Glioblastomas and brain metastases differentiation following an MRI texture analysis-based radiomics approach

PurposeTo evaluate the potential of 2D texture features extracted from magnetic resonance (MR) images for differentiating brain metastasis (BM) and glioblastomas (GBM) following a radiomics approach.MethodsThis retrospective study included 50 patients with BM and 50 with GBM who underwent T1-weighted MRI between December 2010 and January 2017. Eighty-eight rotation-invariant texture features were computed for each segmented lesion using six texture analysis methods. These features were also extracted from the four images obtained after applying the discrete wavelet transform (88 features × 4 images). Three feature selection methods and five predictive models were evaluated. A 5-fold cross-validation scheme was used to randomly split the study group into training (80 patients) and testing (20 patients), repeating the process ten times. Classification was evaluated computing the average area under the receiver operating characteristic curve. Sensibility, specificity and accuracy were also computed. The whole process was tested quantizing the images with different gray-level values to evaluate their influence in the final results.ResultsHighest classification accuracy was obtained using the original images quantized with 128 gray-levels and a feature selection method based on the p-value. The best overall performance was achieved using a support vector machine model with a subset of 32 features (AUC = 0.896 ± 0.067, sensitivity of 82% and specificity of 80%). Naïve Bayes and k-nearest neighbors models showed also valuable results (AUC ≈ 0.8) with a lower number of features (<13), thus suggesting that these models may be more generalizable when using external validations.ConclusionThe proposed radiomics MRI approach is able to discriminate between GBM and BM with high accuracy employing a set of 2D texture features, thus helping in the diagnosis of brain lesions in a fast and non-invasive way.     

Intravoxel incoherent motion radiomics nomogram for predicting tumor treatment responses in nasopharyngeal carcinoma

BackgroundIntravoxel incoherent motion (IVIM) plays an important role in predicting treatment responses in patient with nasopharyngeal carcinoma (NPC). The goal of this study was to develop and validate a radiomics nomogram based on IVIM parametric maps and clinical data for the prediction of treatment responses in NPC patients.MethodsEighty patients with biopsy-proven NPC were enrolled in this study. Sixty-two patients had complete responses and 18 patients had incomplete responses to treatment. Each patient received a multiple b-value diffusion-weighted imaging (DWI) examination before treatment. Radiomics features were extracted from IVIM parametric maps derived from DWI image. Feature selection was performed by the least absolute shrinkage and selection operator method. Radiomics signature was generated by support vector machine based on the selected features. Receiver operating characteristic (ROC) curves and area under the ROC curve (AUC) values were used to evaluate the diagnostic performance of radiomics signature. A radiomics nomogram was established by integrating the radiomics signature and clinical data.ResultsThe radiomics signature showed good prognostic performance to predict treatment response in both training (AUC = 0.906, P<0.001) and testing (AUC = 0.850, P<0.001) cohorts. The radiomic nomogram established by integrating the radiomic signature with clinical data significantly outperformed clinical data alone (C-index, 0.929 vs 0.724; P<0.0001).ConclusionsThe IVIM-based radiomics nomogram provided high prognostic ability to treatment responses in patients with NPC. The IVIM-based radiomics signature has the potential to be a new biomarker in prediction of the treatment responses and may affect treatment strategies in patients with NPC.     

Automated voxel-by-voxel tissue classification for hippocampal segmentation: Methods and validation

The hippocampus is an important structural biomarker for Alzheimer's disease (AD) and has a primary role in the pathogenesis of other neurological and psychiatric diseases. This study presents a fully automated pattern recognition system for an accurate and reproducible segmentation of the hippocampus in structural Magnetic Resonance Imaging (MRI). The method was validated on a mixed cohort of 56 T1-weighted structural brain images, and consists of three processing levels: (a) Linear registration: all brain images were registered to a standard template and an automated method was applied to capture the global shape of the hippocampus. (b) Feature extraction: all voxels included in the previously selected volume were characterized by 315 features computed from local information. (c) Voxel classification: a Random Forest algorithm was used to classify voxels as belonging or not belonging to the hippocampus. In order to improve the classification performance, an adaptive learning method based on the use of the Pearson's correlation coefficient was developed. The segmentation results (Dice similarity index = 0.81 ± 0.03) compare well with other state-of-the art approaches. A validation study was conducted on an independent dataset of 100 T1-weighted brain images, achieving significantly better results than those obtained with FreeSurfer.     

Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography

ObjectivesThe subtype classification of lung adenocarcinoma is important for treatment decision. This study aimed to investigate the deep learning and radiomics networks for predicting histologic subtype classification and survival of lung adenocarcinoma diagnosed through computed tomography (CT) images.MethodsA dataset of 1222 patients with lung adenocarcinoma were retrospectively enrolled from three medical institutions. The anonymised preoperative CT images and pathological labels of atypical adenomatous hyperplasia, adenocarcinoma in situ, minimally invasive adenocarcinoma, invasive adenocarcinoma (IAC) with five predominant components were obtained. These pathological labels were divided into 2-category classification (IAC; non-IAC), 3-category and 8-category. We modeled the classification task of histological subtypes based on modified ResNet-34 deep learning network, radiomics strategies and deep radiomics combined algorithm. Then we established the prognostic models in lung adenocarcinoma patients with survival outcomes. The accuracy (ACC), area under ROC curves (AUCs) and C-index were primarily performed to evaluate the algorithms.ResultsThis study included a training set (n = 802) and two validation cohorts (internal, n = 196; external, n = 224). The ACC of deep radiomics algorithm in internal validation achieved 0.8776, 0.8061 in the 2-category, 3-category classification, respectively. Even in 8 classifications, the AUC ranged from 0.739 to 0.940 in internal set. Further, we constructed a prognosis model that C-index was 0.892(95% CI: 0.846–0.937) in internal validation set.ConclusionsThe automated deep radiomics based triage system has achieved the great performance in the subtype classification and survival predictability in patients with CT-detected lung adenocarcinoma nodules, providing the clinical guide for treatment strategies.     

Patient-specific IMRT QA verification using machine learning and gamma radiomics

Gamma function is the standard methodology for comparing dose distributions. It is calculated in dedicated software, and its results verification is not performed. Thus we developed an automatic tool for patient-specific QA results verification through high accuracy machine learning (ML) models based on the radiomics characteristics extraction from gamma images. We used 158 patient-specific QA tests and extracted 105 radiomics features from each gamma image. Three random forest models were developed (ML I, ML II, and ML III). ML I and ML II verified the gamma image approval using criteria of 2%/2mm/15% threshold and 3%/3mm/15% threshold, respectively. ML III verified if the gamma analyzes software recommended protocol was followed to detect if the TPS grid modification step was done. The models were based on the most important features selected using the mean decreased impurity, and their performances were evaluated. ML I included 25 features. Its accuracy was 0.85 using the test set and 0.84 using dataset B. ML II included 10 features, and its accuracy with the test set was 0.98; the same value was achieved using the never seen data (dataset B). The First-order 10th percentile feature was identified as a feature strongly related to the approved classification. ML III selected 23 features with an accuracy of 0.99 for test set and 0.98 for dataset B. An automatic workflow example for gamma analyses QA results verification could be proposed combining the models to detect grid inconsistencies on software evaluation, followed by the test approval classification.     

Comparative study of radiomics,tumor morphology,and clinicopathological factors in predicting overall survival of patients with rectal cancer before surgery

