Deep Ensemble Approach Research Articles

Predicting Pathological Characteristics of HER2-Positive Breast Cancer from Ultrasound Images: a Deep Ensemble Approach.

The objective is to evaluate the feasibility of utilizing ultrasound images in identifying critical prognostic biomarkers for HER2-positive breast cancer (HER2 + BC). This study enrolled 512 female patients diagnosed with HER2-positive breast cancer through pathological validation at our institution from January 2016 to December 2021. Five distinct deep convolutional neural networks (DCNNs) and a deep ensemble (DE) approach were trained to classify axillary lymph node involvement (ALNM), lymphovascular invasion (LVI), and histological grade (HG). The efficacy of the models was evaluated based on accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), receiver operating characteristic (ROC) curves, areas under the ROC curve (AUCs), and heat maps. DeLong test was applied to compare differences in AUC among different models. The deep ensemble approach, as the most effective model, demonstrated AUCs and accuracy of 0.869 (95% CI: 0.802-0.936) and 69.7% in LVI, 0.973 (95% CI: 0.949-0.998) and 73.8% in HG, thus providing superior classification performance in the context of imbalanced data (p < 0.05 by the DeLong test). On ALNM, AUC and accuracy were 0.780 (95% CI: 0.688-0.873) and 77.5%, which were comparable to other single models. The pretreatment US-based DE model could hold promise as a clinical guidance for predicting pathological characteristics of patients with HER2-positive breast cancer, thereby providing benefit of facilitating timely adjustments in treatment strategies.

A Surrogate Model for Studying Solar Energetic Particle Transport and the Seed Population

AbstractThe high energy particles originating from the Sun, known as solar energetic particles (SEPs), contribute significantly to the space radiation environment, posing serious threats to astronauts and scientific instruments on board spacecraft. The mechanism that accelerates the SEPs to the observed energy ranges, their transport in the inner heliosphere, and the influence of suprathermal seed particle spectrum are open questions in heliophysics. Accurate predictions of the occurrences of SEP events well in advance are necessary to mitigate their adverse effects but prediction based on first principle models still remains a challenge. In this scenario, adopting a machine learning approach to SEP modeling and prediction is desirable. However, the lack of a balanced database of SEP events restrains this approach. We addressed this limitation by generating large data sets of synthetic SEP events sampled from the physics‐based model, Energetic Particle Radiation Environment Module (EPREM). Using this data, we developed neural networks‐based surrogate models to study the seed population parameter space. Our models, EPREM‐S, run thousands to millions of times faster (depending on computer hardware), making simulation‐based inference workflows practicable in SEP studies while providing predictive uncertainty estimates using a deep ensemble approach.

An empirical analysis of deep ensemble approach on COVID-19 and tuberculosis X-ray images

An empirical analysis of deep ensemble approach on COVID-19 and tuberculosis X-ray images

Deep ensemble with Neural Networks to model power curve uncertainty

Renewable energy is an essential driver towards tackling the climate crisis. Given the increasing need to ensure sustainability and reduce greenhouse gas emissions, wind energy has become one of the most relevant areas to consider in today’s society. However, it appears challenging to rely on this type of energy because of the uncertain nature of the wind. It is essential to use monitoring tools that can accurately capture the uncertainty of the underlying environmental processes and estimate the generated power to ensure the reliability of wind-based power systems. One of these monitoring tools is the power curve, which captures the relationship between wind speed and expected power production. Many regression techniques have previously been used to predict this power production, including Artificial Neural Networks (ANN). The problem is that they do not capture the uncertainty of the predicted power production. In contrast with the power curve models already implemented when using ANNs, we propose a novel approach based on deep ensembles to predict the mean and model variance on the expected power production. We used three months of Supervisory Control and Data Acquisition data from more than 40 turbines to set up the experiments. The results show that the mean and standard deviation fit the data well. It also appears that the expected mean curve goes through the center of the cloud of original data points, while the standard deviation accurately captures the spread of the data. We conclude that that the deep ensemble approach provides an accurate prediction mechanism to estimate the power production from the wind speed while also offering an uncertainty measure. Therefore, the deep ensemble approach allows us to construct a power curve with uncertainty information.

