18F-FDG PET/CT Radiomics-Based Multimodality Fusion Model for Preoperative Individualized Noninvasive Prediction of Peritoneal Metastasis in Advanced Gastric Cancer.

PURPOSE: This study was designed to develop and validate a machine learning-based, multimodality fusion (MMF) model using 18F-fluorodeoxyglucose (FDG) PET/CT radiomics and kernelled support tensor machine (KSTM), integrated with clinical factors and nuclear medicine experts' diagnoses to individually predict peritoneal metastasis (PM) in advanced gastric cancer (AGC). METHODS: A total of 167 patients receiving preoperative PET/CT and subsequent surgery were included between November 2006 and September 2020 and were divided into a training and testing cohort. The PM status was confirmed via laparoscopic exploration and postoperative pathology. The PET/CT signatures were constructed by classic radiomic, handcrafted-feature-based model and KSTM self-learning-based model. The clinical nomogram was constructed by independent risk factors for PM. Lastly, the PET/CT signatures, clinical nomogram, and experts' diagnoses were fused using evidential reasoning to establish the MMF model. RESULTS: The MMF model showed excellent performance in both cohorts (area under the curve [AUC] 94.16% and 90.84% in training and testing), and demonstrated better prediction accuracy than clinical nomogram or experts' diagnoses (net reclassification improvement p < 0.05). The MMF model also had satisfactory generalization ability, even in mucinous adenocarcinoma and signet ring cell carcinoma which have poor uptake of 18F-FDG (AUC 97.98% and 89.71% in training and testing). CONCLUSIONS: The 18F-FDG PET/CT radiomics-based MMF model may have significant clinical implications in predicting PM in AGC, revealing that it is necessary to combine the information from different modalities for comprehensive prediction of PM.

Histone demethylase KDM4A mediating macrophage polarization: A potential mechanism of trichloroethylene induced liver injury.

Trichloroethylene (TCE) is a commonly used organic solvent in industry. Our previous studies have found that TCE can cause liver injury accompanied by macrophage polarization, but the specific mechanism is unclear. The epigenetic regulation of macrophage polarization is mainly focused on histone modification. Histone lysine demethylase 4A (KDM4A) is involved in the activation of macrophages. In this study, we used a mouse model we investigated the role of KDM4A in the livers of TCE-drinking mice and found that the expression of KDM4A, M1-type polarization indicators, and related inflammatory factors in the livers of TCE-drinking mice. In the study, BALB/c mice were randomly divided into four groups: 2.5 mg/mL TCE dose group and 5.0 mg/mL TCE dose group, the vehicle control group, and the blank control group. We found that TCE triggered M1 polarization of mouse macrophages, characterized by the expression of CD11c and robust production of inflammatory cytokines. Notably, exposure to TCE resulted in markedly increased expression of KDM4A in macrophages. Functionally, the increased expression of KDM4A significantly impaired the expression of H3K9me3 and H3K9me2 and increased the expression of H3K9me1. In addition, KDM4A potentially represents a novel epigenetic modulator, with its upregulation connected to ß-catenin activation, a signal critical for the pro-inflammatory activation of macrophages. Furthermore, KDM4A inhibitor JIB-04 treatment resulted in a decrease in ß-catenin expression and prevented TCE-induced M1 polarization and the expression of the pro-inflammatory cytokines TNF-α and IL-1ß. These results suggest that the association of KDM4A and Wnt/ß-catenin cooperatively establishes the activation and polarization of macrophages and global changes in H3K9me3/me2/me1. Our findings identify KDM4A as an essential regulator of the polarization of macrophages and the expression of inflammatory cytokines, which might serve as a potential target for preventing and treating liver injury caused by TCE.

Multiscale and multiperception feature learning for pancreatic lesion detection based on noncontrast CT.