We compared the ability of a radiomics model, morphological imaging model, and clinicopathological risk model to predict 3-year overall survival (OS) in 206 patients with rectal cancer who underwent radical surgery and had magnetic resonance imaging, clinicopathological, and OS data available. The patients were randomized to a training cohort (n = 146) and a verification cohort (n = 60). Radiomics features were extracted from preoperative T2-weighted images, and a radiomics score model was constructed. Factors that were significant in the Cox multivariate analysis were used to construct the final morphological tumor model and clinicopathological model. A comprehensive model in the form of a line chart was established by combining the three models. Ten radiomics features significantly related to OS were selected to construct the radiomics feature model and calculate the radiomics score. In the morphological model, mesorectal extension depth and distance between the lower tumor margin and the anal margin were significant prognostic factors. N stage was the only significant clinicopathological factor. The comprehensive model combined with the above factors had the best prediction performance for OS. The C-index had a predictive performance of 0.872 (95% confidence interval [CI]: 0.832–0.912) in the training cohort and 0.944 (95% CI: 0.890–0.990) in the verification cohort, which was better than for any single model. The comprehensive model was divided into high-risk and low-risk groups. Kaplan-Meier curve analysis showed that all factors were significantly correlated with poor OS in the high-risk group. A comprehensive nomogram based on multi-model radiomics features can predict 3-year OS after rectal cancer surgery.     

Multi-parametric MRI lesion heterogeneity biomarkers for breast cancer diagnosis

PurposeTo identify intra-lesion imaging heterogeneity biomarkers in multi-parametric Magnetic Resonance Imaging (mpMRI) for breast lesion diagnosis.MethodsDynamic Contrast Enhanced (DCE) and Diffusion Weighted Imaging (DWI) of 73 female patients, with 85 histologically verified breast lesions were acquired. Non-rigid multi-resolution registration was utilized to spatially align sequences. Four (4) DCE (2^nd post-contrast frame, Initial-Enhancement, Post-Initial-Enhancement and Signal-Enhancement-Ratio) and one (1) DWI (Apparent-Diffusion-Coefficient) representations were analyzed, considering a representative lesion slice. 11 1^st-order-statistics and 16 texture features (Gray-Level-Co-occurrence-Matrix (GLCM) and Gray-Level-Run-Length-Matrix (GLRLM) based) were derived from lesion segments, provided by Fuzzy C-Means segmentation, across the 5 representations, resulting in 135 features. Least-Absolute-Shrinkage and Selection-Operator (LASSO) regression was utilized to select optimal feature subsets, subsequently fed into 3 classification schemes: Logistic-Regression (LR), Random-Forest (RF), Support-Vector-Machine-Sequential-Minimal-Optimization (SVM-SMO), assessed with Receiver-Operating-Characteristic (ROC) analysis.ResultsLASSO regression resulted in 7, 6 and 7 features subsets from DCE, DWI and mpMRI, respectively. Best classification performance was obtained by the RF multi-parametric scheme (Area-Under-ROC-Curve, (AUC) ± Standard-Error (SE), AUC ± SE = 0.984 ± 0.025), as compared to DCE (AUC ± SE = 0.961 ± 0.030) and DWI (AUC ± SE = 0.938 ± 0.032) and statistically significantly higher as compared to DWI. The selected mpMRI feature subset highlights the significance of entropy (1^st-order-statistics and 2^nd-order-statistics (GLCM)) and percentile features extracted from 2^nd post-contrast frame, PIE, SER maps and ADC map.ConclusionCapturing breast intra-lesion heterogeneity, across mpMRI lesion segments with 1^st-order-statistics and texture features (GLCM and GLRLM based), offers a valuable diagnostic tool for breast cancer.     

Challenges of predicting the potential distribution of a slow-spreading invader: a habitat suitability map for an invasive riparian tree

