in current clinical practice, the standard evaluation for axillary lymph node (ALN) status in breast cancer has a low efficiency and is based on an invasive procedure that causes operative-associated complications in many patients. Therefore, we aimed to use machine learning techniques to develop an efficient preoperative magnetic resonance imaging (MRI) radiomics evaluation approach of ALN status and explore the association between radiomics and the tumor microenvironment in patients with early-stage invasive breast cancer.in this retrospective multicenter study, three independent cohorts of patients with breast cancer (n = 1,088) were used to develop and validate signatures predictive of ALN status. After applying the machine learning random forest algorithm to select the key preoperative MRI radiomic features, we used ALN and tumor radiomic features to develop the ALN-tumor radiomic signature for ALN status prediction by the support vector machine algorithm in 803 patients with breast cancer from Sun Yat-sen Memorial Hospital and Sun Yat-sen University Cancer Center (training cohort). By combining ALN and tumor radiomic features with corresponding clinicopathologic information, the multiomic signature was constructed in the training cohort. Next, the external validation cohort (n = 179) of patients from Shunde Hospital of Southern Medical University and Tungwah Hospital of Sun Yat-Sen University, and the prospective-retrospective validation cohort (n = 106) of patients treated with neoadjuvant chemotherapy in prospective phase 3 trials [NCT01503905], were included to evaluate the predictive value of the two signatures, and their predictive performance was assessed by the area under operating characteristic curve (AUC). This study was registered with ClinicalTrials.gov, number NCT04003558.the ALN-tumor radiomic signature for ALN status prediction comprising ALN and tumor radiomic features showed a high prediction quality with AUC of 0·88 in the training cohort, 0·87 in the external validation cohort, and 0·87 in the prospective-retrospective validation cohort. The multiomic signature incorporating tumor and lymph node MRI radiomics, clinical and pathologic characteristics, and molecular subtypes achieved better performance for ALN status prediction with AUCs of 0·90, 0·91, and 0·93 in the training cohort, the external validation cohort, and the prospective-retrospective validation cohort, respectively. Among patients who underwent neoadjuvant chemotherapy in the prospective-retrospective validation cohort, there were significant differences in the key radiomic features before and after neoadjuvant chemotherapy, especially in the gray-level dependence matrix features. Furthermore, there was an association between MRI radiomics and tumor microenvironment features including immune cells, long non-coding RNAs, and types of methylated sites. Interpretation this study presented a multiomic signature that could be preoperatively and conveniently used for identifying patients with ALN metastasis in early-stage invasive breast cancer. The multiomic signature exhibited powerful predictive ability and showed the prospect of extended application to tailor surgical management. Besides, significant changes in key radiomic features after neoadjuvant chemotherapy may be explained by changes in the tumor microenvironment, and the association between MRI radiomic features and tumor microenvironment features may reveal the potential biological underpinning of MRI radiomics.No funding.
Abstract Having a better grasp of the molecular mechanisms underlying carcinogenesis and progression in osteosarcoma would be helpful to find novel therapeutic targets. Different types of cancers have presented abnormal expression of miRNA-101 (miR-101). Nevertheless, we still could not figure out what expression of miR-101 in human osteosarcoma is and its biological function. Thus, we conducted the present study to identify its expression, function, and molecular mechanism in osteosarcoma. We detected the expression of miR-101 in osteosarcoma samples and cell lines. The effects of miR-101 on osteosarcoma cells’ proliferation and invasion were evaluated. Luciferase reporter assay was applied to identify the direct target of miR-101. Compared with adjacent normal specimens and normal bone cell line by using qPCR, the expression levels of miR-101 in osteosarcoma specimens and human osteosarcoma cell lines distinctly decreased. According to function assays, we found that overexpression of miR-101 significantly inhibited the cell proliferation and invasion in osteosarcoma cells. Moreover, we confirmed that zinc finger E-box binding homeobox 2 (ZEB2) was a direct target of miR-101. In addition, overexpression of ZEB2 could rescue the inhibition effect of proliferation and invasion induced by miR-101 in osteosarcoma cells. MiR-101 has been proved to be down-regulated in osteosarcoma and has the ability to suppress osteosarcoma cell proliferation and invasion by directly targetting ZEB2.