Background. Pancreatic cancer is one of the most malignant tumours, demonstrating a poor prognosis and nearly identically high mortality and morbidity, mainly because of the difficulty of early diagnosis and timely treatment for localized stages.Objective. To develop a noncontrast CT (NCCT)-based pancreatic lesion detection model that could serve as an intelligent tool for diagnosing pancreatic cancer early, overcoming the challenges associated with low contrast intensities and complex anatomical structures present in NCCT images.Approach.We design a multiscale and multiperception (MSMP) feature learning network with ResNet50 coupled with a feature pyramid network as the backbone for strengthening feature expressions. We added multiscale atrous convolutions to expand different receptive fields, contextual attention to perceive contextual information, and channel and spatial attention to focus on important channels and spatial regions, respectively. The MSMP network then acts as a feature extractor for proposing an NCCT-based pancreatic lesion detection model with image patches covering the pancreas as its input; Faster R-CNN is employed as the detection method for accurately detecting pancreatic lesions.Main results. By using the new MSMP network as a feature extractor, our model outperforms the conventional object detection algorithms in terms of the recall (75.40% and 90.95%), precision (40.84% and 68.21%), F1 score (52.98% and 77.96%), F2 score (64.48% and 85.26%) and Ap50 metrics (53.53% and 70.14%) at the image and patient levels, respectively.Significance.The good performance of our new model implies that MSMP can mine NCCT imaging features for detecting pancreatic lesions from complex backgrounds well. The proposed detection model is expected to be further developed as an intelligent method for the early detection of pancreatic cancer.

Ultra-thin nanosheets decorated in-situ S-doped 3D interconnected carbon network as interlayer modified Li-S batteries separator for accelerating adsorption-catalytic synergistic process of LiPSs.

The shuttle effect of soluble lithium polysulfides (LiPSs) is primarily responsible for the unstable performance of lithium-sulfur (Li-S) batteries, which has severely impeded their continued development. In order to solve this problem, a special strategy is proposed. Specifically, ultra-thin NiCo based layered double hydroxides (named LDH or NiCo-LDH) nanosheets are implanted into a pre-designed 3D interconnected carbon networks (SPC) to obtain porous composite materials (named SPC-LDH).During the operation of the battery, the 3D interconnected porous carbon mesh was the first to rapidly adsorb LiPSs, and then the LDH on the surface of the carbon mesh was used to realize the catalytic conversion of LiPSs. This facilitates the electrochemical conversion reaction between S substances while addressing the "shuttle effect". As a result, the battery maintains a discharge capacity of 1401.9, 1114.3, 975.5, 880.7, 760.4 and 679.6 mAh g-1 at the current densities of 0.1, 0.2, 0.5, 1, 2 and 3C, respectively. After 200 cycles at 2C, the battery's capacity stays at 732.9 mAh g-1, meaning that the average rate of capacity decay is only 0.007 % per cycle. Moreover, in-situ XRD demonstrates the critical function of PP/SPC-LDH separators in inhibiting LiPSs and encouraging Li2S transformation. The strong affinity of SPC-LDH for Li2S6 is also confirmed by density functional theory (DFT) calculation, offering more theoretical support for the synergistic adsorption process. This work offers a compelling method to develop modified separator materials that can counteract the "shuttle effect" in Li-S batteries.

Computed tomography and guidelines-based human-machine fusion model for predicting resectability of the pancreatic cancer.

BACKGROUND AND AIM: The study aims to develop a hybrid machine learning model for predicting resectability of the pancreatic cancer, which is based on computed tomography (CT) and National Comprehensive Cancer Network (NCCN) guidelines. METHOD: We retrospectively studied 349 patients. One hundred seventy-one cases from Center 1 and 92 cases from Center 2 were used as the primary training cohort, and 66 cases from Center 3 and 20 cases from Center 4 were used as the independent test dataset. Semi-automatic module of ITK-SNAP software was used to assist CT image segmentation to obtain three-dimensional (3D) imaging region of interest (ROI). There were 788 handcrafted features extracted for 3D ROI using PyRadiomics. The optimal feature subset consists of three features screened by three feature selection methods as the input of the SVM to construct the conventional radiomics-based predictive model (cRad). 3D ROI was used to unify the resolution by 3D spline interpolation method for constructing the 3D tumor imaging tensor. Using 3D tumor image tensor as input, 3D kernelled support tensor machine-based predictive model (KSTM), and 3D ResNet-based deep learning predictive model (ResNet) were constructed. Multi-classifier fusion ML model is constructed by fusing cRad, KSTM, and ResNet using multi-classifier fusion strategy. Two experts with more than 10 years of clinical experience were invited to reevaluate each patient based on their CECT following the NCCN guidelines to obtain resectable, unresectable, and borderline resectable diagnoses. The three results were converted into probability values of 0.25, 0.75, and 0.50, respectively, according to the traditional empirical method. Then it is used as an independent classifier and integrated with multi-classifier fusion machine learning (ML) model to obtain the human-machine fusion ML model (HMfML). RESULTS: Multi-classifier fusion ML model's area under receiver operating characteristic curve (AUC; 0.8610), predictive accuracy (ACC: 80.23%), sensitivity (SEN: 78.95%), and specificity (SPE: 80.60%) is better than cRad, KSTM, and ResNet-based single-classifier models and their two-classifier fusion models. This means that three different models have mined complementary CECT feature expression from different perspectives and can be integrated through CFS-ER, so that the fusion model has better performance. HMfML's AUC (0.8845), ACC (82.56%), SEN (84.21%), SPE (82.09%). This means that ML models might learn extra information from CECT that experts cannot distinguish, thus complementing expert experience and improving the performance of hybrid ML models. CONCLUSION: HMfML can predict PC resectability with high accuracy.

