Siemens (Italy)

Abstract Anomalies in time series can be a symptom of an incoming failure. Thus, the detection of anomalous data can both reduce maintenance actions and asset unscheduled stops. To tackle this challenge, we exploit the capabilities of convolutional neural networks (CNNs), fed with images obtained from multivariate time series data, transformed by means of two different approaches. Two CNN architectures are investigated, i.e., VGG-19 and SqueezeNet. The performance of both CNNs fed with images is compared to that of (i) a temporal convolutional network (TCN) fed with time series data and (ii) a support vector machine (SVM) model. In this paper, we present the comprehensive framework, which starts from time series transformation, goes through CNN development and ends with anomaly detection. The framework is applied to field data taken during normal operation of ten SGT-800 gas turbines, located in two different regions. The normal data covers 150 days of operation. Spike faults are implanted in two out of the 20 available measured variables, i.e., compressor discharge temperature and compressor discharge pressure, by considering different combinations of maximum fault magnitude and number of implanted spikes in each time series. The results demonstrate that both CNNs fed with images achieve significantly higher classification accuracy than both a TCN model fed with time series data and an SVM model. Moreover, the Markov transition field (MTF) method always proves more robust than Gramian angular summation field (GASF) method, and also allows higher accuracy values, in the range from 0.85 to 0.99.

In partial arc volumetric modulated arc therapy (VMAT) for treating breast cancer, setting up the limiting gantry positions of the treatment machine is a nontrivial yet repetitive and time-consuming task during planning. Templatized solutions exist but may not provide adequate plan quality. We have developed a two-step beam geometry optimization (2SBGO) method to set up in an automated manner the VMAT start/stop gantry angles and avoidance sector (AvS) angle limits for breast treatments. We compare our preliminary results of 2SBGO to manually created plans. In the first step of the method (based on patient geometry), the initial angles are obtained either from a template, from machine-learning (ML) predictions or manually. A search range around the initial positions is specified for each angle. In the second step (refinement using dosimetric criteria), the parameters are optimized using generalized simulated annealing (GSA). As objective function for GSA, we used the optimizer cost. We tested the method for deep inspiration breath hold and free breathing patients for left- and right-sided breast treatments. As ML models, we trained convolutional neural networks to predict the angles (start/stop angles for both the partial arc and the avoidance sector limits). The training set size was up to 86 patients, the validation set size was fixed to six patients and the test set size to six patients. The initial input before preprocessing was in DICOM format (RT plan and structure files and CT images). The rationale for using ML as first step is to learn from data the ways the beam angles are set and evaluate how good the initial ML solution would be for the final plan quality. We showed that for all the test patients, the 2SBGO leads to plans that are of a comparable dosimetric quality compared with manual plans while eliminating the complex and time-consuming beam geometry (BG) set-up step. Additionally, with the optimization function we used in our approach the ipsilateral lung doses in right-sided treatments are reduced compared with plans with manual angle selection. The ML models were shown to be most useful in a clinical workflow when integrated in the full 2SBGO scheme. The ML models themselves, before the second optimization step, predicted the medial angle within accuracy 3.6°±2.6° (mean±SD) and the AvS limiting values within 10.8°±8.3°. The Wilcoxon paired signed-rank test indicated that there were no statistically significant differences between the distributions of ML predictions and manual choices. We can obtain in an automatic fashion clinically acceptable breast treatment plans, dosimetrically comparable with the manual ones, by combining an initial BG set-up with a dosimetric refinement step for the beam angles. The initial ML-based plans are a useful starting point while they need further refining which the dosimetric fine-tuning provides. The work paves the way to the automation of the BG setup that has the potential for considerable savings of planner's time, and a decrease in the variation of the quality of the plans.

