Electrochemiluminescent CdS Quantum Dots Biosensor for Cancer Mutation Detection at Different Positions on Linear DNA Analytes.

PCR-based techniques routinely employed for the detection of mutated linear DNA molecules, including circulating tumor DNA (ctDNA), require large nucleotide sections on both sides of the mutation for primer annealing. This means that DNA fragments with a mutation positioned closer to the extremities are unlikely to be detected. Thus, sensors capable of recognizing linear DNA with characteristic mutations closer to the ends would be advantageous over the state-of-the-art approaches. Here, an electrochemiluminescence-resonance energy transfer (ECL-RET) biosensor comprising capped CdS quantum dots and hairpin DNA probes labeled with Au nanoparticles was developed for the detection of epidermal growth factor receptor (EGFR) ctDNA carrying the critical T790M lung cancer mutation. The ECL-RET system detected different DNA molecules including single-stranded 18-nucleotides (nt) and 40-nt as well as double-stranded 100-nt with the single nucleotide polymorphism (SNP) coding for T790M located either in the middle or only 7 nt from one end. For all target DNA, the sensor's limits of detection (LODs) were in the aM range, with excellent selectivity. It was the case of 100-nt target linear ctDNA fragments with LODs of 8.1 and 3.4 aM when the EGFR T790M SNP was either in the middle or at the end, respectively. These results show that ECL-RET systems can sense mutations in DNA fragments that would remain undetected by standard techniques.

Global burden of prostate cancer attributable to smoking among males in 204 countries and territories, 1990-2019.

INTRODUCTION: Understanding the latest global spatio-temporal pattern of prostate cancer burden attributable to smoking can help guide effective global health policy. This study aims to elucidate the trends in smoking-related prostate cancer from 1990 to 2019 using Global Burden of Disease (GBD) 2019 study data. METHODS: Data on prostate cancer attributable to smoking were extracted from Global Burden of Disease Study (GBD) 2019. The numbers and age-standardized rates on smoking-related prostate cancer mortality (ASMR) and disability-adjusted life years (ASDR) were analyzed by year, age, region, country, and socio-demographic index (SDI) level. Estimated annual percentage change (EAPC) was calculated to evaluate the temporal trends of ASMR and ASDR from 1990 to 2019. RESULTS: Of all prostate cancer deaths and DALYs globally in 2019, 6% and 6.6% were attributable to smoking, which contributed to 29,298 (95% CI 12,789 to 46,609) deaths and 571,590 (95% CI 253,490 to 917,820) disability-adjusted life-years (DALYs) in 2019. The number of smoking-related deaths and DALYs showed an upward trend, increasing by half from 1990 to 2019, while ASMR and ASDR declined in five sociodemographic indexes (SDI) regions, with the fastest decline in high SDI regions. For geographical regions, Western Europe and East Asia were the high-risk areas of prostate cancer deaths and DALYs attributable to smoking, among which China and the United States were the countries with the heaviest burden. The ASMR has decreased in all age groups, with the fastest decrease occurring in 75-79 years old. The ASMR or ASDR tended to increase in countries with the lowest SDI, but declined in countries with the highest SDI. The EAPC in ASMR or ASDR was highly negatively correlated with Human Development Index (HDI) in 2019, with coefficients 0.46. CONCLUSION: The number of smoking-related prostate cancer deaths and DALYs continued to increase globally, whereas its ASMR and ASDR have been decreasing. This substantial progress is particularly significant in developed regions and vary across geographic regions. Medical strategies to prevent and reduce the burden should be adjusted and implemented based on country-specific disease prevalence.

A novel technique to harvest bone autografts with mild local hyperthermia and enhanced osteogenic bone quality: a preclinical study in dogs.