Glutamine metabolic microenvironment drives M2 macrophage polarization to mediate trastuzumab resistance in HER2-positive gastric cancer.

BACKGROUND: Trastuzumab is a first-line targeted therapy for human epidermal growth factor receptor-2 (HER2)-positive gastric cancer. However, the inevitable occurrence of acquired trastuzumab resistance limits the drug benefit, and there is currently no effective reversal measure. Existing researches on the mechanism of trastuzumab resistance mainly focused on tumor cells themselves, while the understanding of the mechanisms of environment-mediated drug resistance is relatively lacking. This study aimed to further explore the mechanisms of trastuzumab resistance to identify strategies to promote survival in these patients. METHODS: Trastuzumab-sensitive and trastuzumab-resistant HER2-positive tumor tissues and cells were collected for transcriptome sequencing. Bioinformatics were used to analyze cell subtypes, metabolic pathways, and molecular signaling pathways. Changes in microenvironmental indicators (such as macrophage, angiogenesis, and metabolism) were verified by immunofluorescence (IF) and immunohistochemical (IHC) analyses. Finally, a multi-scale agent-based model (ABM) was constructed. The effects of combination treatment were further validated in nude mice to verify these effects predicted by the ABM. RESULTS: Based on transcriptome sequencing, molecular biology, and in vivo experiments, we found that the level of glutamine metabolism in trastuzumab-resistant HER2-positive cells was increased, and glutaminase 1 (GLS1) was significantly overexpressed. Meanwhile, tumor-derived GLS1 microvesicles drove M2 macrophage polarization. Furthermore, angiogenesis promoted trastuzumab resistance. IHC showed high glutamine metabolism, M2 macrophage polarization, and angiogenesis in trastuzumab-resistant HER2-positive tumor tissues from patients and nude mice. Mechanistically, the cell division cycle 42 (CDC42) promoted GLS1 expression in tumor cells by activating nuclear factor kappa-B (NF-κB) p65 and drove GLS1 microvesicle secretion through IQ motif-containing GTPase-activating protein 1 (IQGAP1). Based on the ABM and in vivo experiments, we confirmed that the combination of anti-glutamine metabolism, anti-angiogenesis, and pro-M1 polarization therapy had the best effect in reversing trastuzumab resistance in HER2-positive gastric cancer. CONCLUSIONS: This study revealed that tumor cells secrete GLS1 microvesicles via CDC42 to promote glutamine metabolism, M2 macrophage polarization, and pro-angiogenic function of macrophages, leading to acquired trastuzumab resistance in HER2-positive gastric cancer. A combination of anti-glutamine metabolism, anti-angiogenesis, and pro-M1 polarization therapy may provide a new insight into reversing trastuzumab resistance.

Enhanced Spatial Charge Separation in a Niobium and Tantalum Nitride Core-Shell Photoanode: In Situ Interface Bonding for Efficient Solar Water Splitting.

