Texture Analysis of Three-Dimensional MRI Images May Differentiate Borderline and Malignant Epithelial Ovarian Tumors

Accuracy of machine learning in the preoperative identification of ovarian borderline tumors: a meta-analysis

AIM

The objective of this study is to explore the diagnostic value of machine learning (ML) in borderline ovarian tumors through meta-analysis.

METHODS

Pubmed, Embase, Web of Science, and Cochrane Library databases were comprehensively retrieved from database inception untill February 16, 2023. The Prediction Model Risk of Bias Assessment Tool (PROBAST) was adopted to evaluate the risk of bias in the original studies. Sub-group analyses of ML were conducted according to clinical features and radiomics features. We separately discussed the discriminative value of ML for borderline vs benign and borderline vs malignant tumors.

RESULTS

Eighteen studies involving 12,778 subjects were included in our analysis. The modeling variables mainly consisted of radiomics features (n=13) and a small number of clinical features (n=5). When distinguishing between borderline and benign tumors, the ML model based on radiomic features achieved a c-index of 0.782 (95% CI: 0.732-0.831), sensitivity of 0.75 (95% CI: 0.67-0.82), and specificity of 0.75 (95% CI: 0.67-0.81) in the validation set. When distinguishing between borderline and malignant tumors, the ML model based on radiomic features achieved a c-index of 0.916 (95% CI: 0.891-0.940), sensitivity of 0.86 (95% CI: 0.78-0.91), and specificity of 0.88 (95% CI: 0.82-0.92) in the validation set. In addition, we analyzed the discriminatory ability of radiologists and found that their sensitivity was 0.26 (95% CI: 0.12-0.46) and specificity was 0.94 (95% CI: 0.90-0.97).

CONCLUSIONS

ML has tremendous potential in the preoperative diagnosis and differentiation of borderline ovarian tumors and may be more accurate than radiologists in diagnosing and differentiating borderline ovarian tumors.

Magnetic resonance imaging-based radiomics analysis of the differential diagnosis of ovarian clear cell carcinoma and endometrioid carcinoma: a retrospective study

PURPOSE

To retrospectively evaluate the diagnostic potential of magnetic resonance imaging (MRI)-based features and radiomics analysis (RA)-based features for discriminating ovarian clear cell carcinoma (CCC) from endometrioid carcinoma (EC).

MATERIALS AND METHODS

Thirty-five patients with 40 ECs and 42 patients with 43 CCCs who underwent pretherapeutic MRI examinations between 2011 and 2022 were enrolled. MRI-based features of the two groups were compared. RA-based features were extracted from the whole tumor volume on T2-weighted images (T2WI), contrast-enhanced T1-weighted images (cT1WI), and apparent diffusion coefficient (ADC) maps. The least absolute shrinkage and selection operator (LASSO) regression with tenfold cross-validation method was performed to select features. Logistic regression analysis was conducted to construct the discriminating models. Receiver operating characteristic curve (ROC) analyses were performed to predict CCC.

RESULTS

Four features with the highest absolute value of the LASSO algorithm were selected for the MRI-based, RA-based, and combined models: the ADC value, absence of thickening of the uterine endometrium, absence of peritoneal dissemination, and growth pattern of the solid component for the MRI-based model; Gray-Level Run Length Matrix (GLRLM) Long Run Low Gray-Level Emphasis (LRLGLE) on T2WI, spherical disproportion and Gray-Level Size Zone Matrix (GLSZM), Large Zone High Gray-Level Emphasis (LZHGE) on cT1WI, and GLSZM Normalized Gray-Level Nonuniformity (NGLN) on ADC map for the RA-based model; and the ADC value, spherical disproportion and GLSZM_LZHGE on cT1WI, and GLSZM_NGLN on ADC map for the combined model. Area under the ROC curves of those models were 0.895, 0.910, and 0.956. The diagnostic performance of the combined model was significantly superior (p = 0.02) to that of the MRI-based model. No significant differences were observed between the combined and RA-based models.

CONCLUSION

Conventional MRI-based analysis can effectively distinguish CCC from EC. The combination of RA-based features with MRI-based features may assist in differentiating between the two diseases.