BACKGROUND: Guided bone regeneration (GBR) involves collecting bone autografts with high bio-quality and efficiency. The current non-irrigated low-speed drilling has been limited for broader application in bone autograft harvest due to its low efficiency, inability to conduct buccal cortical perforation, and dependence on simultaneous implant placement. Increasing the drilling speed helps improve the efficiency but may incur thermal-mechanical bone damage. Most studies have addressed thermal reactions during bone drilling on non-vital models, which is irrelevant to clinical scenarios. Little has been known about bone's in vivo thermal profiles under non-irrigated higher-speed drilling and its influences on the resulting bone chips. AIM: A novel technique for bone harvest and cortical perforation via in-situ non-irrigated higher-speed drilling was proposed and investigated for the first time. METHODS: The third mandible premolars of eight beagles were extracted and healed for three months. Sixteen partial edentulous sites (left and right) were randomized into four groups for bone autograft harvest without irrigation: chisel, 50 rpm drilling, 500 rpm drilling, and 1000 rpm drilling. Bone chips were harvested on the buccal plates of the missing tooth. An infrared camera and an implantable thermocouple collaboratively monitored in vivo real-time bone temperature at the drilling sites. In vitro performances of cells from bone chips, including cell number, viability, proliferation, migration, ALP activity, in vitro mineralization, mRNA transcriptional level of osteogenic genes and heat shock protein 70 (HSP-70), and HSP-70 expression at the protein level were also studied. RESULTS: 500 rpm produced mild local hyperthermia with a 2-6 °C temperature rise both on the cortical surface and inside the cortical bone. It also held comparable or enhanced cell performances such as cell number, viability, proliferation, migration, ALP activity, in vitro mineralization, and osteogenic genes expression. CONCLUSIONS: In-situ non-irrigated higher-speed drilling at 500 rpm using a screw drill is versatile, efficient, and thermal friendly and improves the bio-quality of bone chips. Our novel technique holds clinical translational potential in GBR application.

Artificial intelligence for the prediction of acute kidney injury during the perioperative period: systematic review and Meta-analysis of diagnostic test accuracy.

BACKGROUND: Acute kidney injury (AKI) is independently associated with morbidity and mortality in a wide range of surgical settings. Nowadays, with the increasing use of electronic health records (EHR), advances in patient information retrieval, and cost reduction in clinical informatics, artificial intelligence is increasingly being used to improve early recognition and management for perioperative AKI. However, there is no quantitative synthesis of the performance of these methods. We conducted this systematic review and meta-analysis to estimate the sensitivity and specificity of artificial intelligence for the prediction of acute kidney injury during the perioperative period. METHODS: Pubmed, Embase, and Cochrane Library were searched to 2nd October 2021. Studies presenting diagnostic performance of artificial intelligence in the early detection of perioperative acute kidney injury were included. True positives, false positives, true negatives and false negatives were pooled to collate specificity and sensitivity with 95% CIs and results were portrayed in forest plots. The risk of bias of eligible studies was assessed using the PROBAST tool. RESULTS: Nineteen studies involving 304,076 patients were included. Quantitative random-effects meta-analysis using the Rutter and Gatsonis hierarchical summary receiver operating characteristics (HSROC) model revealed pooled sensitivity, specificity, and diagnostic odds ratio of 0.77 (95% CI: 0.73 to 0.81),0.75 (95% CI: 0.71 to 0.80), and 10.7 (95% CI 8.5 to 13.5), respectively. Threshold effect was found to be the only source of heterogeneity, and there was no evidence of publication bias. CONCLUSIONS: Our review demonstrates the promising performance of artificial intelligence for early prediction of perioperative AKI. The limitations of lacking external validation performance and being conducted only at a single center should be overcome. TRIAL REGISTRATION: This study was not registered with PROSPERO.

Subclassification of BI-RADS 4 Magnetic Resonance Lesions: A Systematic Review and Meta-Analysis.