Tantalum nitride (Ta3 N5 ) has emerged as a promising photoanode material for photoelectrochemical (PEC) water splitting. However, the inefficient electron-hole separation remains a bottleneck that impedes its solar-to-hydrogen conversion efficiency. Herein, we demonstrate that a core-shell nanoarray photoanode of NbNx -nanorod@Ta3 N5 ultrathin layer enhances light harvesting and forms a spatial charge-transfer channel, which leads to the efficient generation and extraction of charge carriers. Consequently, an impressive photocurrent density of 7 mA cm-2 at 1.23 VRHE is obtained with an ultrathin Ta3 N5 shell thickness of less than 30 nm, accompanied by excellent stability and a low onset potential (0.46 VRHE ). Mechanistic studies reveal the enhanced performance is attributed to the high-conductivity NbNx core, high-crystalline Ta3 N5 mono-grain shell, and the intimate Ta-N-Nb interface bonds, which accelerate the charge-separation capability of the core-shell photoanode. This study demonstrates the key roles of nanostructure design in improving the efficiency of PEC devices.

Regulating the thickness of the carbon coating layer in iron/carbon heterostructures to enhance the catalytic performance for oxygen evolution reaction.

The exploration of high-performance electrocatalysts for the oxygen evolution reaction (OER) is crucial and urgent for the fast development of green and renewable hydrogen energy. Herein, an ultra-fast and energy-efficient preparation strategy (microwave-assisted rapid in-situ pyrolysis of organometallic compounds induced by carbon nanotube (CNT)) is developed to obtain iron/carbon (Fe/C) heterogeneous materials (Fe/Fe3C particles wrapped by carbon coating layer). The thickness of the carbon coating layer can be adjusted by changing the content and form of carbon in the metal sources during the fast preparation process. Fe/Fe3C-A@CNT using iron acetylacetonate as metal sources possesses unique Fe/C heterogeneous, small Fe/Fe3C particles encapsulated by the thin carbon coating layer (1.77 nm), and obtains the optimal electron penetration effect. The electron penetration effect derives from the redistribution of charge between the surface carbon coating layer and inner Fe/Fe3C nanoparticles efficiently improving both catalytic activity and stability. Therefore, Fe/Fe3C-A@CNT shows efficient OER catalytic activity, just needing a low overpotential of 292 mV to reach a current density of 10 mA cm-2, and long-lasting stability. More importantly, the unique control strategy for carbon thickness in this work provides more opportunity and perspective to prepare robust metal/carbon-based catalytic materials at the nanoscale.

A multimodal model fusing multiphase contrast-enhanced CT and clinical characteristics for predicting lymph node metastases of pancreatic cancer.

Objective.To develop a multimodal model that combines multiphase contrast-enhanced computed tomography (CECT) imaging and clinical characteristics, including experts' experience, to preoperatively predict lymph node metastasis (LNM) in pancreatic cancer patients.Methods.We proposed a new classifier fusion strategy (CFS) based on a new evidential reasoning (ER) rule (CFS-nER) by combining nomogram weights into a previous ER rule-based CFS. Three kernelled support tensor machine-based classifiers with plain, arterial, and venous phases of CECT as the inputs, respectively, were constructed. They were then fused based on the CFS-nER to construct a fusion model of multiphase CECT. The clinical characteristics were analyzed by univariate and multivariable logistic regression to screen risk factors, which were used to construct correspondent risk factor-based classifiers. Finally, the fusion model of the three phases of CECT and each risk factor-based classifier were fused further to construct the multimodal model based on our CFS-nER, named MMM-nER. This study consisted of 186 patients diagnosed with pancreatic cancer from four clinical centers in China, 88 (47.31%) of whom had LNM.Results.The fusion model of the three phases of CECT performed better overall than single and two-phase fusion models; this implies that the three considered phases of CECT were supplementary and complemented one another. The MMM-nER further improved the predictive performance, which implies that our MMM-nER can complement the supplementary information between CECT and clinical characteristics. The MMM-nER had better predictive performance than based on previous classifier fusion strategies, which presents the advantage of our CFS-nER.Conclusion.We proposed a new CFS-nER, based on which the fusion model of the three phases of CECT and MMM-nER were constructed and performed better than all compared methods. MMM-nER achieved an encouraging performance, implying that it can assist clinicians in noninvasively and preoperatively evaluating the lymph node status of pancreatic cancer.

Prenatal low-dose antibiotic exposure and children allergic diseases at 4 years of age: A prospective birth cohort study.