Predicting MET exon 14 skipping mutation in pulmonary sarcomatoid carcinoma by whole-tumour texture analysis combined with clinical and conventional contrast-enhanced computed tomography features

Background

Pulmonary sarcomatoid carcinoma (PSC) is a rare, highly malignant type of non-small cell lung cancer (NSCLC) with a poor prognosis. Targeted drugs for MET exon 14 (METex14) skipping mutation can have considerable clinical benefits. This study aimed to predict METex14 skipping mutation in PSC patients by whole-tumour texture analysis combined with clinical and conventional contrast-enhanced computed tomography (CECT) features.

Methods

This retrospective study included 56 patients with PSC diagnosed by pathology. All patients underwent CECT before surgery or other treatment, and both targeted DNA- and RNA-based next-generation sequencing (NGS) were used to detect METex14 skipping mutation status. The patients were divided into two groups: METex14 skipping mutation and nonmutation groups. Overall, 1,316 texture features of the whole tumour were extracted. We also collected 12 clinical and 20 conventional CECT features. After dimensionality reduction and selection, predictive models were established by multivariate logistic regression analysis. Models were evaluated using the area under the curve (AUC), and the clinical utility of the model was assessed by decision curve analysis.

Results

METex14 skipping mutation was detected in 17.9% of PSCs. Mutations were found more frequently in those (I) who had smaller long- or short-axis diameters (P=0.02, P=0.01); (II) who had lower T stages (I, II) (P=0.02); and (III) with pseudocapsular or annular enhancement (P=0.03). The combined model based on the conventional and texture models yielded the best performance in predicting METex14 skipping mutation with the highest AUC (0.89). The conventional and texture models also had good performance (AUC =0.83 conventional; =0.88 texture).

Conclusions

Whole-tumour texture analysis combined with clinical and conventional CECT features may serve as a noninvasive tool to predict the METex14 skipping mutation status in PSC.

Delta-radiomics Analysis Based on Multi-phase Contrast-enhanced MRI to Predict Early Recurrence in Hepatocellular Carcinoma After Percutaneous Thermal Ablation

RATIONALE AND OBJECTIVES

It is critical to predict early recurrence (ER) after percutaneous thermal ablation (PTA) for hepatocellular carcinoma (HCC). We aimed to develop and validate a delta-radiomics nomogram based on multi-phase contrast-enhanced magnetic resonance imaging (MRI) to preoperatively predict ER of HCC after PTA.

MATERIALS AND METHODS

We retrospectively enrolled 164 patients with HCC and divided them into training, temporal validation, and other-scanner validation cohorts (n = 110, 29, and 25, respectively). The volumes of interest of the intratumoral and/or peritumoral regions were delineated on preoperative multi-phase MR images. Original radiomics features were extracted from each phase, and delta-radiomics features were calculated. Logistic regression was used to train the corresponding radiomics models. The clinical and radiological characteristics were evaluated and combined to establish a clinical-radiological model. A fusion model comprising the best radiomics scores and clinical-radiological risk factors was constructed and presented as a nomogram. The performance of each model was evaluated and recurrence-free survival (RFS) was assessed.

RESULTS

Child-Pugh grade B, high-risk tumor location, and an incomplete/absent tumor capsule were independent predictors of ER. The optimal radiomics model comprised 12 delta-radiomics features with areas under the curve (AUCs) of 0.834, 0.795, and 0.769 in the training, temporal validation, and other-scanner validation cohorts, respectively. The nomogram showed the best predictive performance with AUCs as 0.893, 0.854, and 0.827 in the three datasets. There was a statistically significant difference in RFS between the risk groups calculated using the delta-radiomics model and nomogram.

CONCLUSIONS

The nomogram combined with the delta-radiomic score and clinical-radiological risk factors could non-invasively predict ER of HCC after PTA.

Texture analysis using short-tau inversion recovery magnetic resonance images to differentiate squamous cell carcinoma of the gingiva from medication-related osteonecrosis of the jaw

OBJECTIVES

Despite the difficulty in distinguishing between squamous cell carcinoma (SCC) and medication-related osteonecrosis of the jaw (MRONJ) on the basis of medical imaging examinations, the two conditions have completely different treatment methods and prognoses. Therefore, differentiation of SCC from MRONJ on imaging examinations is very important. This study aimed to distinguish SCC from MRONJ by performing texture analysis using magnetic resonance imaging (MRI) short-tau inversion recovery images.

METHODS

This retrospective case-control study included 14 patients with SCC of the lower gingiva and 35 with MRONJ of the mandible who underwent MRI and computed tomography (CT) for suspected SCC or MRONJ. SCC was identified by histopathological examination of tissues excised during surgery. The radiomics features of SCC and MRONJ were analyzed using the open-access software MaZda version 3.3 (Technical University of Lodz, Institute of Electronics, Poland). CT was used to evaluate the presence or absence of qualitative findings (sclerosis, sequestrum, osteolysis, periosteal reaction, and cellulitis) of SCC and MRONJ.

