Development and validation of prediction models for papillary thyroid cancer structural recurrence using machine learning approaches.

BACKGROUND: Although papillary thyroid cancer (PTC) patients are known to have an excellent prognosis, up to 30% of patients experience disease recurrence after initial treatment. Accurately predicting disease prognosis remains a challenge given that the predictive value of several predictors remains controversial. Thus, we investigated whether machine learning (ML) approaches based on comprehensive predictors can predict the risk of structural recurrence for PTC patients. METHODS: A total of 2244 patients treated with thyroid surgery and radioiodine were included. Twenty-nine perioperative variables consisting of four dimensions (demographic characteristics and comorbidities, tumor-related variables, lymph node (LN)-related variables, and metabolic and inflammatory markers) were analyzed. We applied five ML algorithms-logistic regression (LR), support vector machine (SVM), extreme gradient boosting (XGBoost), random forest (RF), and neural network (NN)-to develop the models. The area under the receiver operating characteristic (AUC-ROC) curve, calibration curve, and variable importance were used to evaluate the models' performance. RESULTS: During a median follow-up of 45.5 months, 179 patients (8.0%) experienced structural recurrence. The non-stimulated thyroglobulin, LN dissection, number of LNs dissected, lymph node metastasis ratio, N stage, comorbidity of hypertension, comorbidity of diabetes, body mass index, and low-density lipoprotein were used to develop the models. All models showed a greater AUC (AUC = 0.738 to 0.767) than did the ATA risk stratification (AUC = 0.620, DeLong test: P < 0.01). The SVM, XGBoost, and RF model showed greater sensitivity (0.568, 0.595, 0.676), specificity (0.903, 0.857, 0.784), accuracy (0.875, 0.835, 0.775), positive predictive value (PPV) (0.344, 0.272, 0.219), negative predictive value (NPV) (0.959, 0.959, 0.964), and F1 score (0.429, 0.373, 0.331) than did the ATA risk stratification (sensitivity = 0.432, specificity = 0.770, accuracy = 0.742, PPV = 0.144, NPV = 0.938, F1 score = 0.216). The RF model had generally consistent calibration compared with the other models. The Tg and the LNR were the top 2 important variables in all the models, the N stage was the top 5 important variables in all the models. CONCLUSIONS: The RF model achieved the expected prediction performance with generally good discrimination, calibration and interpretability in this study. This study sheds light on the potential of ML approaches for improving the accuracy of risk stratification for PTC patients. TRIAL REGISTRATION: Retrospectively registered at www.chictr.org.cn (trial registration number: ChiCTR2300075574, date of registration: 2023-09-08).

Identification of ipsilateral supraclavicular lymph node metastasis in breast cancer based on LASSO regression with a high penalty factor.

Aiming at the problems of small sample size and large feature dimension in the identification of ipsilateral supraclavicular lymph node metastasis status in breast cancer using ultrasound radiomics, an optimized feature combination search algorithm is proposed to construct linear classification models with high interpretability. The genetic algorithm (GA) is used to search for feature combinations within the feature subspace using least absolute shrinkage and selection operator (LASSO) regression. The search is optimized by applying a high penalty to the L1 norm of LASSO to retain excellent features in the crossover operation of the GA. The experimental results show that the linear model constructed using this method outperforms those using the conventional LASSO regression and standard GA. Therefore, this method can be used to build linear models with higher classification performance and more robustness.

A nomogram to preoperatively predict the aggressiveness of non-functional pancreatic neuroendocrine tumors based on CT features.