BACKGROUND: Based on a medical record or questionnaire survey approach, previous epidemiological studies have investigated associations between maternal antibiotic exposure during pregnancy and childhood allergic diseases. However, biomonitoring studies on the prenatal low-dose antibiotic exposure, mainly from the environment and contaminated food, and in relation to children allergic diseases, are missing. OBJECTIVES: This research aimed to examine the associations between prenatal low-dose antibiotic exposure measured at multiple time points and children current allergic diseases at 4 years of age. METHODS: The current study including 2453 mother-child pairs was based on the Ma'anshan Birth Cohort study. Selected 41 antibiotics and their two metabolites, which including human antibiotics (HAs), preferred as human antibiotics (PHAs), veterinary antibiotics (VAs) and preferred as veterinary antibiotics (PVAs), in urine samples from 2453 pregnant women were biomonitored through liquid chromatography-triple quadrupole tandem mass spectrometry. Information on children current allergic diseases were collected via validated questionnaires. Generalized estimating equation were used to explore the associations between the repeated measurements of maternal urinary antibiotic over three trimesters and current allergic diseases in children. RESULTS: The detection rates of nine individual antibiotics in the three trimester during pregnancy are greater than 10%, and the 90th percentile concentration of the detected antibiotics ranges from 0.07 to 22.34 µg/g, and the 95th percentile concentration ranges from 0.17 to 59.57 µg/g. Among the participants, each one-unit concentration increment of sulfamethazine (adjusted OR=1.28, 95% CI: 1.10, 1.49, P-FDR=0.014) in the first trimester and ciprofloxacin (adjusted OR=1.17, 95% CI: 1.07, 1.28, P-FDR=0.008) in the second trimester were associated with an increased risk of current eczema in children. In the third trimester, each one-unit concentration increment of oxytetracycline (adjusted OR=1.90, 95% CI: 1.30, 2.78, P-FDR=0.014) was associated with an increased risk of current asthma in children. Gender-stratified analyses demonstrated that no gender differences were observed in the associations between prenatal antibiotic exposure and current allergic diseases in children. CONCLUSIONS: Maternal exposure to certain specific VAs or PVAs (sulfamethazine, ciprofloxacin and oxytetracycline) in different trimesters was associated with an increased risk of current asthma and current eczema in 4-year-old children. No gender differences were found in these associations. Further studies are warranted to confirm our findings and explore the potential mechanisms.

On the robustness of deep learning-based lung-nodule classification for CT images with respect to image noise.

Robustness is an important aspect when evaluating a method of medical image analysis. In this study, we investigated the robustness of a deep learning (DL)-based lung-nodule classification model for CT images with respect to noise perturbations. A deep neural network (DNN) was established to classify 3D CT images of lung nodules into malignant or benign groups. The established DNN was able to predict malignancy rate of lung nodules based on CT images, achieving the area under the curve of 0.91 for the testing dataset in a tenfold cross validation as compared to radiologists' prediction. We then evaluated its robustness against noise perturbations. We added to the input CT images noise signals generated randomly or via an optimization scheme using a realistic noise model based on a noise power spectrum for a given mAs level, and monitored the DNN's output. The results showed that the CT noise was able to affect the prediction results of the established DNN model. With random noise perturbations at 100 mAs, DNN's predictions for 11.2% of training data and 17.4% of testing data were successfully altered by at least once. The percentage increased to 23.4% and 34.3%, respectively, for optimization-based perturbations. We further evaluated robustness of models with different architectures, parameters, number of output labels, etc, and robustness concern was found in these models to different degrees. To improve model robustness, we empirically proposed an adaptive training scheme. It fine-tuned the DNN model by including perturbations in the training dataset that successfully altered the DNN's perturbations. The adaptive scheme was repeatedly performed to gradually improve DNN's robustness. The numbers of perturbations at 100 mAs affecting DNN's predictions were reduced to 10.8% for training and 21.1% for testing by the adaptive training scheme after two iterations. Our study illustrated that robustness may potentially be a concern for an exemplary DL-based lung-nodule classification model for CT images, indicating the needs for evaluating and ensuring model robustness when developing similar models. The proposed adaptive training scheme may be able to improve model robustness.

Efficacy of Qishen Yiqi Drop Pill for Chronic Heart Failure: An Updated Meta-Analysis of 85 Studies.

