REFINED-CNN framework for survival prediction with high-dimensional features.

Robust and accurate survival prediction of clinical trials using high-throughput genomics data is a fundamental challenge in pharmacogenomics. Current machine learning tools often provide limited predictive performance and model interpretation in these settings. In the present study, we extend the application of REFINED-CNN from regression tasks to making survival predictions, by mapping high-dimensional RNA sequencing data into REFINED images which are conducive to CNN modeling. We show that the REFINED-CNN survival model can be easily adapted to new tasks of a similar nature (e.g., predicting on new cancer types) using transfer learning with a low number of patients. Furthermore, the model can also be interpreted both locally and globally through risk score back propagation that quantifies each feature (e.g., gene) importance in survival prediction task for the patient or cancer type of interest.

Federated learning framework integrating REFINED CNN and Deep Regression Forests.

Summary: Predictive learning from medical data incurs additional challenge due to concerns over privacy and security of personal data. Federated learning, intentionally structured to preserve high level of privacy, is emerging to be an attractive way to generate cross-silo predictions in medical scenarios. However, the impact of severe population-level heterogeneity on federated learners is not well explored. In this article, we propose a methodology to detect presence of population heterogeneity in federated settings and propose a solution to handle such heterogeneity by developing a federated version of Deep Regression Forests. Additionally, we demonstrate that the recently conceptualized REpresentation of Features as Images with NEighborhood Dependencies CNN framework can be combined with the proposed Federated Deep Regression Forests to provide improved performance as compared to existing approaches. Availability and implementation: The Python source code for reproducing the main results are available on GitHub: https://github.com/DanielNolte/FederatedDeepRegressionForests. Contact: ranadip.pal@ttu.edu. Supplementary information: Supplementary data are available at Bioinformatics Advances online.

Prognostic MicroRNA Panel for HCV-Associated HCC: Integrating Computational Biology and Clinical Validation.

Early detection of hepatocellular carcinoma (HCC) will reduce morbidity and mortality rates of this widely spread disease. Dysregulation in microRNA (miRNA) expression is associated with HCC progression. The objective is to identify a panel of differentially expressed miRNAs (DE-miRNAs) to enhance HCC early prediction in hepatitis C virus (HCV) infected patients. Candidate miRNAs were selected using a bioinformatic analysis of microarray and RNA-sequencing datasets, resulting in nine DE-miRNAs (miR-142, miR-150, miR-183, miR-199a, miR-215, miR-217, miR-224, miR-424, and miR-3607). Their expressions were validated in the serum of 44 healthy individuals, 62 non-cirrhotic HCV patients, 67 cirrhotic-HCV, and 72 HCV-associated-HCC patients using real-time PCR (qPCR). There was a significant increase in serum concentrations of the nine-candidate miRNAs in HCC and HCV patients relative to healthy individuals. MiR-424, miR-199a, miR-142, and miR-224 expressions were significantly altered in HCC compared to non-cirrhotic patients. A panel of five miRNAs improved sensitivity and specificity of HCC detection to 100% and 95.12% relative to healthy controls. Distinguishing HCC from HCV-treated patients was achieved by 70.8% sensitivity and 61.9% specificity using the combined panel, compared to alpha-fetoprotein (51.4% sensitivity and 60.67% specificity). These preliminary data show that the novel miRNAs panel (miR-150, miR-199a, miR-224, miR-424, and miR-3607) could serve as a potential non-invasive biomarker for HCC early prediction in chronic HCV patients. Further prospective studies on a larger cohort of patients should be conducted to assess the potential prognostic ability of the miRNAs panel.

Investigation of REFINED CNN ensemble learning for anti-cancer drug sensitivity prediction.

MOTIVATION: Anti-cancer drug sensitivity prediction using deep learning models for individual cell line is a significant challenge in personalized medicine. Recently developed REFINED (REpresentation of Features as Images with NEighborhood Dependencies) CNN (Convolutional Neural Network)-based models have shown promising results in improving drug sensitivity prediction. The primary idea behind REFINED-CNN is representing high dimensional vectors as compact images with spatial correlations that can benefit from CNN architectures. However, the mapping from a high dimensional vector to a compact 2D image depends on the a priori choice of the distance metric and projection scheme with limited empirical procedures guiding these choices. RESULTS: In this article, we consider an ensemble of REFINED-CNN built under different choices of distance metrics and/or projection schemes that can improve upon a single projection based REFINED-CNN model. Results, illustrated using NCI60 and NCI-ALMANAC databases, demonstrate that the ensemble approaches can provide significant improvement in prediction performance as compared to individual models. We also develop the theoretical framework for combining different distance metrics to arrive at a single 2D mapping. Results demonstrated that distance-averaged REFINED-CNN produced comparable performance as obtained from stacking REFINED-CNN ensemble but with significantly lower computational cost. AVAILABILITY AND IMPLEMENTATION: The source code, scripts, and data used in the paper have been deposited in GitHub (https://github.com/omidbazgirTTU/IntegratedREFINED). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Representation of features as images with neighborhood dependencies for compatibility with convolutional neural networks.

Deep learning with Convolutional Neural Networks has shown great promise in image-based classification and enhancement but is often unsuitable for predictive modeling using features without spatial correlations. We present a feature representation approach termed REFINED (REpresentation of Features as Images with NEighborhood Dependencies) to arrange high-dimensional vectors in a compact image form conducible for CNN-based deep learning. We consider the similarities between features to generate a concise feature map in the form of a two-dimensional image by minimizing the pairwise distance values following a Bayesian Metric Multidimensional Scaling Approach. We hypothesize that this approach enables embedded feature extraction and, integrated with CNN-based deep learning, can boost the predictive accuracy. We illustrate the superior predictive capabilities of the proposed framework as compared to state-of-the-art methodologies in drug sensitivity prediction scenarios using synthetic datasets, drug chemical descriptors as predictors from NCI60, and both transcriptomic information and drug descriptors as predictors from GDSC.

A Classification System for Assessment and Home Monitoring of Tremor in Patients with Parkinson's Disease.

BACKGROUND: Tremor is one of the most common symptoms of Parkinson's disease (PD), which is widely being used in the diagnosis procedure. Accurate estimation of PD tremor based on Unified PD Rating Scale (UPDRS) provides aid for physicians in prescription and home monitoring. This article presents a robust design of a classification system to estimate PD patient's hand tremors and the results of the proposed system as compared to the UPDRS. METHODS: A smartphone accelerometer sensor is used for accurate and noninvasive data acquisition. We applied short-time Fourier transform to time series data of 52 PD patients. Features were extracted based on the severity of PD patients' hand tremor. The wrapper method was employed to determine the most discriminative subset of the extracted features. Four different classifiers were implemented for achieving best possible accuracy in the estimation of PD hand tremor based on UPDRS. Of the four tested classifiers, the Naive Bayesian approach proved to be the most accurate one. RESULTS: The classification result for the assessment of PD tremor achieved close to 100% accuracy by selecting an optimum combination of extracted features of the acceleration signal acquired. For home health-care monitoring, the proposed algorithm was also implemented on a cost-effective embedded system equipped with a microcontroller, and the implemented classification algorithm achieved 93.8% average accuracy. CONCLUSIONS: The accuracy result of both implemented systems on MATLAB and microcontroller is acceptable in comparison with the previous works.