Consecutive Serial Non-Contrast CT Scan-Based Deep Learning Model Facilitates the Prediction of Tumor Invasiveness of Ground-Glass Nodules

IntroductionTumors are continuously evolving biological systems which can be monitored by medical imaging. Previous studies only focus on single timepoint images, whether the performance could be further improved by using serial noncontrast CT imaging obtained during nodule follow-up management remains unclear. In this study, we evaluated DL model for predicting tumor invasiveness of GGNs through analyzing time series CT imagesMethodsA total of 168 pathologically confirmed GGN cases (48 noninvasive lesions and 120 invasive lesions) were retrospectively collected and randomly assigned to the development dataset (n = 123) and independent testing dataset (n = 45). All patients underwent consecutive noncontrast CT examinations, and the baseline CT and 3-month follow-up CT images were collected. The gross region of interest (ROI) patches containing only tumor region and the full ROI patches including both tumor and peritumor regions were cropped from CT images. A baseline model was built on the image features and demographic features. Four DL models were proposed: two single-DL model using gross ROI (model 1) or full ROI patches (model 3) from baseline CT images, and two serial-DL models using gross ROI (model 2) or full ROI patches (model 4) from consecutive CT images (baseline scan and 3-month follow-up scan). In addition, a combined model integrating serial full ROI patches and clinical information was also constructed. The performance of these predictive models was assessed with respect to discrimination and clinical usefulness.ResultsThe area under the curve (AUC) of the baseline model, models 1, 2, 3, and 4 were 0.562 [(95% confidence interval (C)], 0.406~0.710), 0.693 (95% CI, 0.538–0.822), 0.787 (95% CI, 0.639–0.895), 0.727 (95% CI, 0.573–0.849), and 0.811 (95% CI, 0.667–0.912) in the independent testing dataset, respectively. The results indicated that the peritumor region had potential to contribute to tumor invasiveness prediction, and the model performance was further improved by integrating imaging scans at multiple timepoints. Furthermore, the combined model showed best discrimination ability, with AUC, sensitivity, specificity, and accuracy achieving 0.831 (95% CI, 0.690–0.926), 86.7%, 73.3%, and 82.2%, respectively.ConclusionThe DL model integrating full ROIs from serial CT images shows improved predictive performance in differentiating noninvasive from invasive GGNs than the model using only baseline CT images, which could benefit the clinical management of GGNs.

Quality Assessment of Protein Docking Models Based on Graph Neural Network

Protein docking provides a structural basis for the design of drugs and vaccines. Among the processes of protein docking, quality assessment (QA) is utilized to pick near-native models from numerous protein docking candidate conformations, and it directly determines the final docking results. Although extensive efforts have been made to improve QA accuracy, it is still the bottleneck of current protein docking systems. In this paper, we presented a Deep Graph Attention Neural Network (DGANN) to evaluate and rank protein docking candidate models. DGANN learns inter-residue physio-chemical properties and structural fitness across the two protein monomers in a docking model and generates their probabilities of near-native models. On the ZDOCK decoy benchmark, our DGANN outperformed the ranking provided by ZDOCK in terms of ranking good models into the top selections. Furthermore, we conducted comparative experiments on an independent testing dataset, and the results also demonstrated the superiority and generalization of our proposed method.

MRI-based Radiomics for Prognosis of Pediatric Diffuse Intrinsic Pontine Glioma: An International Study

Background Diffuse Intrinsic pontine gliomas (DIPGs) are lethal pediatric brain tumors. Presently, MRI is the mainstay of disease diagnosis and surveillance. We identify clinically significant computational features from MRI and create a prognostic machine learning model. Methods We isolated tumor volumes of T1-post contrast (T1) and T2-weighted (T2) MRIs from 177 treatment-naïve DIPG patients from an international cohort for model training and testing. The Quantitative Image Feature Pipeline and PyRadiomics was used for feature extraction. Ten-fold cross-validation of LASSO Cox regression selected optimal features to predict overall survival (OS) in the training dataset and tested in the independent testing dataset. We analyzed model performance using clinical variables (age at diagnosis and sex) only, radiomics only, and radiomics plus clinical variables. Results All selected features were intensity and texture-based on the wavelet filtered images (three T1 grey-level co-occurrence matrix (GLCM) texture features, T2 GLCM texture feature, and T2 first order-mean). This multivariable Cox model demonstrated a concordance of 0.68 [95% CI: 0.61-0.74] in the training dataset, significantly outperforming the clinical-only model (C=0.57 [95% CI: 0.49-0.64]). Adding clinical features to radiomics slightly improved performance (C=0.70 [95% CI: 0.64-0.77]). The combined radiomics and clinical model was validated in the independent testing dataset (C=0.59 [95% CI: 0.51-0.67], Noether’s test p=0.02). Conclusion In this international study, we demonstrate the use of radiomic signatures to create a machine learning model for DIPG prognostication. Standardized, quantitative approaches that objectively measure DIPG changes, including computational MRI evaluation, could offer new approaches to assessing tumor phenotype and serve a future role for optimizing clinical trial eligibility and tumor surveillance.