BACKGROUND: Despite evidence for beneficial effects of Qishen Yiqi Drop Pill (QSYQ) on congestive heart failure, the majority of studies are based on insufficient sample sizes. The aim of this study was to evaluate the therapeutic effects of QSYQ using a meta-analysis approach. Methodology/Principal Findings. All relevant studies published before December 31, 2019, were identified by searches of various databases with key search terms. In total, 85 studies involving 8,579 participants were included. The addition of QSYQ to routine Western medicine increased 6-minute walking distance (SMD = 2.08, 95% CI: 1.72-2.44, p < 0.001), left ventricular ejection fraction (SMD = 1.05, 95% CI: 0.87-1.23, p < 0.001), and cardiac index (SMD = 1.44, 95% CI: 0.92-1.95, p < 0.001) and reduced brain natriuretic peptide (SMD = -2.28, 95% CI: -2.81 to -1.76, p < 0.001), N-terminal prohormone of brain natriuretic peptide (SMD = -2.49, 95% CI: -3.24 to -1.73, p < 0.001), left ventricular end-diastolic dimensions (SMD = -0.92, 95% CI: -1.25 to -0.59, p < 0.001), and left ventricular end-systolic dimensions (SMD = -0.55, 95% CI: -0.89 to -0.21, p < 0.001). The results were stable in subgroup analyses and sensitivity analyses. CONCLUSIONS: Our current meta-analysis indicated that QSYQ combined with Western therapy might be effective in CHF patients. Further researches are needed to identify which subgroups of CHF patients will benefit most and what kind of combination medicines work best.

Multi-Objective-Based Radiomic Feature Selection for Lesion Malignancy Classification.

OBJECTIVE: accurately classifying the malignancy of lesions detected in a screening scan is critical for reducing false positives. Radiomics holds great potential to differentiate malignant from benign tumors by extracting and analyzing a large number of quantitative image features. Since not all radiomic features contribute to an effective classifying model, selecting an optimal feature subset is critical. METHODS: this work proposes a new multi-objective based feature selection (MO-FS) algorithm that considers sensitivity and specificity simultaneously as the objective functions during feature selection. For MO-FS, we developed a modified entropy-based termination criterion that stops the algorithm automatically rather than relying on a preset number of generations. We also designed a solution selection methodology for multi-objective learning that uses the evidential reasoning approach (SMOLER) to automatically select the optimal solution from the Pareto-optimal set. Furthermore, we developed an adaptive mutation operation to generate the mutation probability in MO-FS automatically. RESULTS: we evaluated the MO-FS for classifying lung nodule malignancy in low-dose CT and breast lesion malignancy in digital breast tomosynthesis. CONCLUSION: the experimental results demonstrated that the feature set selected by MO-FS achieved better classification performance than features selected by other commonly used methods. SIGNIFICANCE: the proposed method is general and more effective radiomic feature selection strategy.

A manifold learning regularization approach to enhance 3D CT image-based lung nodule classification.

PURPOSE: Diagnosis of lung cancer requires radiologists to review every lung nodule in CT images. Such a process can be very time-consuming, and the accuracy is affected by many factors, such as experience of radiologists and available diagnosis time. To address this problem, we proposed to develop a deep learning-based system to automatically classify benign and malignant lung nodules. METHODS: The proposed method automatically determines benignity or malignancy given the 3D CT image patch of a lung nodule to assist diagnosis process. Motivated by the fact that real structure among data is often embedded on a low-dimensional manifold, we developed a novel manifold regularized classification deep neural network (MRC-DNN) to perform classification directly based on the manifold representation of lung nodule images. The concise manifold representation revealing important data structure is expected to benefit the classification, while the manifold regularization enforces strong, but natural constraints on network training, preventing over-fitting. RESULTS: The proposed method achieves accurate manifold learning with reconstruction error of ~ 30 HU on real lung nodule CT image data. In addition, the classification accuracy on testing data is 0.90 with sensitivity of 0.81 and specificity of 0.95, which outperforms state-of-the-art deep learning methods. CONCLUSION: The proposed MRC-DNN facilitates an accurate manifold learning approach for lung nodule classification based on 3D CT images. More importantly, MRC-DNN suggests a new and effective idea of enforcing regularization for network training, possessing the potential impact to a board range of applications.

A collection input based support tensor machine for lesion malignancy classification in digital breast tomosynthesis.