OBJECTIVE: This research aims to investigate and evaluate the diagnostic efficacy of magnetic resonance imaging (MRI) in classifying Breast Imaging Reporting and Data System (BI-RADS) 4 lesions into subcategories: 4a, 4b, and 4c, so as to limit biopsies of suspected lesions in the breast. METHODS: The PubMed, Web of Science, Embase, and Cochrane Library foreign language databases were searched for literature published between January 2000 and July 2018. After analyzing the selection, data extraction, and quality assessment, a meta-analysis was performed, including data pooling, heterogeneity testing, and meta-regression. RESULTS: Fourteen articles, including 18 studies, met the inclusion criteria. The diagnostic efficacy of MRI for BI-RADS 4-weighted summary assay sensitivity and specificity were estimated at 0.95 [95% confidence interval (CI), 0.89-0.98] and 0.87 (95% CI, 0.81-0.91), respectively. The positive and negative likelihood ratios were 7.1 (95% CI, 4.7-10.7) and 0.06 (95% CI, 0.02-0.14), respectively. The diagnostic odds ratio was 126 (95% CI, 37-426), and the area under the receiver operating characteristic curve was 0.95 (95% CI, 0.93-0.97). The malignancy ratio of BI-RADS 4a, 4b, and 4c and malignancy range were 2.5% to 18.3%, 23.5% to 57.1%, and 58.0% to 95.2%, respectively. CONCLUSION: Risk stratification of suspected lesions (BI-RADS categories 4a, 4b, and 4c) can be achieved by MRI. The MRI is an effective auxiliary tool to further subclassify BI-RADS 4 lesions and prevent unnecessary biopsy of BI-RADS 4a lesions.

Magnetic Resonance Imaging Texture Analysis in Differentiating Benign and Malignant Breast Lesions of Breast Imaging Reporting and Data System 4: A Preliminary Study.

RATIONALE AND OBJECTIVES: This novel study aims to investigate texture parameters in distinguishing malignant and benign breast lesions classified as Breast Imaging Reporting and Data System 4 in dynamic contrast-enhanced magnetic resonance imaging (MRI). MATERIALS AND METHODS: This retrospective study included 203 patients with 136 breast cancer and 67 benign lesions who underwent breast MRI between November 23, 2016, and August 27, 2018. Co-occurrence matrix-based texture features were extracted from each lesion on T1-weighted contrast-enhanced MRI using MatLab software. The association between texture parameters and breast lesions was analyzed, and the diagnostic model for breast cancer was created. Classification performance was evaluated by the area under the receiver operating characteristic curve, sensitivity, and specificity. RESULTS: Significant differences were seen between malignant and benign lesions for a number of textural features, including contrast, correlation, autocorrelation, dissimilarity, cluster shade, and cluster performance (P < 0.05). After the analysis of the multicollinearity, 5 texture features (contrast, correlation, dissimilarity, cluster shade, and cluster performance) were included for the next principal component analysis. The differentiation accuracy of breast cancer based on the diagnostic model was 0.948 (95% confidence interval, 0.908-0.974). CONCLUSIONS: Texture features that measure randomness, heterogeneity, or homogeneity may reflect underlying growth patterns of breast lesions and show great difference in malignant and benign lesions. Therefore, texture analysis may be a valuable assisted tool for diagnostic analysis on breast.

CT and clinical characteristics that predict risk of EGFR mutation in non-small cell lung cancer: a systematic review and meta-analysis.