Understanding the potential spread of invasive species is essential for land managers to prevent their establishment and restore impacted habitat. Habitat suitability modeling provides a tool for researchers and managers to understand the potential extent of invasive species spread. Our goal was to use habitat suitability modeling to map potential habitat of the riparian plant invader, Russian olive (Elaeagnus angustifolia). Russian olive has invaded riparian habitat across North America and is continuing to expand its range. We compiled 11 disparate datasets for Russian olive presence locations (n = 1,051 points and 139 polygons) in the western US and used Maximum entropy (Maxent) modeling to develop two habitat suitability maps for Russian olive in the western United States: one with coarse-scale water data and one with fine-scale water data. Our models were able to accurately predict current suitable Russian olive habitat (Coarse model: training AUC = 0.938, test AUC = 0.907; Fine model: training AUC = 0.923, test AUC = 0.885). Distance to water was the most important predictor for Russian olive presence in our coarse-scale water model, but it was only the fifth most important variable in the fine-scale model, suggesting that when water bodies are considered on a fine scale, Russian olive does not necessarily rely on water. Our model predicted that Russian olive has suitable habitat further west from its current distribution, expanding into the west coast and central North America. Our methodology proves useful for identifying potential future areas of invasion. Model results may be influenced by locations of cultivated individuals and sampling bias. Further study is needed to examine the potential for Russian olive to invade beyond its current range. Habitat suitability modeling provides an essential tool for enhancing our understanding of invasive species spread.     

Spatial distribution and effects of Lantana camara invasion on soil properties and native vegetation community composition in Tugwi-Mukosi Recreational Park,Masvingo, Zimbabwe

This study investigated spatial distribution and effects of Lantana camara invasion on soil properties and vegetation composition in Tugwi-Mukosi Recreational Park. Supervised classification with the Random Forest classifier was used to map L. camara in the park. The study area was stratified into two categories based on soil type and the extent of L. camara invasion. Stratified random sampling was used in data collection to assess differences in species diversity and soil properties in the park. Results revealed that L. camara covered an area of 1772 ha. Using Random Forest classification, the study obtained an accuracy score of 92.1% and an F1 score of 86.25%. Two-way ANOVA showed significant effect of L. camara invasion on soil moisture (F = 28.143, p = 0.000), organic matter (F = 13.377, p = 0.003), pH (F = 1272.369, p = 0.000), total nitrogen (F = 51.762, p = 0.000), total phosphorus (F = 5.000, p = 0.045), woody species density (F = 4.987, p = 0.027), basal area (F = 10.393, p = 0.001), grass species richness (F = 196.258, p = 0.000) and grass cover (F = 3.637, p = 0.042). These results suggest that L. camara was modifying the soil and vegetation properties of the ecosystem which has implications on biodiversity.     

Early risk-assessment of patients with nasopharyngeal carcinoma: the added prognostic value of MR-based radiomics

ObjectivesTo assess the additive prognostic value of MR-based radiomics in predicting progression-free survival (PFS) in patients with nasopharyngeal carcinoma (NPC)MethodsPatients newly diagnosed with non-metastatic NPC between June 2006 and October 2019 were retrospectively included and randomly grouped into training and test cohorts (7:3 ratio). Radiomic features (n=213) were extracted from T2-weighted and contrast-enhanced T1-weighted MRI. The patients were staged according to the 8^th edition of American Joint Committee on Cancer Staging Manual. The least absolute shrinkage and selection operator was used to select the relevant radiomic features. Univariate and multivariate Cox proportional hazards analyses were conducted for PFS, yielding three different survival models (clinical, stage, and radiomic). The integrated time-dependent area under the curve (iAUC) for PFS was calculated and compared among different combinations of survival models, and the analysis of variance was used to compare the survival models. The prognostic performance of all models was validated using a test set with integrated Brier scores.ResultsThis study included 81 patients (training cohort=57; test cohort=24), and the mean PFS was 57.5 ± 43.6 months. In the training cohort, the prognostic performances of survival models improved significantly with the addition of radiomics to the clinical (iAUC, 0.72–0.80; p=0.04), stage (iAUC, 0.70–0.79; p=0.001), and combined models (iAUC, 0.76–0.81; p<0.001). In the test cohort, the radiomics and combined survival models were robustly validated for their ability to predict PFS.ConclusionIntegration of MR-based radiomic features with clinical and stage variables improved the prediction PFS in patients diagnosed with NPC.     