Digital breast tomosynthesis (DBT) with improved lesion conspicuity and characterization has been adopted in screening practice. DBT-based diagnosis strongly depends on physicians' experience, so an automatic lesion malignancy classification model using DBT could improve the consistency of diagnosis among different physicians. Tensor-based approaches that use the original imaging data as input have shown promising results for many classification tasks. However, DBT data are pseudo-3D volumetric imaging as the slice spacing of DBT is much coarser than that of the in-plane resolution. Thus, directly constructing DBT as the third-order tensor in a conventional tensor-based classifier with introducing additional information to the original DBT data along the slice-spacing dimension will lead to inconsistency across all three dimensions. To avoid such inconsistency, we introduce a collection input based support tensor machine (CISTM)-based classifier that uses the tensor collection as input for classifying lesion malignancy in DBT. In CISTM, instead of introducing the third dimension directly into the geometry construction, the third-dimension structural relationship is related by weight parameters in the decision function, which is dynamically and automatically constructed during the classifier training process and is more consistent with the pseudo-3D nature of DBT. We tested our method on a DBT dataset of 926 images among which 262 were malignant and 664 were benign. We compared our method with the latest tensor-based method, KSTM (kernelled support tensor machine), which does not consider the unique non-uniform resolution property of DBT. Experimental results illustrate that the CISTM-based classifier is effective for classifying breast lesion malignancy in DBT and that it outperforms the KSTM-based classifier.

Predicting lung nodule malignancies by combining deep convolutional neural network and handcrafted features.

To predict lung nodule malignancy with a high sensitivity and specificity for low dose CT (LDCT) lung cancer screening, we propose a fusion algorithm that combines handcrafted features (HF) into the features learned at the output layer of a 3D deep convolutional neural network (CNN). First, we extracted twenty-nine HF, including nine intensity features, eight geometric features, and twelve texture features based on grey-level co-occurrence matrix (GLCM). We then trained 3D CNNs modified from three 2D CNN architectures (AlexNet, VGG-16 Net and Multi-crop Net) to extract the CNN features learned at the output layer. For each 3D CNN, the CNN features combined with the 29 HF were used as the input for the support vector machine (SVM) coupled with the sequential forward feature selection (SFS) method to select the optimal feature subset and construct the classifiers. The fusion algorithm takes full advantage of the HF and the highest level CNN features learned at the output layer. It can overcome the disadvantage of the HF that may not fully reflect the unique characteristics of a particular lesion by combining the intrinsic CNN features. Meanwhile, it also alleviates the requirement of a large scale annotated dataset for the CNNs based on the complementary of HF. The patient cohort includes 431 malignant nodules and 795 benign nodules extracted from the LIDC/IDRI database. For each investigated CNN architecture, the proposed fusion algorithm achieved the highest AUC, accuracy, sensitivity, and specificity scores among all competitive classification models.

[A radiomic approach to differential diagnosis of renal cell carcinoma in patients with hydronephrosis and renal calculi].

ObjectiveTo explore the application of radiomic analysis in differential diagnosis of renal cell carcinoma in patients with hydronephrosis and renal calculi using supervised machine learning methods.MethodThe abdominal CT scan data were retrospectively analyzed for 66 patients with pathologically confirmed hydronephrosis and renal calculi, among whom 35 patients had renal cell carcinoma. In each case 18 non-texture features and 344 texture features were extracted from the region of interest (ROI). Infinite feature selection (InfFS)-based forward feature selection method coupled with support vector machine (SVM) classifier was used to select the optimal feature subset. SVM was trained and performed the prediction using the selected feature subset to classify whether hydronephrosis with renal calculi was associated with renal cell carcinoma.ResultsA total of 12 texture features were selected as the optimal features. The area under curve (AUC), accuracy, sensitivity, specificity, false positive rate and false negative rate of the SVM- InfFS model for predicting accompanying renal tumors in patients with hydronephrosis and calculi were 0.907, 81.0%, 70.0%, 90.9%, 9.1%, and 30.0%, respectively. The diagnostic accuracy, sensitivity, specificity, false positive and false negative rates by the clinicians provided with these classification results were 90.5%, 80.0%, 100%, 0.00%, and 20.0%, respectively.ConclusionThe computer-aided classification model based on supervised machine learning can effectively extract the diagnostic information and improve the diagnostic rate of renal cell carcinoma associated with hydronephrosis and renal calculi.