Ensemble voting decreases false positives in AI second-observer reads for detecting colorectal cancer.

141 Background: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths, and survival can be improved if early, suspect imaging features on CT of the abdomen and pelvis (CTAP) can be routinely identified. At present, up to 40% of these features are undiagnosed on routine CTAP, but this can be improved with a second observer. In this study, we developed a deep ensemble learning method for detecting CRC on CTAP to determine if increasing agreement between ensemble models can decrease the false positives detected by artificial intelligence (AI) second-observer. Methods: 2D U-Net convolutional neural network (CNN) containing 31 million trainable parameters was trained with 58 CRC CT images from Banner MD Anderson (AZ) and MD Anderson Cancer Center (TX) (51 used for training and 7 for validation) and 59 normal CT scans from Banner MD Anderson Cancer Center. 20 of the 25 CRC cases from public domain data (The Cancer Genome Atlas) were used to evaluate the performance of the models. The CRC was segmented using ITK-SNAP open-source software (v. 3.8). To apply the deep ensemble approach, five CNN models were trained independently with random initialization using the same U-Net architect and the same training data. Given a testing CT scan, each of the five trained CNN models was applied to produce tumor segmentation for the testing CT scan. The tumor segmentation results produced by the trained CNN models were then fused using a simple majority voting rule to produce consensus tumor segmentation results. The segmentation was analyzed by the percentage of correct detection, the number of false positives per case, and the Dice similarity coefficient (DSC). If parts of the CRC were flagged by AI, then it was considered correct. A detection was considered false positive if the marked lesion did not overlap with any CRC; contiguous false positives across different slices of CT image were considered a single false positive. DSC measures the quality of the segmentation by measuring the overlap between the ground-truth and AI detected lesion. Results: Our results showed that increasing the agreement between the 5 models dramatically decreases the number of false positives per CT at the expense of slight decrease in accuracy and DSC. This is described in the table. Conclusions: Our results show that AI-based second observer can potentially detect CRC on routine CTAP. Although the initial result yields high false positives per case, ensemble voting is an effective method for decreasing the false positives with a slight decrease in accuracy. This technique can be further improved for eventual clinical application.[Table: see text]

On deep ensemble CNN–SAE based novel agro-market price forecasting

The prices of agro-commodities are highly volatile. Hence it is a challenge to the farmers to ensure fair and remunerative prices of these commodities. As a result, there is a need for prediction of agro market price appropriately. The closing price prediction of one soft commodity product, Cotton 29 mm and one agro-commodity product, Guar gum are chosen. The existing reported methods exhibit poor prediction performance. To alleviate this problem, the current investigation is undertaken for better prediction of closing prices. The deep ensemble approach using convolutional neural network (CNN) and stacked autoencoder (SAE) is employed to improve the prediction performance. For the ensemble strategy, the weights are optimized using three bio-inspired techniques such as genetic algorithm (GA), particle swarm optimization (PSO) and spider monkey optimization (SMO). Eighteen attributes relating to the closing price of each products are considered as input to the proposed models. The simulation based experimental results demonstrate the following contribution of the paper. Firstly, it is observed that CNN outperforms the SAE model in terms of short range prediction and vice versa for long range prediction. Secondly, the prediction performance of all the three ensemble models has been determined. Thirdly, out of three ensemble models, ensemble-SMO (ESMO) shows the best prediction performance in terms of mean square error and coefficient of multiple determination (R2). It is then followed by ensemble-PSO and ensemble-GA respectively. The performance of proposed best ESMO is compared with the Grey wolf optimization based multiquadratic kernel KELM model (GWO-KELM) and it is observed that the proposed ESMO outperforms the GWO-KELM model.