CT radiomics combined with clinical variables for predicting the overall survival of hepatocellular carcinoma patients after hepatectomy

PurposeTo establish a model for assessing the overall survival (OS) of the hepatocellular carcinoma (HCC) patients after hepatectomy based on the clinical and radiomics features.MethodsThis study recruited a total of 267 patients with HCC, which were randomly divided into the training (N = 188) and validation (N = 79) cohorts. In the training cohort, radiomic features were selected with the intra-reader and inter-reader correlation coefficient (ICC), Spearman's correlation coefficient, and the least absolute shrinkage and selection operator (LASSO). The radiomics signatures were built by COX regression analysis and compared the predictive potential in the different phases (arterial, portal, and double-phase) and regions of interest (tumor, peritumor 3 mm, peritumor 5 mm). A clinical-radiomics model (CR model) was established by combining the radiomics signatures and clinical risk factors. The validation cohort was used to validate the proposed models.ResultsA total of 267 patients 86 (45.74%) and 37 (46.84%) patients died in the training and validation cohorts, respectively. Among all the radiomics signatures, those based on the tumor and peritumor (5 mm) (AP-TP5-Signature) showed the best prognostic potential (training cohort 1–3 years AUC:0.774–0.837; validation cohort 1–3 years AUC:0.754–0.810). The CR model showed better discrimination, calibration, and clinical applicability as compared to the clinical model and radiomics features. In addition, the CR model could perform risk-stratification and also allowed for significant discrimination between the Kaplan-Meier curves in most of the subgroups.ConclusionsThe CR model could predict the OS of the HCC patients after hepatectomy.     

Precipitation and winter temperature predict long‐term range‐scale abundance changes in Western North American birds

Predicting biodiversity responses to climate change remains a difficult challenge, especially in climatically complex regions where precipitation is a limiting factor. Though statistical climatic envelope models are frequently used to project future scenarios for species distributions under climate change, these models are rarely tested using empirical data. We used long‐term data on bird distributions and abundance covering five states in the western US and in the Canadian province of British Columbia to test the capacity of statistical models to predict temporal changes in bird populations over a 32‐year period. Using boosted regression trees, we built presence‐absence and abundance models that related the presence and abundance of 132 bird species to spatial variation in climatic conditions. Presence/absence models built using 1970–1974 data forecast the distributions of the majority of species in the later time period, 1998–2002 (mean AUC = 0.79 ± 0.01). Hindcast models performed equivalently (mean AUC = 0.82 ± 0.01). Correlations between observed and predicted abundances were also statistically significant for most species (forecast mean Spearman′s ρ = 0.34 ± 0.02, hindcast = 0.39 ± 0.02). The most stringent test is to test predicted changes in geographic patterns through time. Observed changes in abundance patterns were significantly positively correlated with those predicted for 59% of species (mean Spearman′s ρ = 0.28 ± 0.02, across all species). Three precipitation variables (for the wettest month, breeding season, and driest month) and minimum temperature of the coldest month were the most important predictors of bird distributions and abundances in this region, and hence of abundance changes through time. Our results suggest that models describing associations between climatic variables and abundance patterns can predict changes through time for some species, and that changes in precipitation and winter temperature appear to have already driven shifts in the geographic patterns of abundance of bird populations in western North America.     

Modeling and detection of invasive trees using UAV image and machine learning in a subtropical forest in Brazil

Protected areas play an extremely important role in the conservation of global biodiversity. However, these areas are subject to the introduction of invasive alien species (IAS), which cause damage to native environments. The present study aimed to use images obtained by Unmanned Aerial Vehicles (UAVs) combined with machine learning (ML) algorithms to identify the IAS Hovenia dulcis in a Conservation Unit in southern Brazil. Field data were obtained in a sample area, where the floristic survey of the H. dulcis species was carried out. To obtain remote data, a UAV with a built-in RGB sensor was used. Subsequently, the images were processed for orthomosaic generation and the spatial distribution of the inventoried species, based on manual photointerpretation. Furthermore, in the supervised classification process, four classes of interest were defined: H. dulcis, similar species, shade, and other species. The process involved two approaches (pixel-based - PB and object-based image analysis - OBIA) and two ML algorithms were compared (Random Forest - RF and Support Vector Machine - SVM). Samples were separated into 90% for training and 10% for model validation. For performance analysis, overall accuracy (OA) and Kappa index metrics were calculated. The results show that the RF algorithm in the PB approach had the best performance in the classification of the IAS H. dulcis, presenting a kappa of 0.87 and OA of 91.5%, in the training data set and 90.91% of success in the model validation dataset. Our study demonstrated to be able to reach the results to respond to the raised hypotheses. Furthermore, the UAV-RGB data combined with ML are highly accurate to identify H. dulcis in relation to the other species that make up the forest stratum of the study area.     

