Deep learning to assist composition classification and thyroid solid nodule diagnosis: a multicenter diagnostic study.

OBJECTIVES: This study aimed to propose a deep learning (DL)-based framework for identifying the composition of thyroid nodules and assessing their malignancy risk. METHODS: We conducted a retrospective multicenter study using ultrasound images from four hospitals. Convolutional neural network (CNN) models were constructed to classify ultrasound images of thyroid nodules into solid and non-solid, as well as benign and malignant. A total of 11,201 images of 6784 nodules were used for training, validation, and testing. The area under the receiver-operating characteristic curve (AUC) was employed as the primary evaluation index. RESULTS: The models had AUCs higher than 0.91 in the benign and malignant grading of solid thyroid nodules, with the Inception-ResNet AUC being the highest at 0.94. In the test set, the best algorithm for identifying benign and malignant thyroid nodules had a sensitivity of 0.88, and a specificity of 0.86. In the human vs. DL test set, the best algorithm had a sensitivity of 0.93, and a specificity of 0.86. The Inception-ResNet model performed better than the senior physicians (p < 0.001). The sensitivity and specificity of the optimal model based on the external test set were 0.90 and 0.75, respectively. CONCLUSIONS: This research demonstrates that CNNs can assist thyroid nodule diagnosis and reduce the rate of unnecessary fine-needle aspiration (FNA). CLINICAL RELEVANCE STATEMENT: High-resolution ultrasound has led to increased detection of thyroid nodules. This results in unnecessary fine-needle aspiration and anxiety for patients whose nodules are benign. Deep learning can solve these problems to some extent. KEY POINTS: • Thyroid solid nodules have a high probability of malignancy. • Our models can improve the differentiation between benign and malignant solid thyroid nodules. • The differential performance of one model was superior to that of senior radiologists. Applying this could reduce the rate of unnecessary fine-needle aspiration of solid thyroid nodules.

Prognosis and influencing factors of follicular thyroid cancer.

OBJECTIVES: Follicular thyroid cancer (FTC) is prone to distant metastasis, and patients with distant metastasis often have poor prognosis. In this study, the impact of metastasis and other relevant factors on the prognosis of follicular thyroid carcinoma was examined. METHODS: This was a retrospective study. Data were obtained from Zhejiang Cancer Hospital, Sun Yat-sen University Cancer Center and Hangzhou First People's Hospital affiliated with Zhejiang University School of Medicine, from January 2009 to June 2021 for 153 FTC patients. The patients were assigned into three groups according to their distant metastasis: distant metastasis at initial diagnosis (M1), distant metastasis during follow-up (M2), and no evidence of distant metastasis over the course of the study (M0). Data were collected and summarized on clinical data, laboratory parameters, imaging features, postoperative pathologic subtypes, and metastases. The Cox proportional hazard model was used to perform the univariate and multivariate analysis. Kaplan-Meier curves were used to evaluate cancer-specific survival (CSS). RESULTS: Based on metastasis, the patients were assigned into three groups, including 31 in the M1 group, 15 in the M2 group, and 107 in the M0 group. These individuals were followed up for an average of 5.9 years, and the group included 46 patients with distant metastasis (31 confirmed at diagnosis and 15 found during follow-up). Univariate Cox regression analysis showed that age, Hashimoto's thyroiditis (HT), surgery method, postoperative adjuvant therapy, histologic subtype, nodule size, calcification, TSH, and distant metastasis all impacted prognosis. Multivariate Cox regression analysis suggested that histologic subtype (widely invasive; HR: 7.440; 95% CI: 3.083, 17.954; p < 0.001), nodule size (≥40 mm; HR: 8.622; 95% CI: 3.181, 23.369; p < 0.001) and distant metastasis (positive; HR: 6.727; 95% CI: 2.488, 18.186; p < 0.001) were independent risk factors affecting the prognosis of follicular thyroid cancer. CONCLUSIONS: Histologic subtype, nodule size, and distant metastasis are important risk factors for the prognosis of follicular thyroid cancer. Patients with metastatic follicular thyroid cancer have a poor prognosis, especially with metastasis at the time of initial diagnosis. As a result, this group of patients requires individualized treatment and closer follow-up.