A pilot study using kernelled support tensor machine for distant failure prediction in lung SBRT.

We developed a kernelled support tensor machine (KSTM)-based model with tumor tensors derived from pre-treatment PET and CT imaging as input to predict distant failure in early stage non-small cell lung cancer (NSCLC) treated with stereotactic body radiation therapy (SBRT). The patient cohort included 110 early stage NSCLC patients treated with SBRT, 25 of whom experienced failure at distant sites. Three-dimensional tumor tensors were constructed and used as input for the KSTM-based classifier. A KSTM iterative algorithm with a convergent proof was developed to train the weight vectors for every mode of the tensor for the classifier. In contrast to conventional radiomics approaches that rely on handcrafted imaging features, the KSTM-based classifier uses 3D imaging as input, taking full advantage of the imaging information. The KSTM-based classifier preserves the intrinsic 3D geometry structure of the medical images and the correlation in the original images and trains the classification hyper-plane in an adaptive feature tensor space. The KSTM-based predictive algorithm was compared with three conventional machine learning models and three radiomics approaches. For PET and CT, the KSTM-based predictive method achieved the highest prediction results among the seven methods investigated in this study based on 10-fold cross validation and independent testing.

Shell feature: a new radiomics descriptor for predicting distant failure after radiotherapy in non-small cell lung cancer and cervix cancer.

Distant failure is the main cause of human cancer-related mortalities. To develop a model for predicting distant failure in non-small cell lung cancer (NSCLC) and cervix cancer (CC) patients, a shell feature, consisting of outer voxels around the tumor boundary, was constructed using pre-treatment positron emission tomography (PET) images from 48 NSCLC patients received stereotactic body radiation therapy and 52 CC patients underwent external beam radiation therapy and concurrent chemotherapy followed with high-dose-rate intracavitary brachytherapy. The hypothesis behind this feature is that non-invasive and invasive tumors may have different morphologic patterns in the tumor periphery, in turn reflecting the differences in radiological presentations in the PET images. The utility of the shell was evaluated by the support vector machine classifier in comparison with intensity, geometry, gray level co-occurrence matrix-based texture, neighborhood gray tone difference matrix-based texture, and a combination of these four features. The results were assessed in terms of accuracy, sensitivity, specificity, and AUC. Collectively, the shell feature showed better predictive performance than all the other features for distant failure prediction in both NSCLC and CC cohorts.

Trichloroethylene Alters Th1/Th2/Th17/Treg Paradigm in Mice: A Novel Mechanism for Chemically Induced Autoimmunity.

The role of environmental factors in autoimmune diseases has been increasingly recognized. While major advance has been made in understanding biological pathogen-induced autoimmune diseases, chemically triggered autoimmunity is poorly understood. Trichloroethylene (TCE), a common environmental pollutant, has recently been shown to induce autoimmunity. This study explored whether TCE could cause imbalance of T helper (Th) cell subsets which would contribute to the pathogenesis of TCE-induced medicamentosa-like dermatitis. BALB/c mice were treated with TCE via drinking water at doses of 2.5 or 5.0 mg/mL for 2, 4, 8, 12, and 16 weeks. Trichloroethylene exposure caused time- and dose-dependent increase in Th1, Th2, and Th17 and decrease in regulatory cell (Treg) in the spleen at 2, 4, 8, 12, and 16 weeks, with greatest changes mainly at 4 weeks. These effects were mirrored by similar changes in the expression of their corresponding cytokines interferon-γ, interleukin 4 (IL-4), IL-17A, and IL-10. Mechanistically, these phenotypic changes were accounted for by alterations to their respective master transcription factors T-box expressed in T cells, GATA-binding protein 3, Retinoic acid-related orphan receptor ct (RORct), and forkhead box P3. Of interest, TCE treatment shifted the ratios of Th1/Th2 and Th17/Treg; specifically, TCE increased Th17/Treg. These findings provide the first evidence that TCE exposure significantly changes the Th1/Th2/Th17/Treg paradigm and their specific cytokines driven by altered master transcription factors. This may promote autoimmune reactions in the pathogenesis of TCE-induced skin sensitization and associated damage to other tissues.