RESULTS

Among the 19 texture features selected using MaZda feature-reduction methods, SCC of the gingiva and MRONJ of the mandible revealed differences in two histogram features, one absolute gradient feature, and 16 Gy level co-occurrence matrix features. In particular, the percentile, angular second moment, entropy, and difference entropy exhibited excellent diagnostic performance.

CONCLUSION

Non-contrast-enhanced MRI texture analysis revealed differences in texture parameters between mandibular SCC and mandibular MRONJ. MRI texture analysis can be a new noninvasive quantitative method for distinguishing between SCC and MRONJ.

Assessment of chemotherapy resistance changes in human colorectal cancer xenografts in rats based on MRI histogram features

Purpose

We investigated the value of magnetic resonance imaging (MRI) histogram features, a non-invasive method, in assessing the changes in chemoresistance of colorectal cancer xenografts in rats.

Methods

A total of 50 tumor-bearing mice with colorectal cancer were randomly divided into two groups: control group and 5-fluorouracil (5-FU) group. The MRI histogram characteristics and the expression levels of p53 protein and MRP1 were obtained at 24 h, 48 h, 72 h, 120 h, and 168 h after treatment.

Results

Sixty highly repeatable MRI histogram features were obtained. There were 16 MRI histogram parameters and MRP1 resistance protein differences between groups. At 24 h after treatment, the MRI histogram texture parameters of T2-weighted imaging (T2WI) images (10%, 90%, median, energy, and RootMeanSquared) and D images (10% and Range) were positively correlated with MRP1 (r = 0.925, p = 0.005). At 48 h after treatment, histogram texture parameters of apparent diffusion coefficient (ADC) images (Energy) were positively correlated with the presence of MRP1 resistance protein (r = 0.900, p = 0.037). There was no statistically significant difference between MRI histogram features and p53 protein expression level.

Conclusions

MRI histogram texture parameters based on T2WI, D, and ADC maps can help to predict the change of 5-FU resistance in colorectal cancer in the early stage and provide important reference significance for clinical treatment.

Radiomics nomogram for preoperative differentiation of early-stage serous borderline ovarian tumors and serous malignant ovarian tumors

Objectives

This study aimed to construct a radiomics nomogram and validate its performance in the preoperative differentiation between early-stage (I and II) serous borderline ovarian tumors (SBOTs) and serous malignant ovarian tumors (SMOTs).

Methods

Data were collected from 80 patients with early-stage SBOTs and 102 with early-stage SMOTs (training set: n = 127; validation set: n = 55). Univariate and multivariate analyses were performed to identify the independent clinicoradiological factors. A radiomics signature model was constructed using radiomics features extracted from multidetector computed tomography images of the venous phase, in which the least absolute shrinkage and selection operator regression was employed to lessen the dimensionality of the data and choose the radiomics features. A nomogram model was established by combining independent clinicoradiological factors with the radiomics signature. The performance of nomogram calibration, discrimination, and clinical usefulness was evaluated using training and validation sets.

Results

In terms of clinicoradiological characteristics, age (p = 0.001), the diameter of the solid component (p = 0.009), and human epididymis protein 4 level (p < 0.001) were identified as the independent risk factors of SMOT, for which the area under the curves (AUCs) were calculated to be 0.850 and 0.836 in the training and validation sets, respectively. Nine features were finally selected to construct the radiomics signature model, which exhibited AUCs of 0.879 and 0.826 for the training and validation sets, respectively. The nomogram model demonstrated considerable calibration and discrimination with AUCs of 0.940 and 0.909 for the training and validation sets, respectively. The nomogram model displayed more prominent clinical usefulness than the clinicoradiological and radiomics signature models according to the decision curve analysis.

Conclusions

The nomogram model can be employed as an individualized preoperative non-invasive tool for differentiating early-stage SBOTs from SMOTs.

A systematic review and meta-analysis of CT and MRI radiomics in ovarian cancer: methodological issues and clinical utility

OBJECTIVES

We aimed to present the state of the art of CT- and MRI-based radiomics in the context of ovarian cancer (OC), with a focus on the methodological quality of these studies and the clinical utility of these proposed radiomics models.