OBJECTIVES: To develop a nomogram to predict the aggressiveness of non-functional pancreatic neuroendocrine tumors (NF-pNETs) based on preoperative computed tomography (CT) features. METHODS: This study included 176 patients undergoing radical resection for NF-pNETs. These patients were randomly divided into the training (n = 123) and validation sets (n = 53). A nomogram was developed based on preoperative predictors of aggressiveness of the NF-pNETs which were identified by univariable and multivariable logistic regression analysis. The aggressiveness of NF-pNETs was defined as a composite measure including G3 grading, N+, distant metastases, and/ or disease recurrence. RESULTS: Altogether, the number of patients with highly aggressive NF-pNETs was 37 (30.08 %) and 15 (28.30 %) in the training and validation sets, respectively. Multivariable logistic regression analysis identified that tumor size, biliopancreatic duct dilatation, lymphadenopathy, and enhancement pattern were preoperative predictors of aggressiveness. Those variables were used to develop a nomogram with good concordance statistics of 0.89 and 0.86 for predicting aggressiveness in the training and validation sets, respectively. With a nomogram score of 59, patients with NF-pNETs were divided into low-aggressive and high-aggressive groups. The high-aggressive group had decreased overall survival (OS) and disease-free survival (DFS). Moreover, the nomogram showed good performance in predicting OS and DFS at 3, 5, and 10 years. CONCLUSION: The nomogram integrating CT features helped preoperatively predict the aggressiveness of NF-pNETs and could potentially facilitate clinical decision-making.

Comprehensive split TEV based protein-protein interaction screening reveals TAOK2 as a key modulator of Hippo signalling to limit growth.

The conserved Hippo signalling pathway plays a crucial role in tumour formation by limiting tissue growth and proliferation. At the core of this pathway are tumour suppressor kinases STK3/4 and LATS1/2, which limit the activity of the oncogene YAP1, the primary downstream effector. Here, we employed a split TEV-based protein-protein interaction screen to assess the physical interactions among 28 key Hippo pathway components and potential upstream modulators. This screen led us to the discovery of TAOK2 as pivotal modulator of Hippo signalling, as it binds to the pathway's core kinases, STK3/4 and LATS1/2, and leads to their phosphorylation. Specifically, our findings revealed that TAOK2 binds to and phosphorylates LATS1, resulting in the reduction of YAP1 phosphorylation and subsequent transcription of oncogenes. Consequently, this decrease led to a decrease in cell proliferation and migration. Interestingly, a correlation was observed between reduced TAOK2 expression and decreased patient survival time in certain types of human cancers, including lung and kidney cancer as well as glioma. Moreover, in cellular models corresponding to these cancer types the downregulation of TAOK2 by CRISPR inhibition led to reduced phosphorylation of LATS1 and increased proliferation rates, supporting TAOK2's role as tumour suppressor gene. By contrast, overexpression of TAOK2 in these cellular models lead to increased phospho-LATS1 but reduced cell proliferation. As TAOK2 is a druggable kinase, targeting TAOK2 could serve as an attractive pharmacological approach to modulate cell growth and potentially offer strategies for combating cancer.

The predictive significance of lipid accumulation products for future diabetes in a non-diabetic population from a gender perspective: an analysis using time-dependent receiver operating characteristics.

Objective: The increasing prevalence of diabetes is strongly associated with visceral adipose tissue (VAT), and gender differences in VAT remarkably affect the risk of developing diabetes. This study aimed to assess the predictive significance of lipid accumulation products (LAP) for the future onset of diabetes from a gender perspective. Methods: A total of 8,430 male and 7,034 female non-diabetic participants in the NAGALA (NAfld in the Gifu Area, Longitudinal Analysis) program were included. The ability of LAP to assess the risk of future new-onset diabetes in both genders was analyzed using multivariate Cox regression. Subgroup analysis was conducted to explore the impact of potential modifiers on the association between LAP and diabetes. Additionally, time-dependent receiver operator characteristics (ROC) curves were used to assess the predictive power of LAP in both genders for new-onset diabetes over the next 2-12 years. Results: Over an average follow-up of 6.13 years (maximum 13.14 years), 373 participants developed diabetes. Multivariate Cox regression analysis showed a significant gender difference in the association between LAP and future diabetes risk (P-interaction<0.05): the risk of diabetes associated with LAP was greater in females than males [hazard ratios (HRs) per standard deviation (SD) increase: male 1.20 (1.10, 1.30) vs female 1.35 (1.11, 1.64)]. Subgroup analysis revealed no significant modifying effect of factors such as age, body mass index (BMI), smoking history, drinking history, exercise habits, and fatty liver on the risk of diabetes associated with LAP (All P-interaction <0.05). Time-dependent ROC analysis showed that LAP had greater accuracy in predicting diabetes events occurring within the next 2-12 years in females than males with more consistent predictive thresholds in females. Conclusions: This study highlighted a significant gender difference in the association between LAP and future diabetes risk. The risk of diabetes associated with LAP was greater in females than in males. Furthermore, LAP showed superior predictive ability for diabetes at different time points in the future in females and had more consistent and stable predictive thresholds in females, particularly in the medium and long term.