Objective
To analyze the expression of YTH domain containing family 2 (YTHDF2) in the hepatocellular carcinom (HCC) and the relationship between the expression of YTHDF2 and prognosis. To detect the influence of YTHDF2 in cell proliferation, migration and invasion.
Methods
We detected the expression of YTHDF2 among the paired HCC tissue and adjacent tumor tissue with bioinformatics, real-time quantitative reverse transcriptase-polymerase chain reaction (RT-qPCR) and immunohistochemistry. To analyze the relationship between YTHDF2 expression and clinical characteristics. We established the HCC cells with YTHDF2 knockdown and analyzed the cell proliferation with cell counting kit-8 (CCK-8) assay and detect the cell migration with Transwell and wound scratch assay. To analyze the expression of E-cadherin and N-cadherin that associated with epithelial-mesenchymal transition (EMT) with western blotting.
Results
Bioinformatics analysis showed the high expression of YTHDF2 in HCC tumor and its relationship with poor prognosis. We found that the HCC tumor tissue has a higher expression of YTHDF2 in compared with non-tumor adjacent tissue (17.76±4.78 vs. 12.10±3.43, P<0.01). The expression of YTHDF2 is positively related with BCLC stage, TNM stage and portal vein tumor thrombosis (PVTT) (P<0.01). CCK-8 assay demonstrated that the absorbance at 450 nm of the cell with YTHDF2 knockdown is lower than the negative control, while HepG2 is (1.129±0.178 vs. 1.571±0.226, P<0.01) and Huh7 (1.227±0.251 vs. 1.815±0.302, P<0.05). Transwell assay showed that the number of cell that diffuses through the membrane of cell with YTHDF2 knockdown is lower than the negative control. The wound scratch assay showed that the percent of wound area in cell with YTHDF2 knockdown is lower. We detected the HCC cells with YTHDF2 konckdown and found it had a higher relative expression of E-cadherin (1.780±0.319 vs. 0.916±0.173, P<0.05) and a lower expression of N-cadherin (0.821±0.217 vs. 4.697±0.398, P<0.01).
Conclusion
The high expression of YTHDF2 in HCC tissue showed its relationship with HCC development and YTHDF2 may play an important role in HCC cell proliferation and migration.
Key words:
YTH domain containing family 2; Hepatocellular carcinom; Epithelial-mesenchymal transition
To explore the clinical value of CT guided 125I seeds implantation in the treatment of local recurrent rectal cancer after surgery resection.Twenty-one patients with local recurrent rectal cancer after surgery, 12 males and 9 females, aged 47 (35-69), with the longest diameter of 4. 2 cm (3.4 - 6.4 cm), were treated by CT guided 125I radioactive seeds implantation according to treatment planning system (TPS) or Halarism's experienced function: mCi = Da x 5. Totally, 506 seeds, 24.1 for one patient on average, were implanted. All the patients received PET-CT scan 3 months after the treatment and were followed up for 12.5 months (7-22 months).Complete remission, partial remission, and no change were seen in 16, 3, and 2 patients respectively. All patients survived.CT guided 125I seeds implantation is an effective methods in the treatment of local recurrent rectal cancer after surgery resection.
Background: Cells of the innate and adaptive immune systems play a critical role in the host response to sepsis. However, whether immune cell abundance of peripheral blood has potential application in sepsis patients' prediction of 28-day survival is largely unknown. Here we aim to develop a deep learning model to predict 28-day survival in patients with sepsis. Methods: In this study.a total of 479 sepsis patients were included and patients were randomly divided into training and validation groups in a 9:1 ratio. We built the DeepSurv in TensorFlow, a deep learning survival neural network based model on sepsis patients9 data with 28 immune cells. The algorithm was internally validated and the primary end point was 28-day survival. Results: In the training group, we established a deep learning survival neural network model showed promising results to predict 28-day survival in sepsis patients, patients with low vs high risk score had statistically significantly longer 28-day survival [hazard ratio(HR)=0.022, 95%CI=0.013-0.038, P<0.005]. The immune cell abundance risk score was associated with 28-day survival (AUCs for 7-, 14- and 28-day survival were 0.85, 0.912 an 0.936, respectively). Similarly, in the test group, patients with low vs high risk score had statistically significantly longer 28-day survival[P<0.005] and well AUCs for 14- and 28-day survival. Further, this study identified that model obviously related to immune microenvironment characteristics. Conclusions: This study developed and validated novel deep learning survival neural network model showed reliable individual 28-day survival information in prognostic evaluation and treatment recommendation in patients with sepsis.