Distinct tumor bacterial microbiome in lung adenocarcinomas manifested as radiological subsolid nodules

ObjectivesIncreasing evidence indicates that microbiota dysbiosis in the human body may play vital roles in carcinogenesis. However, the relationship between microbiome and lung cancer remains unclear. In this study, we aimed to characterize the microbiome in early stage of lung adenocarcinoma (LUAD), which presented as subsolid nodules (SSN) or solid nodules (SN).Materials and MethodsWe performed 16S rRNA sequencing of 35 pairs (10 SSN and 25 SN) of LUAD tumor tissues and paired adjacent normal tissues. Machine learning was used to identify microbial signatures and construct predictive models.ResultsSSN has higher microbiome richness and diversity compared with SN (richness p = 0.017, Shannon index p = 0.17), and the microbiome composition of SSN is distinct from that of SN (Bray-Curtis p = 0.013, unweighted unifrac p = 0.001). Phylum Chloroflexi (p = 0.009), Gemmatimonadetes (p = 0.018) and genus including Cloacibacterium (p = 0.003), Subdoligranulum (p = 0.002), and Mycobacterium (p = 0.034) were significantly increased in SSN. Tumor and normal tissues had similar richness and diversity, as well as overall microbiome composition. Probiotics with anti-cancer potential, like Lactobacillus, showed elevated levels in normal tissues (p = 0.018). A random forest model with 20 genera-based biomarkers achieved high accuracy for LUAD prediction (area under curve, AUC = 0.879). Meanwhile, a five genera-based signature can accurately discriminate SSN between SN (AUC = 0.950). Cross-validation of these two models also showed high predictive performance (LUAD AUC = 0.813, SSN AUC = 0.933).ConclusionsThis study demonstrates, for the first time, the tumor bacterial microbiome composition of LUAD manifested as SSN is distinct from that presented as SN, which adds new knowledge to SSN in the perspective of microbiome. Furthermore, microbiome signatures showed good performance to predict LUAD or SSN.     

Regulation of basal and norepinephrine-induced cAMP and ICa in hiPSC-cardiomyocytes: Effects of culture conditions and comparison to adult human atrial cardiomyocytes