METHODS

Original articles investigating radiomics in OC published in PubMed, Embase, Web of Science, and the Cochrane Library between January 1, 2002, and January 6, 2023, were extracted. The methodological quality was evaluated using the radiomics quality score (RQS) and Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Pairwise correlation analyses were performed to compare the methodological quality, baseline information, and performance metrics. Additional meta-analyses of studies exploring differential diagnoses and prognostic prediction in patients with OC were performed separately.

RESULTS

Fifty-seven studies encompassing 11,693 patients were included. The mean RQS was 30.7% (range - 4 to 22); less than 25% of studies had a high risk of bias and applicability concerns in each domain of QUADAS-2. A high RQS was significantly associated with a low QUADAS-2 risk and recent publication year. Significantly higher performance metrics were observed in studies examining differential diagnosis; 16 such studies as well as 13 exploring prognostic prediction were included in a separate meta-analysis, which revealed diagnostic odds ratios of 25.76 (95% confidence interval (CI) 13.50-49.13) and 12.55 (95% CI 8.38-18.77), respectively.

CONCLUSION

Current evidence suggests that the methodological quality of OC-related radiomics studies is unsatisfactory. Radiomics analysis based on CT and MRI showed promising results in terms of differential diagnosis and prognostic prediction.

CRITICAL RELEVANCE STATEMENT

Radiomics analysis has potential clinical utility; however, shortcomings persist in existing studies in terms of reproducibility. We suggest that future radiomics studies should be more standardized to better bridge the gap between concepts and clinical applications.

Computed tomography-based texture assessment for the differentiation of benign, borderline, and early-stage malignant ovarian neoplasms

OBJECTIVE

This study was performed to examine the value of computed tomography-based texture assessment for characterizing different types of ovarian neoplasms.

METHODS

This retrospective study involved 225 patients with histopathologically confirmed ovarian tumors after surgical resection. Two different data sets of thick (5-mm) slices (during regular and portal venous phases) were analyzed. Raw data analysis, principal component analysis, linear discriminant analysis, and nonlinear discriminant analysis were performed to classify ovarian tumors. The radiologist's misclassification rate was compared with the MaZda (texture analysis software) findings. The results were validated with the neural network classifier. Receiver operating characteristic curves were analyzed to determine the performances of different parameters.

RESULTS

Nonlinear discriminant analysis had a lower misclassification rate than the other analyses. Thirty texture parameters significantly differed between the two groups. In the training set, WavEnLH_s-3 and WavEnHL_s-3 were the optimal texture features during the regular phase, while WavEnHH_s-4 and Kurtosis seemed to be the most discriminative features during the portal venous phase. In the validation test, benign versus malignant tumors and benign versus borderline lesions were well-distinguished.

CONCLUSIONS

Computed tomography-based texture features provide a useful imaging signature that may assist in differentiating benign, borderline, and early-stage ovarian cancer.

Nomograms of Combining MRI Multisequences Radiomics and Clinical Factors for Differentiating High-Grade From Low-Grade Serous Ovarian Carcinoma

Objective

To compare the performance of clinical factors, FS-T2WI, DWI, T1WI+C based radiomics and a combined clinic-radiomics model in predicting the type of serous ovarian carcinomas (SOCs).

Methods

In this retrospective analysis, 138 SOC patients were confirmed by histology. Significant clinical factors (P < 0.05, and with the area under the curve (AUC) > 0.7) was retained to establish a clinical model. The radiomics model included FS-T2WI, DWI, and T1WI+C, and also, a multisequence model was established. A total of 1,316 radiomics features of each sequence were extracted; the univariate and multivariate logistic regressions, cross-validations were performed to reduce valueless features and then radiomics signatures were developed. Nomogram models using clinical factors, combined with radiomics features, were developed in the training cohort. The predictive performance was validated by receiver operating characteristic curve (ROC) analysis and decision curve analysis (DCA). A stratified analysis was conducted to compare the differences between the combined radiomics model and the clinical model in identifying low- and high-grade SOC.

Results

The AUC of the clinical model and multisequence radiomics model in the training and validation cohorts was 0.90 and 0.89, 0.91 and 0.86, respectively. By incorporating clinical factors and multi-radiomics signature, the AUC of the radiomic-clinical nomogram in the training and validation cohorts was 0.98 and 0.95. The model comparison results show that the AUC of the combined model is higher than that of the uncombined models (P= 0.05, 0.002).