A cancer cell membrane coated, doxorubicin and microRNA co-encapsulated nanoplatform for colorectal cancer theranostics.

Endogenous microRNAs (miRNA) in tumors are currently under exhaustive investigation as potential therapeutic agents for cancer treatment. Nevertheless, RNase degradation, inefficient and untargeted delivery, limited biological effect, and currently unclear side effects remain unsettled issues that frustrate clinical application. To address this, a versatile targeted delivery system for multiple therapeutic and diagnostic agents should be adapted for miRNA. In this study, we developed membrane-coated PLGA-b-PEG DC-chol nanoparticles (m-PPDCNPs) co-encapsulating doxorubicin (Dox) and miRNA-190-Cy7. Such a system showed low biotoxicity, high loading efficiency, and superior targeting ability. Systematic delivery of m-PPDCNPs in mouse models showed exceptionally specific tumor accumulation. Sustained release of miR-190 inhibited tumor angiogenesis, tumor growth, and migration by regulating a large group of angiogenic effectors. Moreover, m-PPDCNPs also enhanced the sensitivity of Dox by suppressing TGF-ß signal in colorectal cancer cell lines and mouse models. Together, our results demonstrate a stimulating and promising m-PPDCNPs nanoplatform for colorectal cancer theranostics.

Differential Diagnosis of DCIS and Fibroadenoma Based on Ultrasound Images: a Difference-Based Self-Supervised Approach.

Differentiation of ductal carcinoma in situ (DCIS, a precancerous lesion of the breast) from fibroadenoma (FA) using ultrasonography is significant for the early prevention of malignant breast tumors. Radiomics-based artificial intelligence (AI) can provide additional diagnostic information but usually requires extensive labeling efforts by clinicians with specialized knowledge. This study aims to investigate the feasibility of differentially diagnosing DCIS and FA using ultrasound radiomics-based AI techniques and further explore a novel approach that can reduce labeling efforts without sacrificing diagnostic performance. We included 461 DCIS and 651 FA patients, of whom 139 DCIS and 181 FA patients constituted a prospective test cohort. First, various feature engineering-based machine learning (FEML) and deep learning (DL) approaches were developed. Then, we designed a difference-based self-supervised (DSS) learning approach that only required FA samples to participate in training. The DSS approach consists of three steps: (1) pretraining a Bootstrap Your Own Latent (BYOL) model using FA images, (2) reconstructing images using the encoder and decoder of the pretrained model, and (3) distinguishing DCIS from FA based on the differences between the original and reconstructed images. The experimental results showed that the trained FEML and DL models achieved the highest AUC of 0.7935 (95% confidence interval, 0.7900-0.7969) on the prospective test cohort, indicating that the developed models are effective for assisting in differentiating DCIS from FA based on ultrasound images. Furthermore, the DSS model achieved an AUC of 0.8172 (95% confidence interval, 0.8124-0.8219), indicating that our model outperforms the conventional radiomics-based AI models and is more competitive.

Wavelet transformation can enhance computed tomography texture features: a multicenter radiomics study for grade assessment of COVID-19 pulmonary lesions.

