).Methods A total of 151 nodules with PTC in 116 patients confirmed by surgery and pathology were enrolled.The ultrasonographic characteristics were retrospectively analyzed,and the percentage of the perimeter of the nodule that abutted the thyroid capsule (A) were observed.Taking A≥1％,A≥25 ％ and A≥ 50％ as cutoff value respectively,the ETE of PTC was predicted.Results Of the 151 nodules with PTC,ETE presented in 59 (59/151,39.07％),while no ETE (non-ETE) was observed in 92 (92/151,60.93％).Nodules in the thyroid parenchyma (A0) which could be observed with ultrasound was 84 (84/151,55.63％),and 1％≤A＜25％,25％≤A＜50％andA≥50％ was 25 (25/151,16.56％),15 (15/151,9.93％),and 27 (27/151,17.88％),respectively.There were statistically significant differences of nodules between the non-ETE and ETE when A≥1％,A≥25 ％ and A≥50％ (all P ＜0.01).Taking A=50％ as the cutoff value for predicting the presence of ETE,the sensitivity,specificity,accuracy,positive predictive value and negative predictive value was 37.29％ (22/59),94.57％ (87/92),72.19％ (109/151),81.48％ (22/27) and 70.16％ (87/124),respectively.Conclusion High frequency ultrasonography can effectively predict whether PTC has been invaded,which provides important reference information for preselected surgical approach and prognosis of patients.
RESUMEN
Objective To explore ultrasound performances and misdiagnostic causes of aggressive fibromatosis (AF).Methods Ultrasound performance and misdiagnostic causes of 45 patients (47 tumors) with AF confirmed by histopathology were analyzed retrospectively.Results Of 45 patients,ultrasound showed lesions of varying size,and the maximum diameter of 36 masses (36/47,76.60％) were more than 3 cm;37 masses (37/47,78.72％) were irregular pale leaf;45 masses (45/47,95.74％) had no complete capsule;45 masses (45/47,95.74％) were shown as interogeneous internal echo mixed with hyperechoic area in the hypoechoic internal;29 cases were diagnosed correctly,16 cases were misdiagnosed,the diagnostic accordance rate was 64.44％ (29/46).Conclusion AF has certain ultrasonic characteristics.Combined with the patient's medical history and physical sign,ultrasound can significantly improve the preoperative detection rate and diagnostic accuracy of AF.