Multi-institutional validation of a radiomics signature for identification of postoperative progression of soft tissue sarcoma.

BACKGROUND: To develop a magnetic resonance imaging (MRI)-based radiomics signature for evaluating the risk of soft tissue sarcoma (STS) disease progression. METHODS: We retrospectively enrolled 335 patients with STS (training, validation, and The Cancer Imaging Archive sets, n = 168, n = 123, and n = 44, respectively) who underwent surgical resection. Regions of interest were manually delineated using two MRI sequences. Among 12 machine learning-predicted signatures, the best signature was selected, and its prediction score was inputted into Cox regression analysis to build the radiomics signature. A nomogram was created by combining the radiomics signature with a clinical model constructed using MRI and clinical features. Progression-free survival was analyzed in all patients. We assessed performance and clinical utility of the models with reference to the time-dependent receiver operating characteristic curve, area under the curve, concordance index, integrated Brier score, decision curve analysis. RESULTS: For the combined features subset, the minimum redundancy maximum relevance-least absolute shrinkage and selection operator regression algorithm + decision tree classifier had the best prediction performance. The radiomics signature based on the optimal machine learning-predicted signature, and built using Cox regression analysis, had greater prognostic capability and lower error than the nomogram and clinical model (concordance index, 0.758 and 0.812; area under the curve, 0.724 and 0.757; integrated Brier score, 0.080 and 0.143, in the validation and The Cancer Imaging Archive sets, respectively). The optimal cutoff was - 0.03 and cumulative risk rates were calculated. DATA CONCLUSION: To assess the risk of STS progression, the radiomics signature may have better prognostic power than a nomogram/clinical model.

Applying dynamic contrast-enhanced MRI tracer kinetic models to differentiate benign and malignant soft tissue tumors.

BACKGROUND: To explore the potential of different quantitative dynamic contrast-enhanced (qDCE)-MRI tracer kinetic (TK) models and qDCE parameters in discriminating benign from malignant soft tissue tumors (STTs). METHODS: This research included 92 patients (41females, 51 males; age range 16-86 years, mean age 51.24 years) with STTs. The qDCE parameters (Ktrans, Kep, Ve, Vp, F, PS, MTT and E) for regions of interest of STTs were estimated by using the following TK models: Tofts (TOFTS), Extended Tofts (EXTOFTS), adiabatic tissue homogeneity (ATH), conventional compartmental (CC), and distributed parameter (DP). We established a comprehensive model combining the morphologic features, time-signal intensity curve shape, and optimal qDCE parameters. The capacities to identify benign and malignant STTs was evaluated using the area under the curve (AUC), degree of accuracy, and the analysis of the decision curve. RESULTS: TOFTS-Ktrans, EXTOFTS-Ktrans, EXTOFTS-Vp, CC-Vp and DP-Vp demonstrated good diagnostic performance among the qDCE parameters. Compared with the other TK models, the DP model has a higher AUC and a greater level of accuracy. The comprehensive model (AUC, 0.936, 0.884-0.988) demonstrated superiority in discriminating benign and malignant STTs, outperforming the qDCE models (AUC, 0.899-0.915) and the traditional imaging model (AUC, 0.802, 0.712-0.891) alone. CONCLUSIONS: Various TK models successfully distinguish benign from malignant STTs. The comprehensive model is a noninvasive approach incorporating morphological imaging aspects and qDCE parameters, and shows significant potential for further development.

Integrating Tumor and Nodal Radiomics to Predict the Response to Neoadjuvant Chemotherapy and Recurrence Risk for Locally Advanced Gastric Cancer.

AIMS: To develop and evaluate machine learning models using tumor and nodal radiomics features for predicting the response to neoadjuvant chemotherapy (NAC) and recurrence risk in locally advanced gastric cancer (LAGC). BACKGROUND: Early and accurate response prediction is vital to stratify LAGC patients and select proper candidates for NAC. OBJECTIVE: A total of 218 patients with LAGC undergoing NAC followed by gastrectomy were enrolled in our study and were randomly divided into a training cohort (n = 153) and a validation cohort (n = 65). METHODS: We extracted 1316 radiomics features from the volume of interest of the primary lesion and maximal lymph node on venous phase CT images. We built 3 radiomics signatures for distinguishing good responders and poor responders based on tumor radiomics (TR), nodal radiomics (NR), and a combination of the two (TNR), respectively. A nomogram was then developed by integrating the radiomics signature and clinical factors. Kaplan- Meier survival curves were used to evaluate the prognostic value of the nomogram. RESULTS: The TNR signature achieved improved predictive value, with AUCs of 0.755 and 0.744 in the training and validation cohorts. Our proposed nomogram model (TNRN) showed a good performance for GR prediction in the prediction efficacy, calibration ability, and clinical benefit, with AUCs of 0.779 and 0.732 in the training and validation cohorts, superior to the clinical model. Moreover, the TNRN could accurately classify the patients into high-risk and low-risk groups in both training and validation cohorts with regard to postoperative recurrence and metastasis. CONCLUSION: The TNRN performed well in identifying good responders and provided valuable information for predicting progression-free survival time (PFS) in patients with LAGC who underwent NAC.