AI diagnosis of Bethesda category IV thyroid nodules.

Thyroid nodules are a common disease, and fine needle aspiration cytology (FNAC) is the primary method to assess their malignancy. For the diagnosis of follicular thyroid nodules, however, FNAC has limitations. FNAC can classify them only as Bethesda IV nodules, leaving their exact malignant status and pathological type undetermined. This imprecise diagnosis creates difficulties in selecting the follow-up treatment. In this retrospective study, we collected ultrasound (US) image data of Bethesda IV thyroid nodules from 2006 to 2022 from five hospitals. Then, US image-based artificial intelligence (AI) models were trained to identify the specific category of Bethesda IV thyroid nodules. We tested the models using two independent datasets, and the best AI model achieved an area under the curve (AUC) between 0.90 and 0.95, demonstrating its potential value for clinical application. Our research findings indicate that AI could change the diagnosis and management process of Bethesda IV thyroid nodules.

Combined conventional ultrasonography with clinicopathological features to predict axillary status after neoadjuvant therapy for breast cancer: A case-control study.

OBJECTIVES: This study aimed to evaluate the value of a model combining conventional ultrasonography and clinicopathologic features for predicting axillary status after neoadjuvant therapy in breast cancer. METHODS: This retrospective study included 329 patients with lymph node-positive who underwent neoadjuvant systemic treatment (NST) from June 2019 to March 2022. Ultrasound and clinicopathological characteristics of breast lesions and axillary lymph nodes were analyzed before and after NST. The diagnostic efficacy of ultrasound, clinicopathological characteristics, and combined model were evaluated using multivariate logistic regression and receiver operator characteristic curve (ROC) analyses. RESULTS: The area under ROC (AUC) for the ability of the combined model to predict the axillary pathological complete response (pCR) after NST was 0.882, that diagnostic effectiveness was significantly better than that of the clinicopathological model (AUC of 0.807) and the ultrasound feature model (AUC of 0.795). In addition, eight features were screened as independent predictors of axillary pCR, including clinical N stage, ERBB2 status, Ki-67, and after NST the maximum diameter reduction rate and margins of breast lesions, the short diameter, cortical thickness, and fatty hilum of lymph nodes. CONCLUSIONS: The combined model constructed from ultrasound and clinicopathological features for predicting axillary pCR has favorable diagnostic results, which allowed more accurate identification of BC patients who had received axillary pCR after NST. ADVANCES IN KNOWLEDGE: A combined model incorporated ultrasound and clinicopathological characteristics of breast lesions and axillary lymph nodes demonstrated favorable performance in evaluating axillary pCR preoperatively and non-invasively.

Ultrasound and clinicopathological characteristics-based model for prediction of pathologic response to neoadjuvant chemotherapy in HER2-positive breast cancer: a case-control study.

BACKGROUND: The objective of this study was to develop a model combining ultrasound (US) and clinicopathological characteristics to predict the pathologic response to neoadjuvant chemotherapy (NACT) in human epidermal growth factor receptor 2 (HER2)-positive breast cancer. MATERIALS AND METHODS: This is a retrospective study that included 248 patients with HER2-positive breast cancer who underwent NACT from March 2018 to March 2022. US and clinicopathological characteristics were collected from all patients in this study, and characteristics obtained using univariate analysis at p < 0.1 were subjected to multivariate analysis and then the conventional US and clinicopathological characteristics independently associated with pathologic complete response (pCR) from the analysis were used to develop US models, clinicopathological models, and their combined models by the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, and specificity to assess their predictive efficacy. RESULTS: The combined model had an AUC of 0.808, a sensitivity of 88.72%, a specificity of 60.87%, and an accuracy of 75.81% in predicting pCR of HER2-positive breast cancer after NACT, which was significantly better than the clinicopathological model (AUC = 0.656) and the US model (AUC = 0.769). In addition, six characteristics were screened as independent predictors, namely the Clinical T stage, Clinical N stage, PR status, posterior acoustic, margin, and calcification. CONCLUSION: The conventional US combined with clinicopathological characteristics to construct a combined model has a good diagnostic effect in predicting pCR in HER2-positive breast cancer and is expected to be a useful tool to assist clinicians in effectively determining the efficacy of NACT in HER2-positive breast cancer patients.

