Weakly supervised large-scale pancreatic cancer detection using multi-instance learning.

Introduction: Early detection of pancreatic cancer continues to be a challenge due to the difficulty in accurately identifying specific signs or symptoms that might correlate with the onset of pancreatic cancer. Unlike breast or colon or prostate cancer where screening tests are often useful in identifying cancerous development, there are no tests to diagnose pancreatic cancers. As a result, most pancreatic cancers are diagnosed at an advanced stage, where treatment options, whether systemic therapy, radiation, or surgical interventions, offer limited efficacy. Methods: A two-stage weakly supervised deep learning-based model has been proposed to identify pancreatic tumors using computed tomography (CT) images from Henry Ford Health (HFH) and publicly available Memorial Sloan Kettering Cancer Center (MSKCC) data sets. In the first stage, the nnU-Net supervised segmentation model was used to crop an area in the location of the pancreas, which was trained on the MSKCC repository of 281 patient image sets with established pancreatic tumors. In the second stage, a multi-instance learning-based weakly supervised classification model was applied on the cropped pancreas region to segregate pancreatic tumors from normal appearing pancreas. The model was trained, tested, and validated on images obtained from an HFH repository with 463 cases and 2,882 controls. Results: The proposed deep learning model, the two-stage architecture, offers an accuracy of 0.907   ±   0.01, sensitivity of 0.905   ±   0.01, specificity of 0.908   ±   0.02, and AUC (ROC) 0.903   ±   0.01. The two-stage framework can automatically differentiate pancreatic tumor from non-tumor pancreas with improved accuracy on the HFH dataset. Discussion: The proposed two-stage deep learning architecture shows significantly enhanced performance for predicting the presence of a tumor in the pancreas using CT images compared with other reported studies in the literature.

Multiobjective Bayesian Optimization Framework for the Synthesis of Methanol from Syngas Using Interpretable Gaussian Process Models.

Methanol production has gained considerable interest on the laboratory and industrial scale as it is a renewable fuel and an excellent hydrogen energy storehouse. The formation of synthesis gas (CO/H2) and the conversion of synthesis gas to methanol are the two basic catalytic processes used in methanol production. Machine learning (ML) approaches have recently emerged as powerful tools in reaction informatics. Inspired by these, we employ Gaussian process regression (GPR) to the model conversion of carbon monoxide (CO) and selectivity of the methanol product using data sets obtained from experimental investigations to capture uncertainty in prediction values. The results indicate that the proposed GPR model can accurately predict CO conversion and methanol selectivity as compared to other ML models. Further, the factors that influence the predictions are identified from the best GPR model employing "Shapley Additive exPlanations" (SHAP). After interpretation, the essential input features are found to be the inlet mole fraction of CO (Y(CO, in)) and the net inlet flow rate (Fin(nL/min)) for our best prediction GPR models, irrespective of our data sets. These interpretable models are employed for Bayesian optimization in a weighted multiobjective framework to obtain the optimal operating points, namely, maximization of both selectivity and conversion.

Batch-to-Batch Adaptive Iterative Learning Control-Explicit Model Predictive Control Two-Tier Framework for the Control of Batch Transesterification Process.

To harness energy security and reduce carbon emissions, humankind is trying to switch toward renewable energy resources. To this extent, fatty acid methyl esters, also known as biodiesel, are popularly used as a green fuel. Fatty acid methyl esters can be produced by a batch transesterification reaction between vegetable oil and alcohol. Being a batch process, fatty acid methyl esters production is beset with issues such as uncertainties and unsteady state behavior, and therefore, adequate process control measures are necessitated. In this study, we have proposed a novel two-tier framework for the control of the fatty acid methyl esters production process. The proposed approach combines the constrained batch-to-batch iterative learning control technique and explicit model predictive control to obtain the desired concentration of the fatty acid methyl esters. In particular, the batch-to-batch iterative learning control technique is used to generate reactor temperature set-points, which is further utilized to obtain an optimal coolant flow rate by solving a quadratic objective cost function, with the help of explicit model predictive control. Our simulation results indicate that the fatty acid methyl esters concentration trajectory converges to the desired batch trajectory within four batches for uncertainty in activation energy and six batches for uncertainty in both inlet concentration of triglyceride and in activation energy even in the presence of process disturbances. The proposed approach was compared to the heuristic-based approach and constraint iterative learning control approach to showcase its efficacy.