Conclusion

The nomogram models of clinical factors combined with MRI multisequence radiomics signatures can help identifying low- and high-grade SOCs and a provide a more comprehensive, effective method to evaluate preoperative risk stratification for SOCs.

Multi-Region Radiomic Analysis Based on Multi-Sequence MRI Can Preoperatively Predict Microvascular Invasion in Hepatocellular Carcinoma

Objectives

Microvascular invasion (MVI) affects the postoperative prognosis in hepatocellular carcinoma (HCC) patients; however, there remains a lack of reliable and effective tools for preoperative prediction of MVI. Radiomics has shown great potential in providing valuable information for tumor pathophysiology. We constructed and validated radiomics models with and without clinico-radiological factors to predict MVI.

Methods

One hundred and fifteen patients with pathologically confirmed HCC (training set: n = 80; validation set: n = 35) who underwent preoperative MRI were retrospectively recruited. Radiomics models based on multi-sequence MRI across various regions (including intratumoral and/or peritumoral areas) were built using four classification algorithms. A clinico-radiological model was constructed individually and combined with a radiomics model to generate a fusion model by multivariable logistic regression.

Results

Among the radiomics models, the model based on T2WI and arterial phase (T2WI-AP model) in the volume of the liver-HCC interface (VOI_interface) exhibited the best predictive power, with AUCs of 0.866 in the training group and 0.855 in the validation group. The clinico-radiological model exhibited good efficacy (AUC: 0.819 and 0.717, respectively). The fusion model showed excellent predictive ability (AUC: 0.915 and 0.868, respectively), outperforming both the clinico-radiological and the T2WI-AP models in the training and validation sets.

Conclusion

The fusion model of multi-region radiomics achieves an enhanced prediction of the individualized risk estimation of MVI in HCC patients. This may be a beneficial tool for clinicians to improve decision-making in personalized medicine.

A Nomogram Combining MRI Multisequence Radiomics and Clinical Factors for Predicting Recurrence of High-Grade Serous Ovarian Carcinoma

Objective

To develop a combined nomogram based on preoperative multimodal magnetic resonance imaging (mMRI) and clinical information for predicting recurrence in patients with high-grade serous ovarian carcinoma (HGSOC).

Methods

This retrospective study enrolled 141 patients with clinicopathologically confirmed HGSOC, including 65 patients with recurrence and 76 without recurrence. Radiomics features were extracted from the mMRI images (FS-T2WI, DWI, and T1WI+C). L1 regularization-based least absolute shrinkage and selection operator (LASSO) regression was performed to select radiomics features. A multivariate logistic regression analysis was used to build the classification models. A nomogram was established by incorporating clinical risk factors and radiomics Radscores. The area under the curve (AUC) of receiver operating characteristics, accuracy, and calibration curves were assessed to evaluate the performance of classification models and nomograms in discriminating recurrence. Kaplan-Meier survival analysis was used to evaluate the associations between the Radscore or clinical factors and disease-free survival (DFS).

Results

One clinical factor and seven radiomics signatures were ultimately selected to establish the predictive model for this study. The AUCs for identifying recurrence in the training and validation cohorts were 0.76 (0.68, 0.84) and 0.67 (0.53, 0.81) with the clinical model, 0.78 (0.71, 0.86) and 0.74 (0.61, 0.86) with the multiradiomics model, and 0.83 (0.77, 0.90) and 0.78 (0.65, 0.90) with the combined nomogram, respectively. The DFS was significantly shorter in the high-risk group than in the low-risk group.

Conclusion

By incorporating radiomics Radscores and clinical factors, we created a radiomics nomogram to preoperatively identify patients with HGSOC who have a high risk of recurrence, which may serve as a potential tool to guide personalized treatment.

The Feasibility of Combining ADC Value With Texture Analysis of T2WI, DWI and CE-T1WI to Preoperatively Predict the Expression Levels of Ki-67 and p53 of Endometrial Carcinoma

PURPOSE

To evaluate the feasibility of apparent diffusion coefficient (ADC) value combined with texture analysis (TA) in preoperatively predicting the expression levels of Ki-67 and p53 in endometrial carcinoma (EC) patients.

METHODS

Clinical, pathological and MRI findings of 110 EC patients were analyzed retrospectively. The expression levels of Ki-67 and p53 in EC tissues were detected by immunohistochemistry. ADC value was calculated, and three-dimensional (3D) texture features were measured on T₂-weighted images (T₂WI), diffusion-weighted images (DWI), and contrast-enhanced T₁-weighted images (CE-T₁WI). The univariate and multivariate logistic regression and cross-validations were used for the selection of texture features. The receiver operating characteristic (ROC) curve was performed to estimate the diagnostic efficiency of prediction model by the area under the curve (AUC) in the training and validation cohorts.