INTRODUCTION: To systematically analyze CT and clinical characteristics to find out the risk factors of epidermal growth factor receptor (EGFR) mutation in non-small cell lung cancer (NSCLC). Then the significant characteristics were used to set up a mathematic model to predict EGFR mutation in NSCLC. MATERIALS AND METHODS: PubMed, Web of Knowledge and EMBASE up to August 17, 2018 were systematically searched for relevant studies that investigated the evidence of association between CT and clinical characteristics and EGFR mutation in NSCLC. After study selection, data extraction, and quality assessment, the pooled odds ratios (ORs) were calculated. Then from May 2017 to August 2018, all NSCLC received EGFR mutation examination and CT examination in our hospital were chosen to test the prediction model by receiver operating characteristic (ROC) curves. RESULTS: Seventeen original studies met the inclusion criteria. The results showed that the ORs of ground-glass opacity (GGO), air bronchogram, pleural retraction, vascular convergence, smoking history, female gender were, respectively, 1.93 (P = 0.003), 2.09 (P = 0.03), 1.59 (P < 0.01), 1.61 (P = 0.001), 0.28 (P < 0.01), 0.35 (P < 0.01). The result of speculation, cavitation/bubble-like lucency, lesion shape, margin, pathological stage were, respectively, 1.19 (P = 0.32), 0.99 (P = 0.97), 0.82 (P = 0.42), 1.02 (P = 0.90), 0.77 (P = 0.30). 121 NSCLC received EGFR mutation test were included to test the prediction model. The mathematical model based on the results of meta-analysis was: 0.74 × air bronchogram + 0.46 × pleural retraction + 0.48 × vascular convergence - 1.27 × non-smoking history - 1.05 × female. The area under the ROC curve was 0.68. CONCLUSION: Based on the current evidence, GGO presence, air bronchogram, pleural retraction, vascular convergence were significant risk factors of EGFR mutation in NSCLC. And the prediction model can help to predict EGFR mutation status.

Prognostic value of the maximum standardized uptake value of pre-treatment primary lesions in small-cell lung cancer on 18F-FDG PET/CT: a meta-analysis.

Background 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG-PET/CT) has been widely used in clinical practice. However, the prognostic value of the pre-treatment standardized uptake value (SUV) for patients with small-cell lung cancer (SCLC) remains controversial. Purpose To investigate the prognostic role of pre-treatment 18F-FDG PET on SCLC patients by meta-analysis. Material and Methods Extensive literature searches of the PubMed, EMBASE, Web of Science, and Cochrane Library databases were conducted to identify literature published until 5 May 2017. Comparative analyses of the pooled hazard ratios (HRs) for event-free survival (EFS) and overall survival (OS) were performed to assess their correlations with the pre-treatment maximum SUV (SUVmax). Either the fixed- or the random-effects model was adopted, depending on the heterogeneity observed across the studies. Subgroup analyses were performed to assess the robustness of the results. Results Twelve studies with 1062 patients were included. The pooled HR for OS of 11 studies was 1.13 (95% confidence interval [CI] = 1.05-1.22; P = 0.001; I2 = 0%) and the pooled HR for EFS of nine studies was 1.09 (95% CI = 1.02-1.17; P = 0.014; I2 = 0%), indicating that patients with high SUVs may have poorer prognoses. Begg's test detected no significant publication bias. The prognostic role of the SUVmax remained similar in the subgroup analyses. Conclusion Our meta-analysis indicated that the pre-treatment SUVmax of primary lesions can be an important prognostic factor for OS and EFS in patients with SCLC. A high SUVmax may indicate poorer prognosis.

Ultrasound, computed tomography or magnetic resonance imaging - which is preferred for acute appendicitis in children? A Meta-analysis.

BACKGROUND: There is no established consensus about the relative accuracies of US, CT and MRI in childhood appendicitis. OBJECTIVE: To compare, through meta-analysis, the accuracies of US, CT and MRI for clinically suspected acute appendicitis in children. MATERIALS AND METHODS: PubMed, Embase, Web of Science and the Cochrane Library were searched. After study selection, data extraction and quality assessment, the sensitivity, specificity and the area under the curve of summary receiver operating characteristic were calculated and compared. RESULTS: Twenty-seven articles including 29 studies met the inclusion criteria, including 19 studies (9,170 patients) of US, 6 studies (928 patients) of CT and 4 studies (990 patients) of MRI. The analysis showed that the area under the receiver operator characteristics curve of MRI (0.995) was a little higher than that of US (0.987) and CT (0.982; P > 0.05). CONCLUSION: US, CT and MRI have high diagnostic accuracies of clinically suspected acute appendicitis in children overall with no significant difference.