A new xAI framework with feature explainability for tumors decision-making in Ultrasound data: comparing with Grad-CAM.

BACKGROUND AND OBJECTIVE: The value of implementing artificial intelligence (AI) on ultrasound screening for thyroid cancer has been acknowledged, with numerous early studies confirming AI might help physicians acquire more accurate diagnoses. However, the black box nature of AI's decision-making process makes it difficult for users to grasp the foundation of AI's predictions. Furthermore, explainability is not only related to AI performance, but also responsibility and risk in medical diagnosis. In this paper, we offer Explainer, an intrinsically explainable framework that can categorize images and create heatmaps highlighting the regions on which its prediction is based. METHODS: A dataset of 19341 thyroid ultrasound images with pathological results and physician-annotated TI-RADS features is used to train and test the robustness of the proposed framework. Then we conducted a benign-malignant classification study to determine whether physicians perform better with the assistance of an explainer than they do alone or with Gradient-weighted Class Activation Mapping (Grad-CAM). RESULTS: Reader studies show that the Explainer can achieve a more accurate diagnosis while explaining heatmaps, and that physicians' performances are improved when assisted by the Explainer. Case study results confirm that the Explainer is capable of locating more reasonable and feature-related regions than the Grad-CAM. CONCLUSIONS: The Explainer offers physicians a tool to understand the basis of AI predictions and evaluate their reliability, which has the potential to unbox the "black box" of medical imaging AI.

A multi-institutional study of association of sonographic characteristics with cervical lymph node metastasis in unifocal papillary thyroid carcinoma.

Objective: Papillary thyroid carcinoma (PTC) is the most common pathological type of thyroid carcinoma, and is prone to cervical lymph node metastases (CLNM). We aim to evaluate the association between sonographic characteristics of PTC and CLNM before the initial surgery. Methods: Clinical information as well as ultrasonographic measurements and characteristics for 2376 patients from three hospitals were acquired in this retrospective cohort study. Univariate and multivariate logistic analysis were performed to predict CLNM in unifocal PTC patients. Receiver operating characteristic (ROC) curve was created to evaluate diagnostic performance. Results: Univariate analysis showed that gender, age, maximum tumor diameter and volume, cross-sectional and longitudinal aspect ratio, location, echogenicity, margin, and echogenic foci were independently associated with CLNM metastatic status (P < 0.05). Multivariate logistic analysis showed that gender, age, maximum tumor diameter and volume, cross-sectional aspect ratio (CSAR), location, echogenicity, margin, and echogenic foci were independent correlative factors; CSAR showed a significant difference for PTC2 to predict CLNM. The area under the curve (AUC) of the maximum tumor diameter, tumor volume, margin, and echogenic foci was 0.70, 0.69, 0.65, and 0.70, respectively. The multiple-variable linear regression model was constructed with an AUC of 0.77, a specificity of 73.4%, and a sensitivity of 72.3%. Kruskal-Wallis analysis for positive subgroups, maximum tumor diameter and volume, cross-sectional and longitudinal aspect ratio, margin, and echogenic foci showed statistical significance (P < 0.05). Conclusions: Younger age (< 55 years), male, larger tumor, and echogenic foci were high risk factors for CLNM in patients with unifocal PTC. CSAR had a more effective predictive value for CLNM in patients with larger thyroid tumors. A larger tumor with irregular and punctate echogenic foci was also more prone to the lateral neck, and both central and lateral neck metastasis.

Clinical analysis of 13 cases of primary squamous-cell thyroid carcinoma.