Preoperative CT-based deep learning radiomics model to predict lymph node metastasis and patient prognosis in bladder cancer: a two-center study.

OBJECTIVE: To establish a model for predicting lymph node metastasis in bladder cancer (BCa) patients. METHODS: We retroactively enrolled 239 patients who underwent three-phase CT and resection for BCa in two centers (training set, n = 185; external test set, n = 54). We reviewed the clinical characteristics and CT features to identify significant predictors to construct a clinical model. We extracted the hand-crafted radiomics features and deep learning features of the lesions. We used the Minimum Redundancy Maximum Relevance algorithm and the least absolute shrinkage and selection operator logistic regression algorithm to screen features. We used nine classifiers to establish the radiomics machine learning signatures. To compensate for the uneven distribution of the data, we used the synthetic minority over-sampling technique to retrain each machine-learning classifier. We constructed the combined model using the top-performing radiomics signature and clinical model, and finally presented as a nomogram. We evaluated the combined model's performance using the area under the receiver operating characteristic, accuracy, calibration curves, and decision curve analysis. We used the Kaplan-Meier survival curve to analyze the prognosis of BCa patients. RESULTS: The combined model incorporating radiomics signature and clinical model achieved an area under the receiver operating characteristic of 0.834 (95% CI: 0.659-1.000) for the external test set. The calibration curves and decision curve analysis demonstrated exceptional calibration and promising clinical use. The combined model showed good risk stratification performance for progression-free survival. CONCLUSION: The proposed CT-based combined model is effective and reliable for predicting lymph node status of BCa patients preoperatively. CRITICAL RELEVANCE STATEMENT: Bladder cancer is a type of urogenital cancer that has a high morbidity and mortality rate. Lymph node metastasis is an independent risk factor for death in bladder cancer patients. This study aimed to investigate the performance of a deep learning radiomics model for preoperatively predicting lymph node metastasis in bladder cancer patients. KEY POINTS: • Conventional imaging is not sufficiently accurate to determine lymph node status. • Deep learning radiomics model accurately predicted bladder cancer lymph node metastasis. • The proposed method showed satisfactory patient risk stratification for progression-free survival.

Deep learning signature based on multiphase enhanced CT for bladder cancer recurrence prediction: a multi-center study.

Background: Accurate stratification of recurrence risk for bladder cancer (BCa) is essential for precise individualized therapy. This study aimed to develop and validate a model for predicting the risk of recurrence in BCa patients postoperatively using 3-phase enhanced CT images. Methods: We retrospectively enrolled 874 BCa patients across four centers between January 2006 and December 2021. Patients from one center were used as training set, while the remaining patients went into the validation set. We trained a deep learning (DL) model based on convolutional neural networks using 3-phase enhanced CT images. The resulting prediction scores were entered into Cox regression analysis to obtain DL scores and construct a DL signature. DL scores and clinical features were then used as deep learning radioclinical signature. The predictive performance of DL signature was assessed according to concordance index and area under curve compared with deep learning radioclinical signature, clinical model and a widely accepted staging grading system. Recurrence-free survival (RFS) and overall survival (OS) were also predicted in order to further assess survival benefits. Findings: DL signature showed strong power for predicting recurrence (concordance index, 0.869; area under curve, 0.889) in validation set, outperforming other models and system. In addition, we divided RFS and OS into high and low risk groups by selecting appropriate cutoff values for DL signature, and calculated cumulative recurrence risk rates for both groups. Interpretation: Our proposed DL signature shows promising potential as clinical aid for predicting postoperative recurrence risk in BCa and for stratifying the risk of RFS and OS, which can be applied to guide personalized precision therapy. Funding: There are no sources of funding for this manuscript.