RESULTS

Significant differences of the ADC values were found in predicting Ki-67 and p53 (P=0.039, P=0.007). The AUC of the ADC value in predicting the expression levels of Ki-67 and p53 were 0.698, 0.853 and 0.626, 0.702 in the training and validation cohorts. The AUC of the TA model based on T₂WI, DWI, CE-T₁WI, and ADC value combined with T₂WI + DWI + CE-T₁WI in the training and validation cohorts for predicting the expression of Ki-67 were 0.741, 0.765, 0.733, 0.922 and 0.688, 0.691, 0.651, 0.938, respectively, and for predicting the expression of p53 were 0.763, 0.805, 0.781, 0.901 and 0.796, 0.713, 0.657, 0.922, respectively.

CONCLUSION

ADC values combined with TA are beneficial for predicting the expression levels of Ki-67 and p53 in EC patients before surgery, and they provide higher auxiliary diagnostic values for clinical application.

A radiomics nomogram based on contrast-enhanced MRI for preoperative prediction of macrotrabecular-massive hepatocellular carcinoma

BACKGROUND

Macrotrabecular-massive hepatocellular carcinoma (MTM-HCC) represents an aggressive form of hepatocellular carcinoma and is associated with poor survival outcomes.

AIMS

This study aimed to develop a radiomics nomogram based on contrast-enhanced MRI for preoperative prediction of MTM-HCC.

METHODS

This study enrolled 88 patients with histologically confirmed HCC, including 32 MTM-HCCs and 56 Non-MTM-HCCs. The clinical and gadobenate dimeglumine (Gd)-enhanced MRI features were retrospectively reviewed by two abdominal radiologists. The regions of interest (ROIs) on the largest cross-sectional image and two adjacent images of the tumor, from which radiomics features were extracted via MaZda software and a radiomics score (Rad-score) was calculated via Python software. Combined with the Rad-score and independent imaging factors, a radiomics nomogram was constructed using R software. Nomogram performance was estimated with calibration curve.

RESULTS

A total of eleven top weighted radiomics features were selected among five sequences of MR images. There was a significant difference in Rad-score between MTM-HCC and non-MTM-HCC patients (P < 0.001), where patients with MTM-HCC generally had higher Rad-scores (absolute value). After multivariate analysis, radiomics score (OR = 7.794, P < 0.001) and intratumor fat (OR = 9.963, P = 0.014) were determined as independent predictors associated with MTM-HCC. The area under the receiver operating characteristic (ROC) curve of the selected model was 0.813 (95% CI 0.714-0.912) and the optimal cutoff value was 0.60. The nomogram showed overall satisfactory prediction performance (AUC = 0.785 [95% CI 0.684-0.886]).

CONCLUSIONS

A contrast-enhanced MRI-based radiomics nomogram may be useful for preoperative prediction of MTM-HCC in primary HCC patients, allowing opportunity to improve the treatment course and patient outcomes.

Ultrasonography in the Diagnosis of Adnexal Lesions: The Role of Texture Analysis

The classic ultrasonographic differentiation between benign and malignant adnexal masses encounters several limitations. Ultrasonography-based texture analysis (USTA) offers a new perspective, but its role has been incompletely evaluated. This study aimed to further investigate USTA’s capacity in differentiating benign from malignant adnexal tumors, as well as comparing the workflow and the results with previously-published research. A total of 123 adnexal lesions (benign, 88; malignant, 35) were retrospectively included. The USTA was performed on dedicated software. By applying three reduction techniques, 23 features with the highest discriminatory potential were selected. The features’ ability to identify ovarian malignancies was evaluated through univariate, multivariate, and receiver operating characteristics analyses, and also by the use of the k-nearest neighbor (KNN) classifier. Three parameters were independent predictors for ovarian neoplasms (sum variance, and two variations of the sum of squares). Benign and malignant lesions were differentiated with 90.48% sensitivity and 93.1% specificity by the prediction model (which included the three independent predictors), and with 71.43–80% sensitivity and 87.5–89.77% specificity by the KNN classifier. The USTA shows statistically significant differences between the textures of the two groups, but it is unclear whether the parameters can reflect the true histopathological characteristics of adnexal lesions.