Objective: To analyze the clinical features, ultrasonographic manifestations, pathological features, treatment and prognosis of primary thyroid squamous cell carcinoma (PSCTC) and summarize the experience in diagnosis and treatment of this condition. Methods: A retrospective analysis was conducted on patients who were admitted to Zhejiang Cancer Hospital from 2007 to 2021 due to thyroid nodules or thyroid malignant tumors that were ultimately confirmed by postoperative pathology as primary thyroid squamous cell carcinoma. We summarize the general situation, clinical information, laboratory examination, ultrasonic image characteristics, pathological examination, clinical treatment and prognosis of the patients. Results: PSCTC is most often seen in older men and progresses rapidly. In laboratory tests, some patients had elevated levels of tumor markers (CA199, squamous cell carcinoma antigen level), thyroglobulin levels and tumor-related substances, but all these indicators lacked specificity. The ultrasound features of PSCTC are mainly hypoechoic, hard, substantial nodules with gross borders and a grade 1-2 blood flow signal, sometimes with signs of necrosis and calcification. In terms of treatment, PSCTC is mainly surgically resected, though some patients in this study underwent iodine-131 radiation therapy, local radiotherapy, and chemotherapy with unclear results. None of the patients survived for very long after treatment, but the prognosis of patients with highly differentiated squamous carcinoma was significantly better than that of patients with poorly differentiated squamous carcinoma. Papillary thyroid carcinoma may be one of the causes of PSCTC. Conclusion: PSCTC is a malignant tumor with high malignancy and rapid clinical progression. Treatment options are mainly based on surgical resection and can be supplemented with radiotherapy and chemotherapy, but there is still a lack of a standardized treatment management system, and more cases and reports are needed to accumulate data.

The value of ultrasound grayscale ratio in the diagnosis of papillary thyroid microcarcinomas and benign micronodules in patients with Hashimoto's thyroiditis: A two-center controlled study.

Objective: The value of ultrasound grayscale ratio (UGSR) in the diagnosis of papillary thyroid microcarcinomas (PTMCs) and benign micronodules (BMNs) has been recognized by some authors, but studies have not examined these aspects in patients with Hashimoto's thyroiditis (HT). This retrospective study investigated the value of UGSR in the diagnosis of PTMCs and BMNs in patients with HT using data from two medical centers. Methods: Ultrasound images of 428 PTMCs in 368 patients with HT and 225 BMNs in 181 patients with HT in center A were retrospectively analyzed and compared to the ultrasound images of 412 PTMCs in 324 patients with HT and 315 BMNs in 229 patients with HT in medical center B. All of the cases were surgically confirmed. The UGSR was calculated as the ratio of the grayscale value of lesions to the surrounding normal thyroid tissues. The optimal UGSR thresholds for the PTMCs and BMNs in patients with HT from the two medical centers were determined using a receiver operating characteristic (ROC) curve. Furthermore, other statistics, including the area under the curve (AUC), the optimal UGSR threshold, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy of the two medical centers, were pair analyzed in this study. Results: The UGSR of PTMCs and BMNs in patients with HT from medical center A were 0.513 (0.442, 0.592) and 0.857 (0.677, 0.977) (Z = -15.564, p = 0), and those from medical center B were 0.514 (0.431, 0.625) and 0.917 (0.705, 1.131) (Z = -15.564, p = 0). For both medical centers A and B, the AUC, optimal UGSR threshold, sensitivity, specificity, PPV, NPV, and diagnostic accuracy of the UGSR in differentiating between PTMCs and BMNs in patients with HT were 0.870 and 0.889, 0.68 and 0.70, 0.921 and 0.898, 0.747 and 0.759, 0.874 and 0.829, 0.832 and 0.848, and 0.861 and 0.836, respectively. There were no significant differences in the UGSR for the PTMCs between patients from the two medical centers (Z = -0.815, p = 0.415), while there was a significant difference in the UGSR of the BMNs between patients from the two medical centers (Z = -3.637, p = 0). Conclusion: In the context of HT, UGSR still has high sensitivity, accuracy, and stability in differentiating between PTMCs and BMNs, making it a complementary differentiator of thyroid imaging reporting and data systems. However, due to its low specificity, a comprehensive analysis of other ultrasound signs is required.

Research Review of Thermal Ablation in the Treatment of Papillary Thyroid Carcinoma.