Dear Editor, We and others have shown that the tumor mutation burden (TMB) and several underlying oncogenic alterations could provide clinically predictive implications for immune checkpoint inhibitor (ICI).1-3 Protein tyrosine phosphatases (PTPs) consist of a variety classes, and most of them are highly mutated in multiple cancers and are closely interact with innate and acquired immunity regulating immune cell activation and differentiation.4, 5 PTP receptor T (PTPRT) has been found to be the most frequently mutated PTP gene in cancers and could predict poor prognosis;4, 6 however, the association of PTPRT mutation with clinical outcomes of ICI remains unknown. Here, we performed a comprehensive pancancer investigation to clinically validate PTPRT mutation as a predictive biomarker for ICI therapy. We collected clinical and PTPRT mutational data quantified by whole exome sequencing of 2129 cancer patients treated with ICI and 10,814 cancer patients without receiving ICI from the cBioPortal, PubMed, and The Cancer Genome Atlas. The study protocol was approved by the ethics committee of the Sun Yat-sen Memorial Hospital of Sun Yat-sen University. The requirement for informed consent of study participants and the permission to use the patient data were waived because the human data were obtained from publicly available datasets. All analyses were performed according to the STROBE guideline from September 18 through October 1, 2019. Overall survival (OS) were primary outcomes, which were computed using the Kaplan-Meier method and were assessed with the log-rank test and the hazard ratio (HR) calculated by the Cox regression model. The TMB in PTPRT wild-type versus mutant groups were compared with Wilcoxon rank-sum tests. All analyses were performed using R (version 3.4.4) and were considered statistically significant if P values < .05. Among 2129 ICI-treated patients (250 [11.7%] PTPRT mutant; Figure 1A), 596 (28.0%) patients had melanoma, 510 (24.0%) patients had non-small cell lung cancer (NSCLC), and 1023 (48.1%) patients had 12 other cancer types. Patients treated with ICI showed significantly higher TMB in PTPRT mutant group versus PTPRT wild-type group (P < .001; Figure 1B). Thirty-five (6.9%) of 510 NSCLC patients and 151 (25.3%) of 596 melanoma patients harbored PTPRT mutation, who analogously displayed remarkably higher TMB than patients with PTPRT wild-type tumors (P < .001; Figure 1C and D). PTPRT mutations were identified in 687 (6.4%) out of 10,814 patients without receiving ICI across 33 cancer types, among which the mutation frequency was 28.4% in melanoma, 11.1% in esophagogastric adenocarcinoma, 10.9% in endometrial carcinoma, 8.6% in colorectal adenocarcinoma, and 8.0% in NSCLC (Figure 2A). Missense mutations were most commonly observed (82.6%), followed by truncating mutations (15.5%) (Figure 2B). PTPRT mutation resulted in significantly longer OS in 2129 pancancer patients treated with ICI compared with PTPRT wild-type (HR 0.63, 95% CI 0.52-0.77, P < .001; Figure 3A). We further found the clinical usefulness of PRPRT mutation status was most prominent in ICI-treated patients with NSCLC and melanoma. Compared with PTPRT wild-type group, PTPRT mutation group had substantially longer OS in patients with NSCLC and melanoma (HR 0.61, 95% CI 0.48-0.77, P < .001; Figure 3B). However, among ICI-treated patients with cancers except NSCLC and melanoma, no significant difference in OS between PTPRT mutation and wild-type patients was observed (HR 0.95, 95% CI 0.64-1.43; P = .810). We also assessed PTPRT mutation in patients without receiving ICI. Among 10,814 pancancer patients, there was no difference in OS between PTPRT mutant and wild-type (HR 1.00, 95% CI 0.87-1.14; P = 0.980; Figure 3C). Among 986 NSCLC patients (101 [10.2%] PTPRT mutant) and 431 melanoma patients (126 [29.2%] PTPRT mutant), no difference in OS between PTPRT mutant and wild-type patients was observed, either (HR 0.83 95% CI 0.67-1.03; P = .097; Figure 