BackgroundThere is ongoing interest in generating cardiomyocytes derived from human induced pluripotent stem cells (hiPSC) to study human cardiac physiology and pathophysiology. Recently we found that norepinephrine-stimulated calcium currents (I_Ca) in hiPSC-cardiomyocytes were smaller in conventional monolayers (ML) than in engineered heart tissue (EHT). In order to elucidate culture specific regulation of β₁-adrenoceptor (β₁-AR) responses we investigated whether action of phosphodiesterases (PDEs) may limit norepinephrine effects on I_Ca and on cytosolic cAMP in hiPSC-cardiomyocytes. Results were compared to adult human atrial cardiomyocytes.MethodsAdult human atrial cardiomyocytes were isolated from tissue samples obtained during open heart surgery. All patients were in sinus rhythm. HiPSC-cardiomyocytes were dissociated from ML and EHT. Förster-resonance energy transfer (FRET) was used to monitor cytosolic cAMP (Epac1-camps sensor, transfected by adenovirus). I_Ca was recorded by whole-cell patch clamp technique. Cilostamide (300 nM) and rolipram (10 μM) were used to inhibit PDE3 and PDE4, respectively. β₁-AR were stimulated with the physiological agonist norepinephrine (100 μM).ResultsIn adult human atrial cardiomyocytes, norepinephrine increased cytosolic cAMP FRET ratio by +13.7 ± 1.5% (n = 10/9, mean ± SEM, number of cells/number patients) and I_Ca by +10.4 ± 1.5 pA/pF (n = 15/10). This effect was not further increased in the concomitant presence of rolipram, cilostamide and norepinephrine, indicating saturation by norepinephrine alone. In ML hiPSC-cardiomyocytes, norepinephrine exerted smaller increases in cytosolic cAMP and I_Ca (FRET +9.6 ± 0.5% n = 52/21, number of cells/number of ML or EHT, and I_Ca + 1.4 ± 0.2 pA/pF n = 34/7, p < 0.05 each) and both were augmented in the presence of the PDE4 inhibitor rolipram (FRET +16.7 ± 0.8% n = 94/26 and I_Ca + 5.6 ± 1.4 pA/pF n = 11/5, p < 0.05 each). Cilostamide increased the response to norepinephrine on FRET (+12.7 ± 0.5% n = 91/19, p < 0.05), but not on I_Ca. In EHT hiPSC-cardiomyocytes, norepinephrine responses on both, FRET and I_Ca, were larger than in ML (FRET +12.1 ± 0.3% n = 87/32 and I_Ca + 3.3 ± 0.2 pA/pF n = 13/5, p < 0.05 each). Rolipram augmented the norepinephrine effect on I_Ca (+6.2 ± 1.6 pA/pF; p < 0.05 vs. norepinephrine alone, n = 10/4), but not on FRET.ConclusionOur results show culture-dependent differences in hiPSC-cardiomyocytes. In conventional ML but not in EHT, maximum norepinephrine effects on cytosolic cAMP depend on PDE3 and PDE4, suggesting immaturity when compared to the situation in adult human atrial cardiomyocytes. The smaller I_Ca responses to norepinephrine in ML and EHT vs. adult human atrial cardiomyocytes depend at least partially on a non-physiological large impact of PDE4 in hiPSC-cardiomyocytes.     

Automated detection and classification of southern African Roman seabream using mask R-CNN

The availability of relatively cheap, high-resolution digital cameras has led to an exponential increase in the capture of natural environments and their inhabitants. Video-based surveys are particularly useful in the underwater domain where observation by humans can be expensive, dangerous, inaccessible, or destructive to the natural environment. However, a large majority of marine data has never gone through analysis by human experts – a process that is slow, expensive, and not scalable. We test a Mask R-CNN object detection framework for the automated localisation, classification, counting and tracking of fish in unconstrained underwater environments. We present a novel, labelled image dataset of roman seabream (Chrysoblephus laticeps), a fish species endemic to Southern Africa, to train and validate the accuracy of our model. The Mask R-CNN model accurately detected and classified roman seabream on the training dataset (mAP50 = 80.29%), validation dataset (mAP50 = 80.35%), as well as on previously unseen footage (test dataset) (mAP50 = 81.45%). The fact that the model performs well on previously unseen data suggests that it is capable of generalising to new streams of data not included in this research.     

2D and 3D CT Radiomics Features Prognostic Performance Comparison in Non-Small Cell Lung Cancer

OBJECTIVE: To compare 2D and 3D radiomics features prognostic performance differences in CT images of non-small cell lung cancer (NSCLC). METHOD: We enrolled 588 NSCLC patients from three independent cohorts. Two sets of 463 patients from two different institutes were used as the training cohort. The remaining cohort with 125 patients was set as the validation cohort. A total of 1014 radiomics features (507 2D features and 507 3D features correspondingly) were assessed. Based on the dichotomized survival data, 2D and 3D radiomics indicators were calculated for each patient by trained classifiers. We used the area under the receiver operating characteristic curve (AUC) to assess the prediction performance of trained classifiers (the support vector machine and logistic regression). Kaplan–Meier and Cox hazard survival analyses were also employed. Harrell's concordance index (C-Index) and Akaike's information criteria (AIC) were applied to assess the trained models. RESULTS: Radiomics indicators were built and compared by AUCs. In the training cohort, 2D_AUC = 0.653, 3D_AUC = 0.671. In the validation cohort, 2D_AUC = 0.755, 3D_AUC = 0.663. Both 2D and 3D trained indicators achieved significant results (P < .05) in the Kaplan-Meier analysis and Cox regression. In the validation cohort, 2D Cox model had a C-Index = 0.683 and AIC = 789.047; 3D Cox model obtained a C-Index = 0.632 and AIC = 799.409. CONCLUSION: Both 2D and 3D CT radiomics features have a certain prognostic ability in NSCLC, but 2D features showed better performance in our tests. Considering the cost of the radiomics features calculation, 2D features are more recommended for use in the current study.     