Background: This study set out to develop a computed tomography (CT)-based wavelet transforming radiomics approach for grading pulmonary lesions caused by COVID-19 and to validate it using real-world data. Methods: This retrospective study analyzed 111 patients with 187 pulmonary lesions from 16 hospitals; all patients had confirmed COVID-19 and underwent non-contrast chest CT. Data were divided into a training cohort (72 patients with 127 lesions from nine hospitals) and an independent test cohort (39 patients with 60 lesions from seven hospitals) according to the hospital in which the CT was performed. In all, 73 texture features were extracted from manually delineated lesion volumes, and 23 three-dimensional (3D) wavelets with eight decomposition modes were implemented to compare and validate the value of wavelet transformation for grade assessment. Finally, the optimal machine learning pipeline, valuable radiomic features, and final radiomic models were determined. The area under the receiver operating characteristic (ROC) curve (AUC), calibration curve, and decision curve were used to determine the diagnostic performance and clinical utility of the models. Results: Of the 187 lesions, 108 (57.75%) were diagnosed as mild lesions and 79 (42.25%) as moderate/severe lesions. All selected radiomic features showed significant correlations with the grade of COVID-19 pulmonary lesions (P<0.05). Biorthogonal 1.1 (bior1.1) LLL was determined as the optimal wavelet transform mode. The wavelet transforming radiomic model had an AUC of 0.910 in the test cohort, outperforming the original radiomic model (AUC =0.880; P<0.05). Decision analysis showed the radiomic model could add a net benefit at any given threshold of probability. Conclusions: Wavelet transformation can enhance CT texture features. Wavelet transforming radiomics based on CT images can be used to effectively assess the grade of pulmonary lesions caused by COVID-19, which may facilitate individualized management of patients with this disease.

A multifunctional cascade bioreactor based on a layered double oxides composite hydrogel for synergetic tumor chemodynamic/starvation/photothermal therapy.

The field of nanomedicine-catalyzed tumor therapy has achieved a lot of progress; however, overcoming the limitations of the tumor microenvironment (TME) to achieve the desired therapeutic effect remains a major challenge. In this study, a nanocomposite hydrogel (GH@LDO) platform combining the nanozyme CoMnFe-layered double oxides (CoMnFe-LDO) and natural enzyme glucose oxidase (GOX) was engineered to remodel the TME to enhance tumor catalytic therapy. The CoMnFe-LDO is a nanozyme that can convert endogenous H2O2 into reactive oxygen species (ROS) and O2 to achieve chemodynamic therapy (CDT) and alleviate the hypoxic microenvironment. Meanwhile, GOX can catalyze the conversion of glucose and O2 to gluconic acid and H2O2, which not only represses the ATP production of tumor cells to achieve starvation therapy (ST), but also decreases the pH value of TME and supplies extra H2O2 to enhance the CDT effect. Furthermore, this well-designed CoMnFe-LDO possessed a high photothermal conversion efficiency of GH@LDO (66.63%), which could promote the generation of ROS to enhance the CDT effect and achieve photothermal therapy (PTT) under near-infrared light irradiation. The GH@LDO hydrogel performes cascade reaction which overcomes the limitation of the TME and achieves satisfactory CDT/ST/PTT synergetic effects in vitro and in vivo. This work provides a new strategy for remodeling the TME using nanomedicine to achieve precise tumor cascaded catalytic therapy. STATEMENT OF SIGNIFICANCE: At present, the focus of tumor therapy has begun to shift from monotherapy to combination therapy for improving the overall therapeutic effect. In this study, we synthesized a CoMnFe-LDO nanozyme composed of multiple transition metal oxides, which demonstrated improved peroxidase and oxidase activities as well as favorable photothermal conversion capability. The CoMnFe-LDO nanozyme was compounded with an injectable GH hydrogel crosslinked by GOX and horseradish peroxidase (HRP). This nanocomposite hydrogel overcame the limitations of weak acidity, H2O2, and O2 levels in the TME and achieved synergetic CDT, ST, and PTT effects based on the cascaded catalytic actions of CoMnFe-LDO and GOX to H2O2 and glucose.

Comparison of dyslipidemia incidence in Chinese early-stage breast cancer patients following different endocrine therapies: A population-based cohort study.