CT-based deep learning radiomics nomogram for the prediction of pathological grade in bladder cancer: a multicenter study.

BACKGROUND: To construct and assess a computed tomography (CT)-based deep learning radiomics nomogram (DLRN) for predicting the pathological grade of bladder cancer (BCa) preoperatively. METHODS: We retrospectively enrolled 688 patients with BCa (469 in the training cohort, 219 in the external test cohort) who underwent surgical resection. We extracted handcrafted radiomics (HCR) features and deep learning (DL) features from three-phase CT images (including corticomedullary-phase [C-phase], nephrographic-phase [N-phase] and excretory-phase [E-phase]). We constructed predictive models using 11 machine learning classifiers, and we developed a DLRN by combining the radiomic signature with clinical factors. We assessed performance and clinical utility of the models with reference to the area under the curve (AUC), calibration curve, and decision curve analysis (DCA). RESULTS: The support vector machine (SVM) classifier model based on HCR and DL combined features was the best radiomic signature, with AUC values of 0.953 and 0.943 in the training cohort and the external test cohort, respectively. The AUC values of the clinical model in the training cohort and the external test cohort were 0.752 and 0.745, respectively. DLRN performed well on both data cohorts (training cohort: AUC = 0.961; external test cohort: AUC = 0.947), and outperformed the clinical model and the optimal radiomic signature. CONCLUSION: The proposed CT-based DLRN showed good diagnostic capability in distinguishing between high and low grade BCa.

Preoperative contrast-enhanced CT-based radiomics nomogram for differentiating benign and malignant primary retroperitoneal tumors.

OBJECTIVES: This study evaluated the ability of a preoperative contrast-enhanced CT (CECT)-based radiomics nomogram to differentiate benign and malignant primary retroperitoneal tumors (PRT). METHODS: Images and data from 340 patients with pathologically confirmed PRT were randomly placed into training (n = 239) and validation sets (n = 101). Two radiologists independently analyzed all CT images and made measurements. Key characteristics were identified through least absolute shrinkage selection combined with four machine-learning classifiers (support vector machine, generalized linear model, random forest, and artificial neural network back propagation) to create a radiomics signature. Demographic data and CECT characteristics were analyzed to formulate a clinico-radiological model. Independent clinical variables were merged with the best-performing radiomics signature to develop a radiomics nomogram. The discrimination capacity and clinical value of three models were quantified by the area under the receiver operating characteristics (AUC), accuracy, and decision curve analysis. RESULTS: The radiomics nomogram was able to consistently differentiate between benign and malignant PRT in the training and validation datasets, with AUCs of 0.923 and 0.907, respectively. Decision curve analysis manifested that the nomogram achieved higher clinical net benefits than did separate use of the radiomics signature and clinico-radiological model. CONCLUSIONS: The preoperative nomogram is valuable for differentiating between benign and malignant PRT; it can also aid in treatment planning. KEY POINTS: • A noninvasive and accurate preoperative determination of benign and malignant PRT is crucial to identifying suitable treatments and predicting disease prognosis. • Associating the radiomics signature with clinical factors facilitates differentiation of malignant from benign PRT with improved diagnostic efficacy (AUC) and accuracy from 0.772 to 0.907 and from 0.723 to 0.842, respectively, compared with the clinico-radiological model alone. • For some PRT with anatomically special locations and when biopsy is extremely difficult and risky, a radiomics nomogram may provide a promising preoperative alternative for distinguishing benignity and malignancy.

Lumbar MR-based radiomics nomogram for detecting minimal residual disease in patients with multiple myeloma.

OBJECTIVES: Minimal residual disease (MRD) is a standard for assessing treatment response in multiple myeloma (MM). MRD negativity is considered to be the most powerful predictor of long-term good outcomes. This study aimed to develop and validate a radiomics nomogram based on magnetic resonance imaging (MRI) of the lumbar spine to detect MRD after MM treatment. METHODS: A total of 130 MM patients (55 MRD negative and 75 MRD positive) who had undergone MRD testing through next-generation flow cytometry were divided into a training set (n = 90) and a test set (n = 40). Radiomics features were extracted from lumbar spinal MRI (T1-weighted images and fat-suppressed T2-weighted images) by means of the minimum redundancy maximum relevance method and the least absolute shrinkage and selection operator algorithm. A radiomics signature model was constructed. A clinical model was established using demographic features. A radiomics nomogram incorporating the radiomics signature and independent clinical factor was developed using multivariate logistic regression analysis. RESULTS: Sixteen features were used to establish the radiomics signature. The radiomics nomogram included the radiomics signature and the independent clinical factor (free light chain ratio) and showed good performance in detecting the MRD status (area under the curve: 0.980 in the training set and 0.903 in the test set). CONCLUSIONS: The lumbar MRI-based radiomics nomogram showed good performance in detecting MRD status in MM patients after treatment, and it is helpful for clinical decision-making. KEY POINTS: • The presence or absence of minimal residual disease status has a strong predictive significance for the prognosis of patients with multiple myeloma. • A radiomics nomogram based on lumbar MRI is a potential and reliable tool for evaluating minimal residual disease status in MM.