Two Panels of Plasma MicroRNAs as Non-Invasive Biomarkers for Prediction of Recurrence in Resectable NSCLC

The diagnosis of non-small cell lung carcinoma (NSCLC) at an early stage, as well as better prediction of outcome remains clinically challenging due to the lack of specific and robust non-invasive markers. The discovery of microRNAs (miRNAs), particularly those found in the bloodstream, has opened up new perspectives for tumor diagnosis and prognosis. The aim of our study was to determine whether expression profiles of specific miRNAs in plasma could accurately discriminate between NSCLC patients and controls, and whether they are able to predict the prognosis of resectable NSCLC patients. We therefore evaluated a series of seventeen NSCLC-related miRNAs by quantitative real-time (qRT)-PCR in plasma from 52 patients with I-IIIA stages NSCLC, 10 patients with chronic obstructive pulmonary disease (COPD) and 20-age, sex and smoking status-matched healthy individuals. We identified an eleven-plasma miRNA panel that could distinguish NSCLC patients from healthy subjects (AUC = 0.879). A six-plasma miRNA panel was able to discriminate between NSCLC patients and COPD patients (AUC = 0.944). Furthermore, we identified a three-miRNA plasma signature (high miR-155-5p, high miR-223-3p, and low miR-126-3p) that significantly associated with a higher risk for progression in adenocarcinoma patients. In addition, a three-miRNA plasma panel (high miR-20a-5p, low miR-152-3p, and low miR-199a-5p) significantly predicted survival of squamous cell carcinoma patients. In conclusion, we identified two plasma miRNA expression profiles that may be useful for predicting the outcome of patients with resectable NSCLC.     

Pooled analysis of clinical outcome for EGFR TKI‐treated patients with EGFR mutation‐positive NSCLC

Patients with non‐small‐cell lung cancer (NSCLC) appear to gain particular benefit from treatment with epidermal growth factor receptor (EGFR) tyrosine‐kinase inhibitors (TKI) if their disease tests positive for EGFR activating mutations. Recently, several large, controlled, phase III studies have been published in NSCLC patients with EGFR mutation‐positive tumours. Given the increased patient dataset now available, a comprehensive literature search for EGFR TKIs or chemotherapy in EGFR mutation‐positive NSCLC was undertaken to update the results of a previously published pooled analysis. Pooling eligible progression‐free survival (PFS) data from 27 erlotinib studies (n = 731), 54 gefitinib studies (n = 1802) and 20 chemotherapy studies (n = 984) provided median PFS values for each treatment. The pooled median PFS was: 12.4 months (95% accuracy intervals [AI] 11.6–13.4) for erlotinib‐treated patients; 9.4 months (95% AI 9.0–9.8) for gefitinib‐treated patients; and 5.6 months (95% AI 5.3–6.0) for chemotherapy. Both erlotinib and gefitinib resulted in significantly longer PFS than chemotherapy (permutation testing; P = 0.000 and P = 0.000, respectively). Data on more recent TKIs (afatinib, dacomitinib and icotinib) were insufficient at this time‐point to carry out a pooled PFS analysis on these compounds. The results of this updated pooled analysis suggest a substantial clear PFS benefit of treating patients with EGFR mutation‐positive NSCLC with erlotinib or gefitinib compared with chemotherapy.