18 F-FDG PET/CT Imaging of Isolated Myeloid Sarcoma in the Right Humerus.

ABSTRACT: Myeloid sarcoma is a neoplastic mass formed by the infiltration of primitive or immature myeloid cells into organs and tissues outside the bone marrow. It may occur before, at the same time, or manifest as the recurrence of acute myeloid leukemia, myelodysplastic syndromes, and chronic myeloproliferative syndromes. It may involve any organ or tissue, including skin, soft tissue, lymph nodes, and gastrointestinal tract and bone. Isolated humerus involvement is extremely rare. Herein, we present the FDG PET/CT findings of a rare case of isolate myeloid sarcoma in the right humerus, which showed only increased bone density with moderate FDG uptake.

Study on AEMs with Excellent Comprehensive Performance Prepared by Covalently Cross-Linked p-Triphenyl with SEBS Remotely Grafted Piperidine Cations.

A series of SEBS-C6-PIP-yPTP (y = 0-15%) AEMs with good mechanical and chemical stability were prepared by combining the strong rigidity of p-triphenyl, good toughness of SEBS, and excellent stability of PIP cations. After the introduction of a p-triphenyl polymer into the main chain, a clear hydrophilic-hydrophobic phase separation structure was constructed within the membrane, forming a continuous and interconnected ion transport channel to improve ion transport efficiency. Moreover, the molecular chains of the cross-linked AEMs change from chain-like to network-like, and the tighter binding between each molecule increases the tensile strength. The special structure of the six-membered ring makes PIP have a significant constraint effect; when nucleophilic substitution and Hoffman elimination occur at the α and ß positions, the required transition state potential energy increases, making the reaction difficult to occur and improving the alkaline stability of the polymer membrane. The SEBS-C6-PIP-15%PTP membrane has the best mechanical properties (Ts = 38.79 MPa, Eb = 183.09% at 80 °C, 100% RH), the highest ion conductivity (102.02 mS. cm-1 at 80 °C), and the best alkaline stability (6.23% degradation at 80 °C in a 2 M NaOH solution for 1400 h). It can be seen that organic-organic covalent cross-linking is an effective means to improve the comprehensive performance of AEMs.

Correlation between carcinoembryonic antigen (CEA) expression and EGFR mutations in non-small-cell lung cancer: a meta-analysis.

OBJECTIVES: The purpose of this meta-analysis was to investigate the relationship between serum carcinoembryonic antigen (CEA) expression and epidermal growth factor receptor (EGFR) mutation status in non-small cell lung cancer (NSCLC). METHODS: Databases such as PubMed, Cochrane, EMBASE and Google Scholar were systematically searched to identify studies assessing the association of serum CEA expression with EGFR mutations. Across 19 studies, 4168 patients were included between CEA expression and EGFR mutations odds ratio (OR) conjoint analysis of correlations. RESULTS: Compared with CEA-negative NSCLC, CEA-positive tumors had an increased EGFR mutation rate (OR = 1.85, 95% confidence interval: 1.48-2.32, P < 0.00001). This association was observed in both stage IIIB/IV patients (OR = 1.60, 95% CI: 1.18-2.15, P = 0.002) and stage I-IIIA (OR = 1.67, 95% CI: 1.01-2.77, P = 0.05) patients. In addition, CEA expression was associated with exon 19 (OR = 1.97, 95% CI: 1.25-3.11, P = 0.003) and exon 21 (OR = 1.51, 95% CI: 1.07-2.12, P = 0.02) EGFR mutations. In ADC pathological type had also showed the correlation (OR = 1.84, 95% CI: 1.31-2.57, P = 0.0004). CONCLUSIONS: This meta-analysis indicated that serum CEA expression was associated with EGFR mutations in NSCLC patients. The results of this study suggest that CEA level may play a predictive role in the EGFR mutation status of NSCLC patients. Detecting serum CEA expression levels can give a good suggestion to those patients who are confused about whether to undergo EGFR mutation tests. Moreover, it may help better plan of the follow-up treatment.

Achieving optimal trade-off for student dropout prediction with multi-objective reinforcement learning.