3D). These findings indicated that the status of PTPRT mutation was particularly predictive of ICI treatment. To the best of our knowledge, this is the first study to identify the mutation status of PTPRT as a key predictor of ICI efficacy. We found that PTPRT mutation conferred an elevated TMB and better survival during ICI therapy in pancancer and specifically in melanoma and NSCLC, which collaborated with our previous research1 showing a pronounced survival and response benefits of ICI among cancer patients with high TMB. PTPRT has not been suggested to be screened for mutations in current widely used gene panels such as Memorial Sloan Kettering Cancer Center's Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) and FDA-approved FoundationOne CDx (F1CDx). Therefore, PTPRT should be considered together with other known essential genes to expand the landscape of immuno-oncological genomic panel, and should be integrated into multiomics to more fully realize the precision immunotherapy. In-deep characterization of PTP expression pattern could be informative for understanding patterns of immune escape and the selection of candidates for immunotherapy. Moreover, PD-L1 inhibitor atezolizumab plus VGFR inhibitor bevacizumab plus platinum-based chemotherapy was shown to have an encouraging survival benefit in recent randomized IMpower 150 trial.7 We hypothesized that the efficacy of this strategy probably further enhanced through concurrently targeting PTPRT, since PTPRT mutation was demonstrated to be promisingly predictive of immunotherapy efficacy in our study and has been found to determine bevacizumab resistance in the study conducted by Hsu et el.8 The study limitations included a potential random variability in the context of an exploratory analysis contributed by NSCLC and melanoma, our inability to assess the heterogeneity of other treatment between ICI and non-ICI groups and to clarify the mechanisms underlying the interaction between PTPRT mutation and ICI. Future prospective trials with a larger sample size, more detailed clinical treatment information and a longer follow-up are needed to validate the pancancer applicability of PTPRT mutation status and in-deep characterize how PTPRT mutation interact with immune system to influence ICI benefit. In conclusion, PTPRT mutation status could serve as a predictive biomarker for ICI in pancancer and specifically in NSCLC and melanoma. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. This work was supported by the National Science and Technology Major Project (grant number 2020ZX09201021); the Medical Artificial Intelligence Project of Sun Yat-Sen Memorial Hospital (grant number YXRGZN201902); the National Natural Science Foundation of China (grant numbers 81572596, 81972471, and U1601223); the Natural Science Foundation of Guangdong Province (grant number 2017A030313828); the Guangzhou Science and Technology Major Program (grant number 201704020131); the Sun Yat-Sen University Clinical Research 5010 Program (grant number 2018007); the Sun Yat-Sen Clinical Research Cultivating Program (grant number SYS-C-201801); the Guangdong Science and Technology Department (grant number 2017B030314026); and the Special Funds for the Cultivation of Guangdong College Students' Scientific and Technological Innovation (grant number pdjh2019a0212); National Students' Innovation and Entrepreneurship training program (grant number 201910571001); and Guangdong Medical University College Students' Innovation Experiment Project (grant number ZZZF001). The preliminary results were presented in part as a Poster at the ESMO Immuno-Oncology 2019 Congress; December 12, 2019, Geneva, Switzerland.9 The study protocol was approved by the ethics committee of the Sun Yat-sen Memorial Hospital of Sun Yat-sen University. The requirement for informed consent of study participants and the permission to use the patient data were waived because the human data were obtained from publicly available datasets.