Background: Minimally invasive treatment of thyroid tumors has become increasingly common, but has mainly focused on benign thyroid tumors, whereas thermal ablation of thyroid cancer remains controversial. Clinical studies analyzing the efficacy of thermal ablation of papillary thyroid carcinoma (PTC) have been conducted in several countries to verify its safety. Here, we screened and reviewed recent studies on the efficacy and safety of thermal ablation of PTC as well as psychological assessment, patient prognosis, recurrence, and factors affecting ablation. Summary: The most significant controversy surrounding ablative treatment of PTC centers on its effectiveness and safety, and >40 studies have been conducted to address this issue. The studies include papillary thyroid microcarcinoma (PTMC) and non-PTMC, single PTC and multiple PTC, and controlled studies of ablative therapy and surgical treatment. In general, ablation techniques can be carefully performed and promoted under certain conditions and with active follow-up of postoperative patients. Ablation is a promising alternative treatment especially in patients who are inoperable. Conclusions: Clinical studies on PTC ablation have provided new perspectives on local treatment. However, because PTC grows very slowly, it is an indolent tumor; therefore, studies with larger sample sizes and extended post-procedure follow-ups are necessary to confirm the investigators' hypotheses.

Medical lesion segmentation by combining multimodal images with modality weighted UNet.

PURPOSE: Automatic segmentation of medical lesions is a prerequisite for efficient clinic analysis. Segmentation algorithms for multimodal medical images have received much attention in recent years. Different strategies for multimodal combination (or fusion), such as probability theory, fuzzy models, belief functions, and deep neural networks, have also been developed. In this paper, we propose the modality weighted UNet (MW-UNet) and attention-based fusion method to combine multimodal images for medical lesion segmentation. METHODS: MW-UNet is a multimodal fusion method which is based on UNet, but we use a shallower layer and fewer feature map channels to reduce the amount of network parameters, and our method uses the new multimodal fusion method called fusion attention. It uses weighted sum rule and fusion attention to combine feature maps in intermediate layers. During training, all the weight parameters are updated through backpropagation like other parameters in the network. We also incorporate residual blocks into MW-UNet to further improve segmentation performance. The comparison between the automatic multimodal lesion segmentations and the manual contours was quantified by (1) five metrics including Dice, 95% Hausdorff Distance (HD95), volumetric overlap error (VOE), relative volume difference (RVD), and mean-Intersection-over-Union (mIoU); (2) Number of parameters and flops to calculate the complexity of the network. RESULTS: The proposed method is verified on ZJCHD, which is the data set of contrast-enhanced computed tomography (CECT) for Liver Lesion Segmentation taken from Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Hangzhou, China. For accuracy evaluation, we use 120 patients with liver lesions from ZJCHD, of which 100 are used for fourfold cross-validation (CV) and 20 are used for hold-out (HO) test. The mean Dice was 90.55 ± 14.44 % $90.55 \pm 14.44\%$ and 89.31 ± 19.07 % $89.31 \pm 19.07\%$ for HO and CV tests, respectively. The corresponding HD95, VOE, RVD, and mIoU of the two tests are 1.95 ± 1.83 and 2.67 ± 3.35 mm, 13.11 ± 15.83 and 13.13 ± 18.52 % $13.13 \pm 18.52 \%$ , 12.20 ± 18.20 and 13.00 ± 21.82 % $13.00 \pm 21.82 \%$ , and 83.79 ± 15.83 and 82.35 ± 20.03 % $82.35 \pm 20.03 \%$ . The parameters and flops of our method is 4.04 M and 18.36 G, respectively. CONCLUSIONS: The results show that our method performs well on multimodal liver lesion segmentation. It can be easily extended to other multimodal data sets and other networks for multimodal fusion. Our method is the potential to provide doctors with multimodal annotations and assist them with clinical diagnosis.

A Preliminary Study on Exploring a potential Ultrasound Method for Predicting Cervical Cancer.