Student dropout prediction (SDP) in educational research has gained prominence for its role in analyzing student learning behaviors through time series models. Traditional methods often focus singularly on either prediction accuracy or earliness, leading to sub-optimal interventions for at-risk students. This issue underlines the necessity for methods that effectively manage the trade-off between accuracy and earliness. Recognizing the limitations of existing methods, this study introduces a novel approach leveraging multi-objective reinforcement learning (MORL) to optimize the trade-off between prediction accuracy and earliness in SDP tasks. By framing SDP as a partial sequence classification problem, we model it through a multiple-objective Markov decision process (MOMDP), incorporating a vectorized reward function that maintains the distinctiveness of each objective, thereby preventing information loss and enabling more nuanced optimization strategies. Furthermore, we introduce an advanced envelope Q-learning technique to foster a comprehensive exploration of the solution space, aiming to identify Pareto-optimal strategies that accommodate a broader spectrum of preferences. The efficacy of our model has been rigorously validated through comprehensive evaluations on real-world MOOC datasets. These evaluations have demonstrated our model's superiority, outperforming existing methods in achieving optimal trade-off between accuracy and earliness, thus marking a significant advancement in the field of SDP.

Pathogenic factors of maxillary sinus mucosal thickening observed by cone-beam computed tomography (CBCT).

OBJECTIVE: To explore the pathogenic factors associated with maxillary sinus mucosal thickening with Cone-beam computed Tomography (CBCT). METHODS: From 2016 through 2020, 93 patients with periapical periodontitis or periodontitis in the maxillary posterior dental region were selected. RESULTS: The preoperative thickness of the periodontitis group was significantly higher than that of the periapical periodontitis group (P < 0.05). The difference achieves statistical significance for the comparison of the thickness change with various severity of inflammation (F = 54.824, P = 0.000), the change with time (F = 312.741, P = 0.000). and the change with the interaction severity of inflammation and time（F = 86.132, P = 0.000). CONCLUSIONS: Patients with maxillary sinus mucosa thickening caused by periodontitis and periapical periodontitis should be extracted their infectious teeth and get thoroughly debridement. Maxillary sinus augmentation can perform favorable efforts 3-6 months after extracting teeth.

Risk factors for air embolism following computed tomography-guided percutaneous transthoracic needle biopsy: a systematic review and meta-analysis

To quantitatively analyze the risk factors for air embolism following computed tomography (CT)-guided percutaneous transthoracic needle biopsy (PTNB) and qualitatively review their characteristics. The databases of PubMed, Embase, Web of Science, Wanfang Data, VIP information, and China National Knowledge Infrastructure were searched on January 4, 2021, for studies reporting the occurrence of air embolisms following CT-guided PTNB. After study selection, data extraction, and quality assessment, the characteristics of the included cases were qualitatively and quantitatively analyzed. A total of 154 cases of air embolism following CT-guided PTNB were reported. The reported incidence was 0.06% to 4.80%, and 35 (22.73%) patients were asymptomatic. An unconscious or unresponsive state was the most common symptom (29.87%). Air was most commonly found in the left ventricle (44.81%), and 104 (67.53%) patients recovered without sequelae. Air location (P < 0.001), emphysema (P = 0.061), and cough (P = 0.076) were associated with clinical symptoms. Air location (P = 0.015) and symptoms (P < 0.001) were significantly associated with prognosis. Lesion location [odds ratio (OR): 1.85, P = 0.017], lesion subtype (OR: 3.78, P = 0.01), pneumothorax (OR: 2.16, P = 0.003), hemorrhage (OR: 3.20, P < 0.001), and lesions located above the left atrium (OR: 4.35, P = 0.042) were significant risk factors for air embolism. Based on the current evidence, a subsolid lesion, being located in the lower lobe, the presence of pneumothorax or hemorrhage, and lesions located above the left atrium were significant risk factors for air embolism.

Hotspots and frontiers in PSMA research for prostate cancer: a bibliometric and visualization analysis over the past 20 years.