Dear Editor, Currently, the association of patients' sex with survival benefit of immune checkpoint inhibitor (ICI) is being actively investigated, but inconsistent results have been produced. A meta-analysis1 reported that male patients had significantly lager overall survival (OS) benefits from ICI versus control treatment than did female patients. However, an updated meta-analysis2 found no significant difference in OS between sexes. This conflict indicated that the sex-related effects on ICI efficacy involved complex and unknown elements of tumor microenvironment. The phase III KEYNOTE-024 trial3 enrolled cancer patients with PD-L1 expression on at least 50% tumor cells and found a strikingly improved OS benefit with ICI compared with chemotherapy in male but only a minimal improvement in female. However, neither male nor female had significant OS benefit from ICI over chemotherapy in some trials recruiting cancer patients with a lower PD-L1 expression threshold (≥1%).4, 5 Therefore, we hypothesize that PD-L1 expression has essential impact on the clinical usefulness of sex. This study, based on a post hoc analysis of prospective individual patient data from five clinical trials including OAK, POPLAR, IMvigor210, KEYNOTE-001, and CheckMate-012, and a meta-analysis of nine randomized controlled trials (RCTs), is the first to clarify the sex-related difference in ICI efficacy by using PD-L1 expression (Figure 1A). We further evaluated the landscape of tumor immune microenvironment to explore potential factors underpinning this difference. Full methods were described in the Supporting Information Methods. Characteristics and references of included trials and patients were described in the Supporting Information Result 1 and Tables S1 and S2. The individual-patient level analysis of OAK, POPLAR, and IMvigor210 trials showed that for patients receiving ICI, there was no significant difference in OS between male and female (HR 0.85, 95% CI 0.72-1.02; P = .073; Figure S2A). However, stratified by PD-L1 expression of 1%, OS for patients with PD-L1 expression <1% was significantly longer for female compared with male (HR 0.58, 95% CI 0.42-0.80; P < .001), but OS for patients with PD-L1 expression≥1% did not significantly differed between the sexes (HR 1.01, 95% CI 0.79-1.29; P = .910) (Figure 1B,C and Table S3). Results were consistent stratified by PD-L1 expression of 50%. Overall survival was significantly longer for female than for male at PD-L1 threshold of <50% (HR 0.76, 95% CI 0.61-0.95; P = .013), but OS did not significantly differ between the sexes at PD-L1 threshold of 1–49% (HR 0.76, 95% CI 0.52-1.13; P = .173) or ≥50% (HR 0.64, 95% CI 0.31-1.33; P = .231) (Figure S2B-D and Table S3). When examining chemotherapy, sex was not associated with OS regardless of PD-L1 expression (see details in Supporting Information Results 2, Figure 1D-E, and Figure S3). The individual-patient level analysis of POPLAR and OAK trials found that ICI improved OS compared with chemotherapy in both female (HR 0.67, 95% CI 0.53-0.84; P < .001) and male (HR 0.74, 95% CI 0.62-0.88; P < .001) (Figure S4A,B). However, among patients with PD-L1 expression <1%, the benefit from ICI versus chemotherapy was significantly different in female (HR 0.57, 95% CI 0.38-0.85; P = .006; Figure 2A), but there was no significant difference between the two treatments in male (HR 0.93, 95% CI 0.68-1.26; P = .621; Figure 2B). Conversely, among patients with PD-L1 expression ≥1%, there was no significant difference between the two treatments in female (HR 0.69, 95% CI 0.47-1.01; P = .057; Figure 2C), whereas the benefit from ICI over chemotherapy was significantly different in male (HR 0.71, 95% CI 0.54-0.95; P = .022; Figure 2D). Further, we improved the evidence for patients with PD-L1 expression ≥1% by pooling the individual patient-level result with the meta-analysis result of eight other trials, which showed that although female patients showed significant OS benefit from ICI over chemotherapy (eight RCTs, 1602 patients; HR 0.86, 95% CI 0.74-1.00; P = .05), this benefit only remained significant in subsequent line (four RCTs, 850 patients; HR 0.77, 95% CI 0.62-0.96; P = .02), without significant benefits for patients in first-line setting (four RCTs, 752 patients; HR 0.96, 95% CI 0.81-1.15; P = .67) or in subgroups by regimen and ICI class (Figure 2E, Figures S5 and S6 and Table S4). Male patients with PD-L1 expression ≥1% could derive significant OS benefit from ICI over chemotherapy (nine RCTs, 3166 patients; HR 0.76, 95% CI 0.67-0.85; P < .01; Figure 2F). This benefit remained significant in both first line (four RCTs, 1393 patients; HR 0.79, 95% CI 0.65-0.97; P = .02) and subsequent line (five RCTs, 1723 patients; HR 0.73, 95% CI 0.62-0.85; P < .01), and in other tested subgroups (Figure 2F, Figures S7 and S8 and Table S4). There were no significant differences in effects on OS between subgroups. Similar findings were observed stratified by PD-L1 expression of 50% (see details in Supporting Information Results 3 and Figures S9 and S10). Finally, central memory T cells (Tcm), rather than tumor mutation burden (TMB) and neoantigen burden (NAB), was found to potentially correlated with the OS differences between the sexes (see details in Supporting Information Results 4; Figures S11-S24 and Tables S5-S7). To our knowledge, this is