An MRI-Based Radiomics Nomogram to Assess Recurrence Risk in Sinonasal Malignant Tumors.

BACKGROUND: Sinonasal malignant tumors (SNMTs) have a high recurrence risk, which is responsible for the poor prognosis of patients. Assessing recurrence risk in SNMT patients is a current problem. PURPOSE: To establish an MRI-based radiomics nomogram for assessing relapse risk in patients with SNMT. STUDY TYPE: Retrospective. POPULATION: A total of 143 patients with 68.5% females (development/validation set, 98/45 patients). FIELD STRENGTH/SEQUENCE: A 1.5-T and 3-T, fat-suppressed fast spin echo (FSE) T2-weighted imaging (FS-T2WI), FSE T1-weighted imaging (T1WI), and FSE contrast-enhanced T1WI (T1WI + C). ASSESSMENT: Three MRI sequences were used to manually delineate the region of interest. Three radiomics signatures (T1WI and FS-T2WI sequences, T1WI + C sequence, and three sequences combined) were built through dimensional reduction of high-dimensional features. The clinical model was built based on clinical and MRI features. The Ki-67-based and tumor-node-metastasis (TNM) model were established for comparison. The radiomics nomogram was built by combining the clinical model and best radiomics signature. The relapse-free survival analysis was used among 143 patients. STATISTICAL TESTS: The intraclass/interclass correlation coefficients, univariate/multivariate Cox regression analysis, least absolute shrinkage and selection operator Cox regression algorithm, concordance index (C index), area under the curve (AUC), integrated Brier score (IBS), DeLong test, Kaplan-Meier curve, log-rank test, optimal cutoff values. A P value < 0.05 was considered statistically significant. RESULTS: The T1 + C-based radiomics signature had best prognostic ability than the other two signatures (T1WI and FS-T2WI sequences, and three sequences combined). The radiomics nomogram had better prognostic ability and less error than the clinical model, Ki-67-based model, and TNM model (C index, 0.732; AUC, 0.765; IBS, 0.185 in the validation set). The cutoff values were 0.2 and 0.7 and then the cumulative risk rates were calculated. DATA CONCLUSION: A radiomics nomogram for assessing relapse risk in patients with SNMT may provide better prognostic ability than the clinical model, Ki-67-based model, and TNM model. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 5.

The predictive potential of contrast-enhanced computed tomography based radiomics in the preoperative staging of cT4 gastric cancer.

Background: The accuracy of preoperative staging is crucial for cT4 stage gastric cancer patients. The aim of this study was to develop the radiomics model and evaluate its predictive potential for differentiating preoperative cT4 stage gastric cancer patients into pT4b and no-pT4b patients. Methods: A multicenter retrospective analysis of 704 gastric cancer patients with preoperative contrast-enhanced computed tomography (CE-CT) staging cT4 between January 2008 and December 2021. These patients were divided into the training cohort (478 patients, the Affiliated Hospital of Qingdao University) and validation cohort (226 patients, the Weihai Wendeng District People's Hospital). According to the pathological stage of the tumors, the patients were divided into pT4b or no-pT4b stage. In the training cohort, the clinical and radiomics features were analyzed to construct the clinical model, tri-phase radiomics signatures and nomogram. Two kinds of methods were employed to achieve dimensionality reduction: (I) the least absolute shrinkage and selection operator (LASSO); and (II) the minimum redundancy maximum relevance (mRMR) algorithms. We utilized Logistic regression, support vector machine (SVM), Decision tree and Adaptive boosted tree (AdaBoost) algorithms as the machine learning classifiers. The nomogram was constructed on the clinical characteristics and the Rad-score. The performance of the models was evaluated by receiver operating characteristic (ROC) area under the curve (AUC), Decision Curve Analysis (DCA) curve and calibration curve. Results: The 345 pT4b and 359 no-pT4b stage patients were included in this study. In the validation cohort, the AUC of the clinical model was 0.793 (95% CI: 0.732-0.855). The tri-phase radiomics features combined with the SVM algorithm was the best radiomics signature with an AUC of 0.862 (95% CI: 0.812-0.912). The nomogram was the best predictive model of all with an AUC of 0.893 (95% CI: 0.834-0.927). In the training and validation cohorts, the calibration curves and DCA curves of the nomogram showed satisfactory result. Conclusions: CE-CT-based radiomics nomogram offers good accuracy and stability in differentiating preoperative cT4 stage gastric cancer patients into pT4b and non-pT4b stages, which has a great clinical relevance for selecting the course of treatment for cT4 stage gastric cancer patients.