Background: The level of cervical cancer screening in underdeveloped countries is far behind that of developed countries mostly because current cervical cancer screening methods are difficult to implement in underdeveloped countries. The use of non-invasive, repeatable, and low-cost ultrasound needs to be accessed. Methods: The Canadian Sonix TOUCH ultrasound system and transvaginal ultrasound probe were used to record ultrasound radio frequency (RF) signals from cervical tissues of 69 patients with cervical cancer and 37 healthy women. The self-compiled RF time series signal analysis software was used to extract 3 different dimensions of parameters of the region of interest (ROI), including time domain, frequency domain, and fractal dimension (FD). Fourteen spectrum characteristic parameters were extracted, of which structure function method FD (SFD) and Higuchi FD belonged to FD parameters; slope, intercept, midbandfit, S1, S2, S3, and S4 were frequency domain parameters; and fuzzy entropy, kurtosis, peak, cross zero count, and cross zero standard deviation (Std) were time domain parameters. Results: The average values of the five time-domain characteristic parameters of cervical cancer tissues were smaller than those of normal cervical tissues (fuzzy entropy: 1.70±0.29 vs. 1.83±0.20; kurtosis: 0.347±0.03 vs. 0.350±0.02; peak: 1989.9±166.8 vs. 2024.69±187.5; cross zero count: 3.77±0.31 vs. 3.81±0.29; cross zero Std: 1.26±0.17 vs. 1.33±0.14), although the differences were not statistically significant (P = 0.130, 0.326, 0.618, 0.442, and 0.204, respectively). The average values ​​of the two FD characteristic parameters and the seven frequency domain characteristic parameters of cervical cancer tissues were larger than those of normal tissues (SFD: 1.84±0.28 vs. 1.46±0.39; Higuchi FD: 1.71±0.30 vs. 1.28±0.30; slope: -0.32±0.08 vs. -0.26±0.05; intercept: 0.48±0.02 vs. 0.46±0.02; midbandfit: 0.35±0.03 vs. 0.33±0.03; S1: 15.66±1.01 vs. 13.57±1.69; S2: 10.12±0.69 vs. 9.32±1.27; S3: 9.44±1.12 vs. 8.66±1.09; S4: 7.67±1.01 vs. 6.43±0.65), and the differences were statistically significant (P < 0.05). No effective parameters were found to identify cervical squamous cell carcinoma tissues with different levels of differentiation (P > 0.05). Conclusion: Quantitative analysis of RF time series signals based on ultrasound RF flow is expected to become a simple and non-invasive imaging method for cervical cancer diagnosis. However, whether it can be applied to the identification of early small cervical cancer lesions remains to be determined.

A Modified Technique for Valve Regurgitation After Partial Atrioventricular Septal Defect Repair.

Given its complex pathologic anatomy, recurrent left atrioventricular valve regurgitation after partial atrioventricular septal defect repair remains a challenge for surgical correction. Here, we introduce a modified bridging technique by shortening the anteroposterior leaflet distance in selected patients with inadequate coaptation to compensate for the short leaflet height, specifically that of the anterior leaflet.

Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules.

Objective: This study was designed to distinguish benign and malignant thyroid nodules by using deep learning(DL) models based on ultrasound dynamic videos. Methods: Ultrasound dynamic videos of 1018 thyroid nodules were retrospectively collected from 657 patients in Zhejiang Cancer Hospital from January 2020 to December 2020 for the tests with 5 DL models. Results: In the internal test set, the area under the receiver operating characteristic curve (AUROC) was 0.929(95% CI: 0.888,0.970) for the best-performing model LSTM Two radiologists interpreted the dynamic video with AUROC values of 0.760 (95% CI: 0.653, 0.867) and 0.815 (95% CI: 0.778, 0.853). In the external test set, the best-performing DL model had AUROC values of 0.896(95% CI: 0.847,0.945), and two ultrasound radiologist had AUROC values of 0.754 (95% CI: 0.649,0.850) and 0.833 (95% CI: 0.797,0.869). Conclusion: This study demonstrates that the DL model based on ultrasound dynamic videos performs better than the ultrasound radiologists in distinguishing thyroid nodules.

LncRNA RUSC1-AS1 promotes osteosarcoma progression through regulating the miR-340-5p and PI3K/AKT pathway.