the first study to reveal that PD-L1 expression has decisive effect on sex-associated differences in ICI efficacy. Previous meta-analyses included patients across varieties of PD-L1 thresholds, which might explain their inconsistency.1, 2 Previous research indicated that sex differences in mutational landscape might explain ICI efficacy differentially associated with sex,6 but neither TMB nor NAB was significantly differed between male and female at any PD-L1 expression thresholds in our study. Instead, Tcm probably has potential impact on sex differences in ICI efficacy. However, with only bladder cancer patients from IMvigor210 trial having RNA sequencing data, we were unable to evaluate the role of Tcm in other cancer types. Additionally, multi-omics have been shown to mutually predict ICI efficacy7, 8; therefore, future studies are warranted to comprehensively investigate sex difference in immune landscape using multi-omics across different cancer types. This study found the survival benefits of ICI in male and female were greatly influenced by PD-L1 expression, especially in NSCLC. At PD-L1 expression <1%, ICI should be recommended for female but not for male, and Tcm might be essential to drive this recommendation. We suggest that sex and PD-L1 expression should be jointly taken into account in the clinical decision making for ICI in cancer. We sincerely thank all of the investigators and patients who participated in the included trials. Li, Chen, Zhang, Zhong, Ou, Hu, Yu, and Yao jointly designed the study and drafted of the manuscript. Hu, Yu, and Yao provided supervision. Li and Yao obtained funding. All authors contributed to the data collection, data analysis and interpretation, manuscript revision, and approval of the final version. National Science and Technology Major Project; Grant Number: 2020ZX09201021; Medical Artificial Intelligence Project of Sun Yat-Sen Memorial Hospital; Grant Number: YXRGZN201902; National Natural Science Foundation of China; Grant Numbers: 81572596, 81972471, and U1601223; Natural Science Foundation of Guangdong Province; Grant Number: 2017A030313828; Guangzhou Science and Technology Major Program; Grant Number: 201704020131; Guangdong Science and Technology Department; Grant Number: 2017B030314026; Sun Yat-Sen University Clinical Research 5010 Program; Grant Number: 2018007; Sun Yat-Sen Clinical Research Cultivating Program; Grant Number: SYS-C-201801; Special Funds for the Cultivation of Guangdong College Students' Scientific and Technological Innovation; Grant Number: pdjh2019a0212; National Students' Innovation and Entrepreneurship Training Program; Grant Number: 201910571001. The study protocol was approved by the ethics committee of the Sun Yat-sen Memorial Hospital of Sun Yat-sen University. The requirement for informed consent of study participants was waived because the human data were obtained from publicly available datasets. The authors declare that they have no conflict of interest. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request (yaoherui@mail.sysu.edu.cn). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Background: Sepsis commonly causes acute respiratory distress syndrome (ARDS), and ARDS contributes to poor prognosis in sepsis patients. Early prediction of ARDS for sepsis patients remains a clinical challenge. This study aims to develop and validate chest computed tomography (CT) radiomic-based signatures for early prediction of ARDS and assessment of individual severity in sepsis patients. Methods: In this ambispective observational cohort study, a deep learning model, a sepsis-induced acute respiratory distress syndrome (SI-ARDS) prediction neural network, will be developed to extract radiomics features of chest CT from sepsis patients. The datasets will be collected from these retrospective and prospective cohorts, including 400 patients diagnosed with sepsis-3 definition during a period from 1 May 2015 to 30 May 2022. 160 patients of the retrospective cohort will be selected as a discovering group to reconstruct the model and 40 patients of the retrospective cohort will be selected as a testing group for internal validation. Additionally, 200 patients of the prospective cohort from two hospitals will be selected as a validating group for external validation. Data pertaining to chest CT, clinical information, immune-associated inflammatory indicators and follow-up will be collected. The primary outcome is to develop and validate the model, predicting in-hospital incidence of SI-ARDS. Finally, model performance will be evaluated using the area under the curve (AUC) of receiver operating characteristic (ROC), sensitivity and specificity, using internal and external validations. Discussion: Present studies reveal that early identification and classification of the SI-ARDS is essential to improve prognosis and disease management. Chest CT has been sought as a useful diagnostic tool to identify ARDS. However, when characteristic imaging findings were clearly presented, delays in diagnosis and treatment were impossible to avoid. In this ambispective cohort study, we hope to develop a novel model incorporating radiomic signatures and clinical signatures to provide an easy-to-use and individualized prediction of SI-ARDS occurrence and severe degree in patients at early stage.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)