Research on imbalance machine learning methods for MR[Formula: see text]WI soft tissue sarcoma data.

BACKGROUND: Soft tissue sarcoma is a rare and highly heterogeneous tumor in clinical practice. Pathological grading of the soft tissue sarcoma is a key factor in patient prognosis and treatment planning while the clinical data of soft tissue sarcoma are imbalanced. In this paper, we propose an effective solution to find the optimal imbalance machine learning model for predicting the classification of soft tissue sarcoma data. METHODS: In this paper, a large number of features are first obtained based on [Formula: see text]WI images using the radiomics methods.Then, we explore the methods of feature selection, sampling and classification, get 17 imbalance machine learning models based on the above features and performed extensive experiments to classify imbalanced soft tissue sarcoma data. Meanwhile, we used another dataset splitting method as well, which could improve the classification performance and verify the validity of the models. RESULTS: The experimental results show that the combination of extremely randomized trees (ERT) classification algorithm using SMOTETomek and the recursive feature elimination technique (RFE) performs best compared to other methods. The accuracy of RFE+STT+ERT is 81.57% , which is close to the accuracy of biopsy, and the accuracy is 95.69% when using another dataset splitting method. CONCLUSION: Preoperative predicting pathological grade of soft tissue sarcoma in an accurate and noninvasive manner is essential. Our proposed machine learning method (RFE+STT+ERT) can make a positive contribution to solving the imbalanced data classification problem, which can favorably support the development of personalized treatment plans for soft tissue sarcoma patients.

Deep Learning Radiomics Nomogram to Predict Lung Metastasis in Soft-Tissue Sarcoma: A Multi-Center Study.

Objectives: To build and evaluate a deep learning radiomics nomogram (DLRN) for preoperative prediction of lung metastasis (LM) status in patients with soft tissue sarcoma (STS). Methods: In total, 242 patients with STS (training set, n=116; external validation set, n=126) who underwent magnetic resonance imaging were retrospectively enrolled in this study. We identified independent predictors for LM-status and evaluated their performance. The minimum redundancy maximum relevance (mRMR) method and least absolute shrinkage and selection operator (LASSO) algorithm were adopted to screen radiomics features. Logistic regression, decision tree, random forest, support vector machine (SVM), and adaptive boosting classifiers were compared for their ability to predict LM. To overcome the imbalanced distribution of the LM data, we retrained each machine-learning classifier using the synthetic minority over-sampling technique (SMOTE). A DLRN combining the independent clinical predictors with the best performing radiomics prediction signature (mRMR+LASSO+SVM+SMOTE) was established. Area under the receiver operating characteristics curve (AUC), calibration curves, and decision curve analysis (DCA) were used to assess the performance and clinical applicability of the models. Result: Comparisons of the AUC values applied to the external validation set revealed that the DLRN model (AUC=0.833) showed better prediction performance than the clinical model (AUC=0.664) and radiomics model (AUC=0.799). The calibration curves indicated good calibration efficiency and the DCA showed the DLRN model to have greater clinical applicability than the other two models. Conclusion: The DLRN was shown to be an accurate and efficient tool for LM-status prediction in STS.

Multi-parametric MRI-based radiomics signature for preoperative prediction of Ki-67 proliferation status in sinonasal malignancies: a two-centre study.