Application of machine learning in understanding plant virus pathogenesis: trends and perspectives on emergence, diagnosis, host-virus interplay and management.

BACKGROUND: Inclusion of high throughput technologies in the field of biology has generated massive amounts of data in the recent years. Now, transforming these huge volumes of data into knowledge is the primary challenge in computational biology. The traditional methods of data analysis have failed to carry out the task. Hence, researchers are turning to machine learning based approaches for the analysis of high-dimensional big data. In machine learning, once a model is trained with a training dataset, it can be applied on a testing dataset which is independent. In current times, deep learning algorithms further promote the application of machine learning in several field of biology including plant virology. MAIN BODY: Plant viruses have emerged as one of the principal global threats to food security due to their devastating impact on crops and vegetables. The emergence of new viral strains and species help viruses to evade the concurrent preventive methods. According to a survey conducted in 2014, plant viruses are anticipated to cause a global yield loss of more than thirty billion USD per year. In order to design effective, durable and broad-spectrum management protocols, it is very important to understand the mechanistic details of viral pathogenesis. The application of machine learning enables precise diagnosis of plant viral diseases at an early stage. Furthermore, the development of several machine learning-guided bioinformatics platforms has primed plant virologists to understand the host-virus interplay better. In addition, machine learning has tremendous potential in deciphering the pattern of plant virus evolution and emergence as well as in developing viable control options. CONCLUSIONS: Considering a significant progress in the application of machine learning in understanding plant virology, this review highlights an introductory note on machine learning and comprehensively discusses the trends and prospects of machine learning in the diagnosis of viral diseases, understanding host-virus interplay and emergence of plant viruses.

Data reconciliation frameworks for dynamic operation of hybrid renewable energy systems.

Hybrid renewable energy systems (HRES) are a nexus of various renewable energy sources that have been proposed as a solution to circumvent various issues of renewable energy systems when installed in isolation. During the operation of an HRES, efficient demand-side management of energy and real-time power trading requires accurate estimation of real-time variables for each sub-component. Generally, these variables are corrupted by measurement errors. To address this issue, in this study, we present various frameworks for reconciliation strategies that can be used to rectify the inconsistencies in the sensor measurements of HRES. Specifically, in this study, we evaluate the efficacy of various static and dynamic reconciliation strategies such as Regularized Particle filter (RPF), Ensemble Kalman filter (EnKF), and Extended Kalman filter (EKF) of a candidate HRES system with a solar panel, a fuel cell, and an electrolyzer. Proposed frameworks are evaluated using various simulation-based validation studies. To this end, we have considered different operational scenarios, namely, (i) single-rate sampling, (ii) multi-rate sampling, and (iii) sensor outage, to make the study comprehensive. Simulation results indicate that RPF yields the best estimation accuracy for all three operational scenarios with a performance improvement of 75% from EKF and by 50% from EnKF, with only a fractional increment in computational time.

An adaptive, interacting, cluster-based model for predicting the transmission dynamics of COVID-19.

The SARS-CoV-2 driven disease COVID-19 is pandemic with increasing human and monetary costs. COVID-19 has put an unexpected and inordinate degree of pressure on healthcare systems of strong and fragile countries alike. To launch both containment and mitigation measures, each country requires estimates of COVID-19 incidence as such preparedness allows agencies to plan efficient resource allocation and to design control strategies. Here, we have developed a new adaptive, interacting, and cluster-based mathematical model to predict the granular trajectory of COVID-19. We have analyzed incidence data from three currently afflicted countries of Italy, the United States of America, and India. We show that our approach predicts state-wise COVID-19 spread for each country with reasonable accuracy. We show that Rt, as the effective reproduction number, exhibits significant spatial variations in these countries. However, by accounting for the spatial variation of Rt in an adaptive fashion, the predictive model provides estimates of the possible asymptomatic and undetected COVID-19 cases, both of which are key contributors in COVID-19 transmission. We have applied our methodology to make detailed predictions for COVID19 incidences at the district and state level in India. Finally, to make the models available to the public at large, we have developed a web-based dashboard, namely "Predictions and Assessment of Corona Infections and Transmission in India" (PRACRITI, see http://pracriti.iitd.ac.in), which provides the detailed Rt values and a three-week forecast of COVID cases.