BACKGROUND: Prostate-specific membrane antigen (PSMA)-targeted imaging and therapy have significantly changed the management of patients with prostate cancer (PCa) at different disease stages. This advancement has attracted the attention of scholars, leading to a prolific output of scholarly publications. This study comprehensively outlines the knowledge framework associated with PSMA-based diagnosis and treatment of PCa through the application of bibliometric analysis, and discusses the potential research trends and foci. METHODS: Articles and reviews related to PSMA for prostate cancer from 2003 to 2022 were retrieved from Web of Science Core Collection. VOSviewer, Citespace, and R-bibliometrix were primarily employed to execute and visually represent co-authorship, co-citation, and co-occurrence analysis of countries, institutions, authors, references and keywords in this field. RESULTS: A total of 3830 papers were included. The papers on the field of PSMA-based PCa therapy and imaging had been continuously increased since 2003, but the rate has slowed from 2020. The United States made the largest contribution in this field, in terms of publications 997 (26.03%), H-index (110) and total citations (53,167 times). We identified the most productive institution were Technical University of Munich, and Australian institutions had become very active in recent years. Journal of Nuclear Medicine was the most prominent journal in this field. Professors Matthias Eiber and Martin G Pomper made great achievements, while Ali Afshar-Oromieh was the most co-cited author. According to the result of keywords and topics analysis, "ga-68 labeled psma ligand", "radiation dosimetry" and "HBED-CC" were major research areas in the near future, while "Extended pelvic lymph node dissection" was considered to be the future research foci. CONCLUSIONS: The field of psma-based PCa therapy and imaging is in the stage of vigorous development and has a bright prospect. The United States and Germany have achieved outstanding results in this area, while Australia has recently developed rapidly. It is foreseeable that more research foci will be lied in the early detection of pelvic lymph nodes and the multimodal imaging-guided surgery.

Predicting N2 lymph node metastasis in presurgical stage I-II non-small cell lung cancer using multiview radiomics and deep learning method.

BACKGROUND: Accurate diagnosis of N2 lymph node status of the resectable stage I-II non-small cell lung cancer (NSCLC) before surgery is crucial, while there is lack of corresponding method clinically. PURPOSE: To develop and validate a model to quantitively predict the N2 lymph node metastasis in presurgical clinical stage I-II NSCLC using multiview radiomics and deep learning method. METHODS: In this study, 140 NSCLC patients were enrolled and randomly divided into training and test sets. Univariate and multiple analysis method were used step by step to establish the clinical model; Then a multiview radiomics modeling scheme was designed, in which the optimal input feature set was determined by subcategorizing radiomics features (C1: original; C2: LoG and C3: wavelet) and comparison of corresponding radiomics model. The minimum-redundancy maximum-relevance (mRMR) selection and the least absolute shrinkage and selection operator (LASSO) algorithm were used for the feature selection and construction of each radiomics model (Rad). Next, an end-to-end ResNet18 architecture and transfer learning techniques were designed to construct a deep learning model (DL). Subsequently, the screened clinical risk factors and constructed Rad and DL models were combined and compared and a nomogram was constructed. Finally, the diagnostic performance of all constructed models were evaluated and compared using receiver operating characteristic curve (ROC) analysis, Delong test, Calibration analysis, Hosmer-Lemeshow test, and decision curves, respectively. RESULTS: Carcinoma embryonic antigen (CEA) level and spiculation were screened to make up the Clinical model, while seven radiomics features in the optimal input feature set C2 + C3 were selected to construct the Rad. DL was constructed by training on 1.8 million natural images and small sample data of our N2 lymph node volume of interest (VOI) images. Except for the Clinical model, all other models showed good predictive accuracy and consistency in both training set and test set. DL (area under curve (AUC): 0.83) was better than Rad (AUC: 0.76) in predictive accuracy, but their difference was not significant (p = 0.45). The combined models showed better diagnostic performance than the model only clinical or image risk factors were used (AUC for Clinical, Rad + DL, Rad + Clinical, DL + Clinical, and Rad + DL + Clinical were respectively 0.66, 0.86, 0.82, 0.86, and 0.88). Finally, the Rad + DL + Clinical model with the best diagnostic performance was selected to draw the final nomogram for clinical use. CONCLUSION: This study proposes a nomogram based on multiview radiomics, deep learning, and clinical features that can be efficiently used to quantitively predict presurgical N2 diseases in patients with clinical stage I-II NSCLC.