Dysregulation of long noncoding RNA (lncRNA) is frequently involved in the progression and development of osteosarcoma. LncRNA RUSC1-AS1 is reported to be upregulated and acts as an oncogene in hepatocellular carcinoma, cervical cancer and breast cancer. However, its role in osteosarcoma has not been studied yet. In the present study, we investigated the role of RUSC1-AS1 in osteosarcoma both in vitro and in vivo. The results showed that the expression of RUSC1-AS1 was significantly upregulated in osteosarcoma cell line U2OS and HOS compared to that in human osteoblast cell line hFOB1.19. Similar results were found in human samples. Silencing RUSC1-AS1 by siRNA significantly inhibited U2OS and HOS cell proliferation and invasion, measured by CCK-8 and transwell assay. Besides, knockdown of RUSC1-AS1 increased cell apoptosis in osteosarcoma cell lines. In addition, RUSC1-AS1 promoted the epithelial-mesenchymal transition (EMT) process of osteosarcoma cells. In vivo experiments confirmed that RUSC1-AS1 knockdown had an inhibitory effect on osteosarcoma tumor growth. Mechanically, we showed that RUSC1-AS1 directly binds to and inhibits miR-340-5p and activates the PI3K/AKT signaling pathway. In conclusion, our study demonstrated that RUSC1-AS1 promoted osteosarcoma development both in vitro and in vivo through sponging to miR-340-5p and activating the PI3K/AKT signaling pathway. Therefore, RUSC1-AS1 becomes a potential therapeutic target for osteosarcoma.

AI detection of mild COVID-19 pneumonia from chest CT scans.

OBJECTIVES: An artificial intelligence model was adopted to identify mild COVID-19 pneumonia from computed tomography (CT) volumes, and its diagnostic performance was then evaluated. METHODS: In this retrospective multicenter study, an atrous convolution-based deep learning model was established for the computer-assisted diagnosis of mild COVID-19 pneumonia. The dataset included 2087 chest CT exams collected from four hospitals between 1 January 2019 and 31 May 2020. The true positive rate, true negative rate, receiver operating characteristic curve, area under the curve (AUC) and convolutional feature map were used to evaluate the model. RESULTS: The proposed deep learning model was trained on 1538 patients and tested on an independent testing cohort of 549 patients. The overall sensitivity was 91.5% (195/213; p < 0.001, 95% CI: 89.2-93.9%), the overall specificity was 90.5% (304/336; p < 0.001, 95% CI: 88.0-92.9%) and the general AUC value was 0.955 (p < 0.001). CONCLUSIONS: A deep learning model can accurately detect COVID-19 and serve as an important supplement to the COVID-19 reverse transcription-polymerase chain reaction (RT-PCR) test. KEY POINTS: • The implementation of a deep learning model to identify mild COVID-19 pneumonia was confirmed to be effective and feasible. • The strategy of using a binary code instead of the region of interest label to identify mild COVID-19 pneumonia was verified. • This AI model can assist in the early screening of COVID-19 without interfering with normal clinical examinations.

Ultrasonic identification and regression analysis of 294 thyroid follicular tumors.

CONTEXT: We analyzed the clinical features and ultrasound image features of follicular thyroid carcinoma (FTC) and follicular thyroid adenoma (FTA). AIMS: This study aimed to identify ultrasonographic differences and correlations between FTC and FTA. Meanwhile, ultrasonographic manifestations of thyroid follicular tumor were also retrospectively analyzed. SETTINGS AND DESIGN: Using pathological results as the gold standard, the clinical and ultrasonic image characteristics of FTA and FTC were statistically analyzed, and the differences were analyzed. MATERIALS AND METHODS: A total of 304 patients who were diagnosed with FTC or FTA by pathology after thyroidectomy from March 2009 to March 2018 were enrolled in this study. Their ultrasonic images were analyzed; image features were extracted and correlation analyses for these features were conducted. Differences in ultrasonic images between FTC and FTA were also compared. STATISTICAL ANALYSIS USED: Independent sample t-test; Wilcoxon rank sum test; A Chi-square test: Univariate and multivariate logistic regression analyses. RESULTS: When performing ultrasound diagnosis, attention should be paid to identify FTC and FTA in terms of age, nodular goiter conditions, nodular boundary conditions, internal echo, calcification, blood flow signals, thyroid imaging reporting and data system (TI-RADS) grading and cystic solidity conditions. Moreover, a multivariate logistic regression showed that the boundaries were unclear, and cystic degeneration, TI-RADS, hypoecho, nodular goiter, macrocalcification and microcalcification were associated with FTC. Among them, macrocalcification is a protective factor for thyroid follicular tumors, and other indicators are risk factors. CONCLUSION: Ultrasound can provide valuable information for the identification of follicular neoplasms, but further research in this area is still necessary.