Background: There is lack of large-scale real-world research evidence showing the impact of endocrine therapy on blood lipids in Chinese breast cancer patients, especially those with premenopausal breast cancer. Based on a large breast cancer cohort at West China Hospital, we aimed to compare the risk of dyslipidemia between premenopausal and postmenopausal women based on the endocrine therapy used. Methods: A total of 1,883 early-stage breast cancer (EBC) patients who received endocrine monotherapy [selective estrogen receptor modulator (SERM) and aromatase inhibitor (AI), with or without ovarian function suppression] with normal blood lipid levels at baseline were retrospectively included between October 2008 and April 2017. Dyslipidemia was defined as an abnormality in cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein, and total cholesterol (TC) levels. The risk accumulation function was used to calculate the incidence of dyslipidemia in order to assess the absolute risk, while the multivariate Cox regression model was used to calculate the relative risk of dyslipidemia between the groups. Results: Patients with EBC were followed up for 60 months to monitor their blood lipid levels. The accumulated 5-year incidence of dyslipidemia in postmenopausal patients was higher than that in premenopausal patients (adjusted HR [95% confidence interval], 1.25 [1.01-1.56], 41.7% vs. 31.2%, p = 0.045). In premenopausal patients, the risk of abnormal TC was significantly higher in the OFS+AI group compared with that in the SERM group (adjusted HR [95% CI], 6.24 [3.19-12.20], p < 0.001, 5-year abnormal rates: 21.5% vs. 2.4%), and that of abnormal LDL-C level also increased (adjusted HR [95% CI], 10.54 [3.86-28.77], p < 0.001, 5-year abnormal rates: 11.1% vs. 0.9%). In postmenopausal patients, the risk of abnormal TC or LDL-C levels showed a similar trend in the AI and SERM groups. Conclusions: In addition to postmenopausal patients, dyslipidemia is also common in premenopausal Chinese patients with EBC who received endocrine therapy. Irrespective of menopausal status, AI treatment increases the risk of TC/LDL-C dyslipidemia than SERM treatment.

A superlattice composite of Zn-Fe layered double hydroxide and graphene oxide for antitumor application.

A superlattice composite of Zn-Fe layered double hydroxide and graphene oxide was fabricated on the titanium surface and showed lamellar morphology. It was found for the first time that this superlattice composite could inhibit cell adhesion and proliferation, and cause cell death of the cholangiocarcinoma cell line RBE cells in vitro and show tumor inhibition effect in vivo.

Development and Validation of a Novel Radiomics-Based Nomogram With Machine Learning to Preoperatively Predict Histologic Grade in Pancreatic Neuroendocrine Tumors.

Backgroud: Tumor grade is the determinant of the biological aggressiveness of pancreatic neuroendocrine tumors (PNETs) and the best current tool to help establish individualized therapeutic strategies. A noninvasive way to accurately predict the histology grade of PNETs preoperatively is urgently needed and extremely limited. Methods: The models training and the construction of the radiomic signature were carried out separately in three-phase (plain, arterial, and venous) CT. Mann-Whitney U test and least absolute shrinkage and selection operator (LASSO) were applied for feature preselection and radiomic signature construction. SVM-linear models were trained by incorporating the radiomic signature with clinical characteristics. An optimal model was then chosen to build a nomogram. Results: A total of 139 PNETs (including 83 in the training set and 56 in the independent validation set) were included in the present study. We build a model based on an eight-feature radiomic signature (group 1) to stratify PNET patients into grades 1 and 2/3 groups with an AUC of 0.911 (95% confidence intervals (CI), 0.908-0.914) and 0.837 (95% CI, 0.827-0.847) in the training and validation cohorts, respectively. The nomogram combining the radiomic signature of plain-phase CT with T stage and dilated main pancreatic duct (MPD)/bile duct (BD) (group 2) showed the best performance (training set: AUC = 0.919, 95% CI = 0.916-0.922; validation set: AUC = 0.875, 95% CI = 0.867-0.883). Conclusions: Our developed nomogram that integrates radiomic signature with clinical characteristics could be useful in predicting grades 1 and 2/3 PNETs preoperatively with powerful capability.

Preparation of PEG/ZIF-8@HF drug delivery system for melanoma treatment via oral administration.