OBJECTIVE: To assess the predictive ability of a multi-parametric MRI-based radiomics signature (RS) for the preoperative evaluation of Ki-67 proliferation status in sinonasal malignancies. METHODS: A total of 128 patients with sinonasal malignancies that underwent multi-parametric MRIs at two medical centres were retrospectively analysed. Data from one medical centre (n = 77) were used to develop the predictive models and data from the other medical centre (n = 51) constitute the test dataset. Clinical data and conventional MRI findings were reviewed to identify significant predictors. Radiomics features were determined using maximum relevance minimum redundancy and least absolute shrinkage and selection operator algorithms. Subsequently, RSs were established using a logistic regression (LR) algorithm. The predictive performance of RSs was assessed using calibration, decision curve analysis (DCA), accuracy, and AUC. RESULTS: No independent predictors of high Ki-67 proliferation were observed based on clinical data and conventional MRI findings. RS-T1, RS-T2, and RS-T1c (contrast enhancement T1WI) were established based on a single-parametric MRI. RS-Combined (combining T1WI, FS-T2WI, and T1c features) was developed based on multi-parametric MRI and achieved an AUC and accuracy of 0.852 (0.733-0.971) and 86.3%, respectively, on the test dataset. The calibration curve and DCA demonstrated an improved fitness and benefits in clinical practice. CONCLUSIONS: A multi-parametric MRI-based RS may be used as a non-invasive, dependable, and accurate tool for preoperative evaluation of the Ki-67 proliferation status to overcome the sampling bias in sinonasal malignancies. KEY POINTS: • Multi-parametric MRI-based radiomics signatures (RSs) are used to preoperatively evaluate the proliferation status of Ki-67 in sinonasal malignancies. • Radiomics features are determined using maximum relevance minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) algorithms. • RSs are established using a logistic regression (LR) algorithm.

Biodegradable Ru-Containing Polycarbonate Micelles for Photoinduced Anticancer Multitherapeutic Agent Delivery and Phototherapy Enhancement.

The lack of selectivity between tumor and healthy cells, along with inefficient reactive oxygen species production in solid tumors, are two major impediments to the development of anticancer Ru complexes. The development of photoinduced combination therapy based on biodegradable polymers that can be light activated in the "therapeutic window" would be beneficial for enhancing the therapeutic efficacy of Ru complexes. Herein, a biodegradable Ru-containing polymer (poly(DCARu)) is developed, in which two different therapeutics (the drug and the Ru complex) are rationally integrated and then conjugated to a diblock copolymer (MPEG-b-PMCC) containing hydrophilic poly(ethylene glycol) and cyano-functionalized polycarbonate with good degradability and biocompatibility. The polymer self-assembles into micelles with high drug loading capacity, which can be efficiently internalized into tumor cells. Red light induces the generation of singlet oxygen and the release of anticancer drug-Ru complex conjugates from poly(DCARu) micelles, hence inhibiting tumor cell growth. Furthermore, the phototherapy of polymer micelles demonstrates remarkable inhibition of tumor growth in vivo. Meanwhile, polymer micelles exhibit good biocompatibility with blood and healthy tissues, which opens up opportunities for multitherapeutic agent delivery and enhanced phototherapy.

Imaging features of lipoma arborescens.

BACKGROUND: Lipoma arborescens is characterized by the villous proliferation of the synovium and diffuse hyperplasia of fatty tissue in the subsynovial layer, almost always with a periarticular lesion. According to past articles, fewer cases have depicted the imaging features of lipoma arborescens. PURPOSE: To evaluate the computed tomography (CT) and magnetic resonance imaging (MRI) features of lipoma arborescens. MATERIAL AND METHODS: The imaging features of 15 patients with pathologically proven lipoma arborescens were retrospectively analyzed including lesion number, shape, location, size, margins, attenuation on CT, and signal intensity and enhancement patterns on MR images. RESULTS: All cases (n=15) showed joint or bursa effusion. The primary lesion attached to the bursa wall adjacent to the bone in 13 cases and to the lateral wall in two cases. CT shows a mass with fatty tissue attenuation values in the suprapatellar pouch (n=3) or subdeltoid bursa with an erosion of the humeral head (n=2), among them two cases showed branched slightly high density in the center of the fat density tissue. Fifteen cases on routine MRI display mass-like subsynovial mass with fatty tissue signal on all of the sequences and suppression of the signal on fat-suppression sequences. Among them, five lesions showed branched slightly low T1-weighted imaging, high proton density-weighted imaging, and T2-weighted imaging signals in the center. It showed the enhancement of the linear surface and the bursa wall in contrast-enhanced MRI (n=3). CONCLUSION: Lipoma arborescens have specific CT and MRI features that enable them to make a conclusive diagnosis of this rare condition, which helps the diagnosis before treatment.