The protective role of omentin-1 in IL-1ß-induced chondrocyte senescence.

Osteoarthritis is a common type of degenerative joint disease. Inflammation-related chondrocyte senescence plays a major role in the pathogenesis of osteoarthritis. Omentin-1 is a newly identified anti-inflammatory adipokine involved in lipid metabolism. In this study, we examined the biological function of omentin-1 in cultured chondrocytes. The presence of omentin-1 potently suppresses IL-1ß-induced cellular senescence as revealed by staining with senescence-associated beta-galactosidase (SA-ß-Gal). At the cellular level, omentin-1 attenuates IL-1ß-induced G1 phase cell-cycle arrest. Mechanistically, we demonstrate that omentin-1 reduced IL-1ß-induced expression of senescent factors including caveolin-1, p21, and PAI-1 as well as p53 acetylation through ameliorating SIRT1 reduction. Notably, silencing of SIRT1 abolishes IL-1ß-induced senescence along with the induction of p21 and PAI-1, suggesting that the action of omentin-1 in chondrocytes is dependent on SIRT1. Collectively, our results revealed the molecular mechanism through which the adipokine omentin-1 exerts a beneficial effect, thereby protecting chondrocytes from senescence. Thus, omentin-1 could have clinical implication in the treatment of osteoarthritis.

Ack1 overexpression promotes metastasis and indicates poor prognosis of hepatocellular carcinoma.

Despite the substantial data supporting the oncogenic role of Ack1, the predictive value and biologic role of Ack1 in hepatocellular carcinoma (HCC) metastasis remains unknown. In this study, both correlations of Ack1 expression with prognosis of HCC, and the role of Ack1 in metastasis of HCC were investigated in vitro and in vivo. Our results showed that Ack1 was overexpressed in human HCC tissues and cell lines. High Ack1 expression was associated with HCC metastasis and determined as a significant and independent prognostic factor for HCC after liver resection. Ack1 promoted HCC invasion and metastasis in vitro and in vivo. Mechanistically, we confirmed that Ack1 enhanced invasion and metastasis of HCC via EMT by mediating AKT phosphorylation. In conclusion, our study shows Ack1 is a novel prognostic biomarker for HCC and promotes metastasis of HCC via EMT by activating AKT signaling.

Mesohepatectomy for centrally located large hepatocellular carcinoma: Indications, techniques, and outcomes.

BACKGROUND: Whether mesohepatectomy should be performed for large hepatocellular carcinoma (HCC) located in the central part of the liver is controversial, and the safety and long-term survival after this operation remain to be investigated. METHODS: Between January 2002 and December 2012, 696 patients with HCC located in the central part of the liver who received liver resection in our hospital were included in this study. These patients were divided into three groups: 158 patients with large HCC (tumor size >5.0 cm) and 192 patients with small HCC (tumor size ≤ 5.0 cm) who received mesohepatectomy were classified as the mesohepatectomy for large HCC (MHG-L) group and the mesohepatectomy for small HCC (MHG-S) groups, respectively, and 346 patients with large HCC who received hemihepatectomy or less were classified as the non-mesohepatectomy for large HCC (NMHG-L) group. The operative indications, techniques, and outcomes of the three groups were analyzed retrospectively. RESULTS: There were no substantial differences among the three groups in in-hospital mortality or postoperative complication rates. The overall survival and disease-free survival were not different between the MHG-L group and the NMHG-L group or between the MHG-L group and the MHG-S group. Univariable and multivariable analyses of the MHG-L mesohepatectomy group indicated that cirrhosis, tumor number, and vascular invasion were independent risk factors of poor long-term survival of mesohepatectomy. In the MHG-L and NMHG-L groups, solitary large hepatocellular carcinoma had better long-term survival than nodular large hepatocellular carcinoma. CONCLUSION: Mesohepatectomy is safe and efficacious for BCLC B/C patients who have centrally located large HCC, especially for solitary tumors, with good survival outcomes.