Melanoma is one of the highly malignant tumors whose incidence and fatality rates have been increased year by year. However, in addition to early surgical resection, there still lacks specific targeted drugs and treatment strategies. In this study, it was discovered that hinokiflavone (HF) encapsulated in zeolitic imidazolate framework-8 (ZIF-8) exhibited a superior anti-melanoma effect in vitro and in vivo. HF was encapsulated in ZIF-8 through a one-step synthesis method, and polyethylene glycol (PEG-2000) was used to optimize the size and dispersion of the drug-loaded complex (PEG/ZIF-8@HF). The results show that the prepared PEG/ZIF-8@HF has a high encapsulation efficiency (92.12%) and can achieve selective drug release in an acidic microenvironment. The results of in vitro anti-melanoma experiments indicate that PEG/ZIF-8@HF shows up-regulation of reactive oxygen species (ROS) levels and can restrain the migration and invasion of B16F10 cells. Moreover, in vivo animal experiments further confirm that PEG/ZIF-8@HF shows anti-tumor effect by up-regulating the pro-apoptotic proteins caspase-3 and caspase-8, and down-regulating the migration-promoting invasion protein MMP-9. This study developed a safe and effective oral administration of HF based on the high-efficiency delivery ZIF-8 system, which provides an effective treatment strategy for melanoma.

RNA editing regulates lncRNA splicing in human early embryo development.

RNA editing is a co- or post-transcriptional modification through which some cells can make discrete changes to specific nucleotide sequences within an RNA molecule after transcription. Previous studies found that RNA editing may be critically involved in cancer and aging. However, the function of RNA editing in human early embryo development is still unclear. In this study, through analyzing single cell RNA sequencing data, 36.7% RNA editing sites were found to have a have differential editing ratio among early embryo developmental stages, and there was a great reprogramming of RNA editing rates at the 8-cell stage, at which most of the differentially edited RNA editing sites (99.2%) had a decreased RNA editing rate. In addition, RNA editing was more likely to occur on RNA splicing sites during human early embryo development. Furthermore, long non-coding RNA (lncRNA) editing sites were found more likely to be on RNA splicing sites (odds ratio = 2.19, P = 1.37×10-8), while mRNA editing sites were less likely (odds ratio = 0.22, P = 8.38×10-46). Besides, we found that the RNA editing rate on lncRNA had a significantly higher correlation coefficient with the percentage spliced index (PSI) of lncRNA exons (R = 0.75, P = 4.90×10-16), which indicated that RNA editing may regulate lncRNA splicing during human early embryo development. Finally, functional analysis revealed that those RNA editing-regulated lncRNAs were enriched in signal transduction, the regulation of transcript expression, and the transmembrane transport of mitochondrial calcium ion. Overall, our study might provide a new insight into the mechanism of RNA editing on lncRNAs in human developmental biology and common birth defects.

Network-based protein-protein interaction prediction method maps perturbations of cancer interactome.

The perturbations of protein-protein interactions (PPIs) were found to be the main cause of cancer. Previous PPI prediction methods which were trained with non-disease general PPI data were not compatible to map the PPI network in cancer. Therefore, we established a novel cancer specific PPI prediction method dubbed NECARE, which was based on relational graph convolutional network (R-GCN) with knowledge-based features. It achieved the best performance with a Matthews correlation coefficient (MCC) = 0.84±0.03 and an F1 = 91±2% compared with other methods. With NECARE, we mapped the cancer interactome atlas and revealed that the perturbations of PPIs were enriched on 1362 genes, which were named cancer hub genes. Those genes were found to over-represent with mutations occurring at protein-macromolecules binding interfaces. Furthermore, over 56% of cancer treatment-related genes belonged to hub genes and they were significantly related to the prognosis of 32 types of cancers. Finally, by coimmunoprecipitation, we confirmed that the NECARE prediction method was highly reliable with a 90% accuracy. Overall, we provided the novel network-based cancer protein-protein interaction prediction method and mapped the perturbation of cancer interactome. NECARE is available at: https://github.com/JiajunQiu/NECARE.

Advances in Drug Resistance of Esophageal Cancer: From the Perspective of Tumor Microenvironment.