Deep learning radiomic nomogram to predict recurrence in soft tissue sarcoma: a multi-institutional study.

OBJECTIVES: To evaluate the performance of a deep learning radiomic nomogram (DLRN) model at predicting tumor relapse in patients with soft tissue sarcomas (STS) who underwent surgical resection. METHODS: In total, 282 patients who underwent MRI and resection for STS at three independent centers were retrospectively enrolled. In addition, 113 of the 282 patients received additional contrast-enhanced MRI scans. We separated the participants into a development cohort and an external test cohort. The development cohort consisted of patients from one center and the external test cohort consisted of patients from two other centers. Two MRI-based DLRNs for prediction of tumor relapse after resection of STS were established. We universally tested the DLRNs and compared them with other prediction models constructed by using widespread adopted predictors (i.e., staging systems and Ki67) instead of radiomics features. RESULTS: The DLRN1 model incorporated plain MRI-based radiomics signature into the clinical data, and the DLRN2 model integrated radiomics signature extracted from plain and contrast-enhanced MRI with the clinical predictors. Across both study sets, the two MRI-based DLRNs had relatively better prognostic capability (C index ≥ 0.721 and median AUC ≥ 0.746; p < 0.05 compared with most other models and predictors) and less opportunity for prediction error (integrated Brier score ≤ 0.159). The decision curve analysis indicates that the DLRNs have greater benefits than staging systems, Ki67, and other models. We selected appropriate cutoff values for the DLRNs to divide STS recurrence into three risk strata (low, medium, and high) and calculated those groups' cumulative risk rates. CONCLUSION: The DLRNs were shown to be a reliable and externally validated tool for predicting STS recurrence by comparing with other prediction models. KEY POINTS: • The prediction of a high recurrence rate of STS before emergence of local recurrence can help to determine whether more active treatment should be implemented. • Two MRI-based DLRNs for prediction of tumor relapse were shown to be a reliable and externally validated tool for predicting STS recurrence. • We used the DLRNs to divide STS recurrence into three risk strata (low, medium, and high) to facilitate more targeted postoperative management in the clinic.

Radiomics Nomogram Based on Contrast-enhanced CT to Predict the Malignant Potential of Gastrointestinal Stromal Tumor: A Two-center Study.

RATIONALE AND OBJECTIVES: Contrast-enhanced computed tomography (CE-CT) was used to establish radiomics nomogram to evaluate the malignant potential of gastrointestinal stromal tumors (GISTs). MATERIALS AND METHODS: A total of 500 GIST patients were enrolled in this study and divided into training cohort (n = 346, our center) and validation cohort (n = 154, another center). Minimum redundancy maximum relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) algorithms were used to select the feature subset with the best discriminant features from the three phases image, and five classifiers were used to establish four radiomics signatures. Preoperative radiomics nomogram was constructed by adding the clinical features determined by multivariate logistic regression analysis. The performance of radiomics signatures and nomogram were evaluated by area under the curve (AUC) of the receiver operating characteristic (ROC). The calibration of nomogram was appraised by calibration curve. RESULTS: A total of 13 radiomic features were extracted from tri-phase combined CE-CT images. Tri-phase combined CE-CT features + Support Vector Machine (SVM) was the best combination at predicting the malignant potential of GIST, with an AUC of 0.895 (95% CI 0.858-0.931) in the training cohort and 0.847 (95% CI 0.778-0.917) in the validation cohort. The nomogram also had good calibration. In the training cohort and the validation cohort, preoperative radiomics nomogram reached AUCs of 0.927 and 0.905, respectively, which were higher than clinical. CONCLUSION: The radiomics nomogram had a good predictive effect and generalization on the malignant potential of GIST, which could effectively help guide preoperative clinical decision.

Preoperative CT-Based Deep Learning Model for Predicting Risk Stratification in Patients With Gastrointestinal Stromal Tumors.