A Human-Algorithm Integration System for Hip Fracture Detection on Plain Radiography: System Development and Validation Study

Background Hip fracture is the most common type of fracture in elderly individuals. Numerous deep learning (DL) algorithms for plain pelvic radiographs (PXRs) have been applied to improve the accuracy of hip fracture diagnosis. However, their efficacy is still undetermined. Objective The objective of this study is to develop and validate a human-algorithm integration (HAI) system to improve the accuracy of hip fracture diagnosis in a real clinical environment. Methods The HAI system with hip fracture detection ability was developed using a deep learning algorithm trained on trauma registry data and 3605 PXRs from August 2008 to December 2016. To compare their diagnostic performance before and after HAI system assistance using an independent testing dataset, 34 physicians were recruited. We analyzed the physicians’ accuracy, sensitivity, specificity, and agreement with the algorithm; we also performed subgroup analyses according to physician specialty and experience. Furthermore, we applied the HAI system in the emergency departments of different hospitals to validate its value in the real world. Results With the support of the algorithm, which achieved 91% accuracy, the diagnostic performance of physicians was significantly improved in the independent testing dataset, as was revealed by the sensitivity (physician alone, median 95%; HAI, median 99%; P<.001), specificity (physician alone, median 90%; HAI, median 95%; P<.001), accuracy (physician alone, median 90%; HAI, median 96%; P<.001), and human-algorithm agreement [physician alone κ, median 0.69 (IQR 0.63-0.74); HAI κ, median 0.80 (IQR 0.76-0.82); P<.001. With the help of the HAI system, the primary physicians showed significant improvement in their diagnostic performance to levels comparable to those of consulting physicians, and both the experienced and less-experienced physicians benefited from the HAI system. After the HAI system had been applied in 3 departments for 5 months, 587 images were examined. The sensitivity, specificity, and accuracy of the HAI system for detecting hip fractures were 97%, 95.7%, and 96.08%, respectively. Conclusions HAI currently impacts health care, and integrating this technology into emergency departments is feasible. The developed HAI system can enhance physicians’ hip fracture diagnostic performance.

Rapid vessel segmentation and reconstruction of head and neck angiograms using 3D convolutional neural network

The computed tomography angiography (CTA) postprocessing manually recognized by technologists is extremely labor intensive and error prone. We propose an artificial intelligence reconstruction system supported by an optimized physiological anatomical-based 3D convolutional neural network that can automatically achieve CTA reconstruction in healthcare services. This system is trained and tested with 18,766 head and neck CTA scans from 5 tertiary hospitals in China collected between June 2017 and November 2018. The overall reconstruction accuracy of the independent testing dataset is 0.931. It is clinically applicable due to its consistency with manually processed images, which achieves a qualification rate of 92.1%. This system reduces the time consumed from 14.22 ± 3.64 min to 4.94 ± 0.36 min, the number of clicks from 115.87 ± 25.9 to 4 and the labor force from 3 to 1 technologist after five months application. Thus, the system facilitates clinical workflows and provides an opportunity for clinical technologists to improve humanistic patient care.

A Human-Algorithm Integration System for Hip Fracture Detection on Plain Radiography: System Development and Validation Study (Preprint)

BACKGROUND Hip fracture is the most common type of fracture in elderly individuals. Numerous deep learning (DL) algorithms for plain pelvic radiographs (PXRs) have been applied to improve the accuracy of hip fracture diagnosis. However, their efficacy is still undetermined. OBJECTIVE The objective of this study is to develop and validate a human-algorithm integration (HAI) system to improve the accuracy of hip fracture diagnosis in a real clinical environment. METHODS The HAI system with hip fracture detection ability was developed using a deep learning algorithm trained on trauma registry data and 3605 PXRs from August 2008 to December 2016. To compare their diagnostic performance before and after HAI system assistance using an independent testing dataset, 34 physicians were recruited. We analyzed the physicians’ accuracy, sensitivity, specificity, and agreement with the algorithm; we also performed subgroup analyses according to physician specialty and experience. Furthermore, we applied the HAI system in the emergency departments of different hospitals to validate its value in the real world. RESULTS With the support of the algorithm, which achieved 91% accuracy, the diagnostic performance of physicians was significantly improved in the independent testing dataset, as was revealed by the sensitivity (physician alone, median 95%; HAI, median 99%; <i>P</i>&lt;.001), specificity (physician alone, median 90%; HAI, median 95%; <i>P</i>&lt;.001), accuracy (physician alone, median 90%; HAI, median 96%; <i>P</i>&lt;.001), and human-algorithm agreement [physician alone κ, median 0.69 (IQR 0.63-0.74); HAI κ, median 0.80 (IQR 0.76-0.82); <i>P</i>&lt;.001. With the help of the HAI system, the primary physicians showed significant improvement in their diagnostic performance to levels comparable to those of consulting physicians, and both the experienced and less-experienced physicians benefited from the HAI system. After the HAI system had been applied in 3 departments for 5 months, 587 images were examined. The sensitivity, specificity, and accuracy of the HAI system for detecting hip fractures were 97%, 95.7%, and 96.08%, respectively. CONCLUSIONS HAI currently impacts health care, and integrating this technology into emergency departments is feasible. The developed HAI system can enhance physicians’ hip fracture diagnostic performance.