Drug resistance represents the major obstacle to get the maximum therapeutic benefit for patients with esophageal cancer since numerous patients are inherently or adaptively resistant to therapeutic agents. Notably, increasing evidence has demonstrated that drug resistance is closely related to the crosstalk between tumor cells and the tumor microenvironment (TME). TME is a dynamic and ever-changing complex biological network whose diverse cellular and non-cellular components influence hallmarks and fates of tumor cells from the outside, and this is responsible for the development of resistance to conventional therapeutic agents to some extent. Indeed, the formation of drug resistance in esophageal cancer should be considered as a multifactorial process involving not only cancer cells themselves but cancer stem cells, tumor-associated stromal cells, hypoxia, soluble factors, extracellular vesicles, etc. Accordingly, combination therapy targeting tumor cells and tumor-favorable microenvironment represents a promising strategy to address drug resistance and get better therapeutic responses for patients with esophageal cancer. In this review, we mainly focus our discussion on molecular mechanisms that underlie the role of TME in drug resistance in esophageal cancer. We also discuss the opportunities and challenges for therapeutically targeting tumor-favorable microenvironment, such as membrane proteins, pivotal signaling pathways, and cytokines, to attenuate drug resistance in esophageal cancer.

A Novel Multiresolution-Statistical Texture Analysis Architecture: Radiomics-Aided Diagnosis of PDAC Based on Plain CT Images.

Early screening of PDAC (pancreatic ductal adenocarcinoma) based on plain CT (computed tomography) images is of great significance. Therefore, this work conducted a radiomics-aided diagnosis analysis of PDAC based on plain CT images. We explored a novel MSTA (multiresolution-statistical texture analysis) architecture to extract texture features and built machine learning models to classify PDACs and HPs (healthy pancreases). We also performed significance tests of differences to analyze the relationships between histopathological characteristics and texture features. The MSTA architecture originates from the analysis of histopathological characteristics and combines multiresolution analysis and statistical analysis to extract texture features. The MSTA architecture achieved better experimental results than the traditional architecture that scales the coefficient matrices of the multiresolution analysis. In the validation of the classifications, the MSTA architecture achieved an accuracy of 81.19% and an AUC (area under the ROC (receiver operating characteristic) curve) of 0.88 (95% confidence interval: 0.84-0.92). In the test of the classifications, it achieved an accuracy of 77.66% and an AUC of 0.79 (95% confidence interval: 0.71-0.87). Moreover, the significance tests of differences showed that the extracted texture features may be relevant to the histopathological characteristics. The MSTA architecture is beneficial for the radiomics-aided diagnosis of PDAC based on plain CT images. Its texture features can potentially enhance radiologists' imaging interpretation abilities.

Overexpression of Fatty Acid 2-Hydroxylase is Associated with an Increased Sensitivity to Cisplatin by Ovarian Cancer and Better Prognoses.

Background: Recent discoveries indicate that the enzyme fatty acid 2-hydroxylase (FA2H) is associated with biological behavior and can be used for outcome prediction in several types of cancers. Such relevancy, however, between FA2H and ovarian cancer is not clear. Therefore, we carried out this study to compare the expression of FA2H with the clinicopathological features of ovarian cancer. Methods: Using the Oncomine database, we examined the expression levels of the FA2H gene in ovarian cancer tissues and their adjacent noncancerous tissues that had been evaluated by quantitative reverse-transcription polymerase chain reaction (PCR) analyses. We performed Kaplan-Meier curve analyses for overall survival and progression-free survival. In addition, relationships between the FA2H expression levels and clinicopathological features of ovarian cancer were analyzed. Finally, FA2H small interfering RNAs (siRNAs) or negative control siRNAs were separately transfected into OVCAR-3 and SKOV-3 cells to explore the downstream effects. From these results, Gli1 expression was tested by real-time PCR, and the effects of FA2H expression levels on the sensitivity of ovarian cancer cells to cisplatin chemotherapy was evaluated using sulforhodamine B assays. Results: Compared with the adjacent tissues, FA2H was expressed at lower levels in the ovarian cancer tissues. In survival analyses, decreased FA2H was significantly associated with poorer survival outcome in multiple subtypes of ovarian cancer. In addition, FA2H expression was significantly associated with Fédération Internationale de Gynécologie et d'Obstétrique (FIGO) stage, differentiation, lymph node involvement, tumor size, ascites, CA125 levels, and pelvic involvement. Knockdown of FA2H expression by siRNAs in the OVCAR-3 and SKOV-3 cell lines reduced their sensitivity to cisplatin, via modulation of GLI Family Zinc Finger 1 (Gli1) gene expression. Conclusion: Our results demonstrate that FA2H is a biomarker for ovarian cancer and it may serve as a useful prognostic factor.