OBJECTIVE: To develop and evaluate a deep learning model (DLM) for predicting the risk stratification of gastrointestinal stromal tumors (GISTs). METHODS: Preoperative contrast-enhanced CT images of 733 patients with GISTs were retrospectively obtained from two centers between January 2011 and June 2020. The datasets were split into training (n = 241), testing (n = 104), and external validation cohorts (n = 388). A DLM for predicting the risk stratification of GISTs was developed using a convolutional neural network and evaluated in the testing and external validation cohorts. The performance of the DLM was compared with that of radiomics model by using the area under the receiver operating characteristic curves (AUROCs) and the Obuchowski index. The attention area of the DLM was visualized as a heatmap by gradient-weighted class activation mapping. RESULTS: In the testing cohort, the DLM had AUROCs of 0.90 (95% confidence interval [CI]: 0.84, 0.96), 0.80 (95% CI: 0.72, 0.88), and 0.89 (95% CI: 0.83, 0.95) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. In the external validation cohort, the AUROCs of the DLM were 0.87 (95% CI: 0.83, 0.91), 0.64 (95% CI: 0.60, 0.68), and 0.85 (95% CI: 0.81, 0.89) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. The DLM (Obuchowski index: training, 0.84; external validation, 0.79) outperformed the radiomics model (Obuchowski index: training, 0.77; external validation, 0.77) for predicting risk stratification of GISTs. The relevant subregions were successfully highlighted with attention heatmap on the CT images for further clinical review. CONCLUSION: The DLM showed good performance for predicting the risk stratification of GISTs using CT images and achieved better performance than that of radiomics model.

Antidepressant-like effect of Ganoderma lucidum spore polysaccharide-peptide mediated by upregulation of prefrontal cortex brain-derived neurotrophic factor.

A 28-kDa polysaccharide-peptide (PGL) with antidepressant-like activities was isolated from spores of the mushroom Ganoderma lucidum. It was unadsorbed on DEAE-cellulose. Its internal amino acid sequences manifested pronounced similarity with proteins from the mushrooms Lentinula edodes and Agaricus bisporus. The monosaccharides present in 28-kDa PGL comprised predominantly of glucose (over 90%) and much fewer galactose, mannose residues, and other residues. PGL manifested antidepressant-like activities as follows. It enhanced viability and DNA content in corticosterone-injured PC12 cells(a cell line derived from a pheochromocytoma of the rat adrenal medulla with an embryonic origin from the neural crest containing a mixture of neuroblastic cells and eosinophilic cells) and reduced LDH release. A single acute PGL treatment shortened the duration of immobility of mice in both tail suspension and forced swimming tests. PGL treatment enhanced sucrose preference and shortened the duration of immobility in mice exposed to chronic unpredictable mild stress (CUMS). Chronic PGL treatment reversed the decline in mouse brain serotonin and norepinephrine levels but did not affect dopamine levels. PGL decreased serum corticosterone levels and increased BDNF mRNA and protein levels and increased synapsin I and PSD95 levels in the prefrontal cortex. This effect was completely blocked by pretreatment with the BDNF antagonist K252a, indicating that PGL increased synaptic proteins in a BDNF-dependent manner.Key points• An antidepressive polysaccharide-peptide PGL was isolated from G. lucidum spores.• PGL protected PC12 nerve cells from the toxicity of corticosterone.• PGL upregulated BDNF expression and influenced key factors in the prefrontal cortex.

A novel immune-related genes signature after bariatric surgery is histologically associated with non-alcoholic fatty liver disease.

Increasing evidence shows that immune-related genes (IRGs) play an important role in bariatric surgery (BS). We identified differentially expressed immune-related genes (DEIRGs) of adipose tissue after BS by analysing the two expression profiles of GEO (GSE59034 and GSE29409). Subsequently, enrichment analysis, GSEA and PPI networks were examined to identify the hub IRGs and related pathways. The performance of the signature was evaluated by area under the curve (AUC) of the receiver operating characteristic (ROC). CIBERSORT algorithm was used to evaluate the relative abundance of infiltrated immune cells.42 DEIRGs were found between the GSE59034 and GSE29409 datasets. The AUC of the signature was 0.904 and 0.865 in the GSE58979 and GSE48452, respectively. Interestingly, the signature also showed good performance in diagnosing non-alcoholic fatty liver disease (NAFLD) (AUC was 0.834 and 0.800, respectively). The number of neutrophils, macrophages M2, macrophages M0 and dendritic cells activated decreased significantly. After BS, the infiltration of T cells regulatory, monocytes, mast cells resting and plasma cells in adipose tissue increased. The novel proposed IRGs signature reveals the underlying immune mechanism of BS and is a promising biomarker for distinguishing the severity of NAFLD. This will provide new insights into strategies for treating obesity and NAFLD.