A nomogram predicting the severity of COVID-19 based on initial clinical and radiologic characteristics.

Aim: This study aimed to build an easy-to-use nomogram to predict the severity of COVID-19. Patients & methods: From December 2019 to January 2020, patients confirmed with COVID-19 in our hospital were enrolled. The initial clinical and radiological characteristics were extracted. Univariate and multivariate logistic regression were used to identify variables for the nomogram. Results: In total, 104 patients were included. Based on statistical analysis, age, levels of neutrophil count, creatinine, procalcitonin and numbers of involved lung segments were identified for nomogram. The area under the curve was 0.939 (95% CI: 0.893-0.984). The calibration curve showed good agreement between prediction of nomogram and observation in the primary cohort. Conclusion: An easy-to-use nomogram with great discrimination was built to predict the severity of COVID-19.

CT-Guided Percutaneous Core Needle Biopsy in Typing and Subtyping Lung Cancer: A Comparison to Surgery.

Background: Lung cancer histologic types and subtypes are closely associated with treatment selection and prognosis prediction. In this study, we aim to evaluate the suitability of computed tomography-guided percutaneous core needle biopsy (CT-guided PCNB) in typing and subtyping lung cancer. Methods: From August 2007 to December 2015, the patients who underwent CT-guided PCNB and lung lesion resection were retrospectively collected and analyzed. All pathological sections were reassessed in consensus by 2 junior pathologists (group A) and 2 senior pathologists (group B), respectively. All cases were diagnosed on 3 levels: first, malignant and benign diagnosis; second, histologic types diagnosis; and third, histologic subtypes diagnosis and compared with surgery results. Pearson chi-square test was used to compare the differences of diagnostic accuracy between pathologists in group A and group B. Results: A cohort of 160 patients was included in this study. On the first level, the diagnostic accuracy was 90.63% (group A) and 94.38% (group B), (P = .20). On the second level, the diagnostic accuracy for malignant lesions, adenocarcinoma (ADC), and squamous cell carcinoma (SQC) were, respectively, 72.66%, 84.72%, and 69.05% (group A) and 76.98%, 90.28%, and 71.43% (group B) (P > .05). On the third level, the diagnostic accuracy for ADC subtypes were 26.39% (group A) and 55.56% (group B) (P < 0.01); for SQC subtypes were 28.57% (group A) and 38.10% (group B) (P = 0.36). Conclusion: Small specimens obtained by CT-guided PCNB were suitable for the diagnosis of lung cancer histologic types, which may contribute to the selection of a suitable treatment strategy for the unresectable lung cancers. While for the diagnosis of subtypes, discussion with experienced pathologists was recommended.

A Predictive Model to Differentiate Between Second Primary Lung Cancers and Pulmonary Metastasis.

RATIONALE AND OBJECTIVES: To develop and validate a nomogram for differentiating second primary lung cancers (SPLCs) from pulmonary metastases (PMs). MATERIALS AND METHODS: A total of 261 lesions from 253 eligible patients were included in this study. Among them, 195 lesions (87 SPLCs and 108 PMs) were used in the training cohort to establish the diagnostic model. Twenty-one clinical or imaging features were used to derive the model. Sixty-six lesions (32 SPLCs and 34 PMs) were included in the validation set. RESULTS: After analysis, age, lesion distribution, type of lesion, air bronchogram, contour, spiculation, and vessel convergence sign were considered to be significant variables for distinguishing SPLCs from PMs. Subsequently, these variables were selected to establish a nomogram. The model showed good distinction in the training set (area under the curve = 0.97) and the validation set (area under the curve = 0.92). CONCLUSION: This study found that the nomogram calculated from clinical and radiological characteristics could accurately classify SPLCs and PMs.