NeuroCS: A Tool to Predict Cleavage Sites of Neuropeptide Precursors

Background: Neuropeptides are a class of bioactive peptides produced from neuropeptide precursors through a series of extremely complex processes, mediating neuronal regulations in many aspects. Accurate identification of cleavage sites of neuropeptide precursors is of great significance for the development of neuroscience and brain science. Objective: With the explosive growth of neuropeptide precursor data, it is pretty much needed to develop bioinformatics methods for predicting neuropeptide precursors’ cleavage sites quickly and efficiently. Method : We started with processing the neuropeptide precursor data from SwissProt and NueoPedia into two sets of data, training dataset and testing dataset. Subsequently, six feature extraction schemes were applied to generate different feature sets and then feature selection methods were used to find the optimal feature subset of each. Thereafter the support vector machine was utilized to build models for different feature types. Finally, the performance of models were evaluated with the independent testing dataset. Results: Six models are built through support vector machine. Among them the enhanced amino acid composition-based model reaches the highest accuracy of 91.60% in the 5-fold cross validation. When evaluated with independent testing dataset, it also showed an excellent performance with a high accuracy of 90.37% and Area under Receiver Operating Characteristic curve up to 0.9576. Conclusion: The performance of the developed model was decent. Moreover, for users’ convenience, an online web server called NeuroCS is built, which is freely available at http://i.uestc.edu.cn/NeuroCS/dist/index.html#/. NeuroCS can be used to predict neuropeptide precursors’ cleavage sites effectively.

csDMA: an improved bioinformatics tool for identifying DNA 6 mA modifications via Chou’s 5-step rule

DNA N6-methyldeoxyadenosine (6 mA) modifications were first found more than 60 years ago but were thought to be only widespread in prokaryotes and unicellular eukaryotes. With the development of high-throughput sequencing technology, 6 mA modifications were found in different multicellular eukaryotes by using experimental methods. However, the experimental methods were time-consuming and costly, which makes it is very necessary to develop computational methods instead. In this study, a machine learning-based prediction tool, named csDMA, was developed for predicting 6 mA modifications. Firstly, three feature encoding schemes, Motif, Kmer, and Binary, were used to generate the feature matrix. Secondly, different algorithms were selected into the prediction model and the ExtraTrees model received the best AUC of 0.878 by using 5-fold cross-validation on the training dataset. Besides, the ExtraTrees model also received the best AUC of 0.893 on the independent testing dataset. Finally, we compared our method with state-of-the-art predictors and the results shown that our model achieved better performance than existing tools.

A Two-Layer Learning Architecture for Multi-Class Protein Folds Classification

Classification of protein folds plays a very important role in the protein structure discovery process, especially when traditional sequence alignment methods fail to yield convincing structural homologies. In this chapter, we have developed a two-layer learning architecture, named TLLA, for multi-class protein folds classification. In the first layer, OET-KNN (Optimized Evidence-Theoretic K Nearest Neighbors) is used as the component classifier to find the most probable K-folds of the query protein. In the second layer, we use support vector machine (SVM) to build the multi-class classifier just on the K-folds, generated in the first layer, rather than on all the 27 folds. For multi-feature combination, ensemble strategy based on voting is selected to give the final classification result. The standard percentage accuracy of our method at ~63% is achieved on the independent testing dataset, where most of the proteins have <25% sequence identity with those in the training dataset. The experimental evaluation based on a widely used benchmark dataset has shown that our approach outperforms the competing methods, implying our approach might become a useful vehicle in the literature.

A Two-Layer Learning Architecture for Multi-Class Protein Folds Classification

Classification of protein folds plays a very important role in the protein structure discovery process, especially when traditional sequence alignment methods fail to yield convincing structural homologies. In this chapter, we have developed a two-layer learning architecture, named TLLA, for multi-class protein folds classification. In the first layer, OET-KNN (Optimized Evidence-Theoretic K Nearest Neighbors) is used as the component classifier to find the most probable K-folds of the query protein. In the second layer, we use support vector machine (SVM) to build the multi-class classifier just on the K-folds, generated in the first layer, rather than on all the 27 folds. For multi-feature combination, ensemble strategy based on voting is selected to give the final classification result. The standard percentage accuracy of our method at ~63% is achieved on the independent testing dataset, where most of the proteins have <25% sequence identity with those in the training dataset. The experimental evaluation based on a widely used benchmark dataset has shown that our approach outperforms the competing methods, implying our approach might become a useful vehicle in the literature.