Gold/SH-functionalized nanographene oxide/polyamidamine/poly(ethylene glycol) nanocomposites for enhanced non-enzymatic hydrogen peroxide detection.

Hydrogen peroxide (H2O2) is an important mediator in biological medicine, disease diagnosis and environmental analyses and therefore it is essential to develop a detection approach for H2O2 in physical environments. Herein, we designed and prepared a series of AuNP-containing nanocomposites (AuNPs@NGO-PEG, AuNPs@G1-PAMAM-NGO-PEG and AuNPs@G3-PAMAM-NGO-PEG) for enhanced non-enzymatic H2O2 detection. We firstly demonstrated functionalized nanographene oxide (NGO) based materials, which combined advantages of biocompatible poly(ethylene glycol) (PEG), hyperbranched polyamidamine (PAMAM) dendrimer and thiol active site, as compatible platforms. Gold nanoparticles (AuNPs) were then aptly in situ grown on these functionalized NGO based materials via the reduction of HAuCl4 under mild conditions, i.e. AuNPs@NGO-PEG, AuNPs@G1-PAMAM-NGO-PEG and AuNPs@G3-PAMAM-NGO-PEG nanocomposites, which possess stable and uniform AuNPs standing on the functionalized NGO sheets. For H2O2 detection, these nanocomposites were cast on a glassy carbon electrode (GCE) conveniently, i.e. GCE/AuNPs@NGO-PEG, GCE/AuNPs@G1-PAMAM-NGO-PEG and GCE/AuNPs@G3-PAMAM-NGO-PEG. It is evident that these GCEs could be applied as efficient non-enzymatic H2O2 detectors resulting from the corresponding cyclic voltammetric curves and typical ready-state amperometric curves. GCE/AuNPs@G1-PAMAM-NGO-PEG exhibited the fastest electron transfer rate among these modified GCEs. We envisage that these GCEs could provide efficient sensors for H2O2 detection and a new strategy for sensor design.

A radiomics signature to identify malignant and benign liver tumors on plain CT images.

BACKGROUND: In regular examinations, it may be difficult to visually identify benign and malignant liver tumors based on plain computed tomography (CT) images. RCAD (radiomics-based computer-aided diagnosis) has proven to be helpful and provide interpretability in clinical use. OBJECTIVE: This work aims to develop a CT-based radiomics signature and investigate its correlation with malignant/benign liver tumors. METHODS: We retrospectively analyzed 168 patients of hepatocellular carcinoma (malignant) and 117 patients of hepatic hemangioma (benign). Texture features were extracted from plain CT images and used as candidate features. A radiomics signature was developed from the candidate features. We performed logistic regression analysis and used a multiple-regression coefficient (termed as R) to assess the correlation between the developed radiomics signature and malignant/benign liver tumors. Finally, we built a logistic regression model to classify benign and malignant liver tumors. RESULTS: Thirteen features were chosen from 1223 candidate features to constitute the radiomics signature. The logistic regression analysis achieved an R = 0.6745, which was much larger than Rα = 0.3703 (the critical value of R at significant level α = 0.001). The logistic regression model achieved an average AUC of 0.87. CONCLUSIONS: The developed radiomics signature was statistically significantly correlated with malignant/benign liver tumors (p < 0.001). It has potential to help enhance physicians' diagnostic abilities and play an important role in RCADs.