ObjectivesThis study aimed to assess whether pharmacokinetic parameters derived from DCE-MRI can stratify Programmed Death-Ligand 1 (PD-L1) expression in NSCLC. The secondary aim was to identify a suitable pharmacokinetic model configuration for anisotropic temporally-spaced DCE-MRI sequences, considering Tofts variants, population-averaged arterial input functions (AIF), and bolus arrival time (BAT) estimation methods.Materials and methodsFrom April 2021 to May 2023, patients with locally advanced non-small cell lung cancer (NSCLC) were prospectively enrolled. Tumors were categorized based on: PD-L1 absence/presence (threshold 1%) and hyperexpression/hypoexpression (threshold 50%). Pharmacokinetic parameters were extracted using several candidate configurations; fit quality was evaluated using coefficient of determination (R²). Mann–Whitney U-test and ROC-AUC were used to assess correlation with PD-L1 for the best-fit configuration.ResultsThirty-eight patients (mean age 68 ± 9 years, 28 men) were included. PD-L1 expression was present in 25 patients (66%) and absent in 13 (34%). PD-L1 was hyperexpressed in 13 (34%) patients and hypoexpressed in 25 (66%). Voxel-wise pharmacokinetic parameters were extracted using the best-fit configuration—extended Tofts model (ETM) with Georgiou AIF and Peak-Gradient (PG) BAT estimation (R2 = 0.79). Ktrans median (0.25 vs. 0.12 min−¹, p = 0.02), Ktrans standard deviation (0.32 vs. 0.23 min−¹, p = 0.01) and Kep median (1.09 vs. 0.59 min−¹, p = 0.02) were significantly higher in PD-L1 < 50% group (ROC-AUC 0.71–0.76).ConclusionDCE-MRI pharmacokinetic parameters could stratify PD-L1 hypo/hyperexpression in NSCLC. The ETM with PG BAT estimation method and Georgiou AIF was the best-performing pharmacokinetic configuration.Key PointsQuestionCould Dynamic Contrast-Enhanced (DCE) MRI offer a safe and non-invasive way to assess Programmed Death-Ligand 1 (PD-L1) expression?FindingsQuantitative DCE-MRI parameters Ktrans (the volume transfer rate) and Kep (the efflux rate constant) show potential for distinguishing PD-L1 hyperexpression from hypoexpression.Clinical relevancePreliminary results suggest that DCE-MRI could be a safe method to stratify PD-L1 hypo/hyperexpression in non-small cell lung cancer, potentially optimizing treatment decisions, given the high cost of immunotherapy.Graphical

Abstract The current energy market requires that gas turbines (GTs) run efficiently and reliably, thus improving their sustainability. To this aim, condition monitoring is fundamental for GT manufacturers to track performance degradation and ensure fault detection in advance. Thanks to the huge amount of data acquired during GT operation, data-driven methodologies can be successfully employed for fault detection and prediction. Therefore, this paper develops an unsupervised data-driven methodology for the prognostics of GT abrupt faults. The methodology comprises i) an autoencoder (AE) neural network aimed at modeling GT operation in the absence of fault symptoms and ii) a threshold-based criterion for the detection of anomalous deviations between actual data and AE predictions. A sensitivity analysis on the anomaly threshold is carried out to identify the optimal set-up of the methodology. The paper considers GT trip as archetypal of an abrupt fault. Since no labeled data are available in the case of trip occurrence, the time point at which trip symptoms arise has to be identified. Furthermore, the paper also investigates which measurable variables are most effective for predicting GT trip, in order to provide useful insights on trip prediction. The methodology is validated by considering a real-world case study comprising data collected from six GTs before trip occurrence. The results prove that the optimally tuned methodology correctly classifies the 80% of data. Another relevant result is that fuel mass flow rate, exhaust gas temperature, and compressor discharge temperature are the most useful measurable variables for the detection of trip symptoms.

Traditionally, the generating units connected at the HV level are the only providers of ancillary services. This is now extended to distributed resources connected at the MV/LV level. However, due to the huge number, it is hard to manage them directly in a centralized Ancillary Services Market. In this paper, an optimization tool to dispatch local flexibility according to regulated local market schemes is proposed. A MILP is formulated to integrate both continuous and discrete flexibility bids and the network constraints. The tool performance is evaluated on the 69 buses distribution grid.

The verification of electric traction motors requires the consideration of different attributes – electromagnetics, thermal, structural integrity and acoustics performance. To achieve an optimal balance of these attributes, the geometry, which is detailed to instill confidence in verification, undergoes modifications that must be considered in subsequent iterations. Therefore, the turnaround time for this multi-attribute iterative process is critical. This paper demonstrates an associative workflow between the geometry and simulation models in the multi-attribute verification of an in-wheel motor. Besides these process efficiency considerations, this paper also considers electric motor NVH analysis where the rotor noise contributions are significant. This is true for axial-flux, transverse-flux and outer-rotor radial-flux machines.

Abstract The current energy scenario requires that gas turbines (GTs) operate at their maximum efficiency and highest reliability. Trip is one of the most disrupting events that reduces GT availability and increases maintenance costs. To tackle the challenge of GT trip prediction, this paper presents a methodology that has the goal of monitoring the early warnings raised during GT operation and trigger an alert to avoid trip occurrence. The methodology makes use of an autoencoder (prediction model) and a three-stage criterion (detection procedure). The autoencoder is first trained to reconstruct safe operation data and subsequently tested on new data collected before trip occurrence. The trip detection criterion checks whether the individually tested data points should be classified as normal or anomalous (first stage), provides a warning if the anomaly score over a given time frame exceeds a threshold (second stage), and, finally, combines consecutive warnings to trigger a trip alert in advance (third stage). The methodology is applied to a real-world case study composed of a collection of trips, of which the causes may be different, gathered from various GTs in operation during several years. Historical observations of gas path measurements taken during three days of GT operation before trip occurrence are employed for the analysis. Once optimally tuned, the methodology provides a trip alert with a reliability equal to 75% at least ten hours in advance before trip occurrence.

When researchers are faced with building machine learning (ML) radiomic models, the first choice they have to make is what model to use. Naturally, the goal is to use the model with the best performance. But what is the best model? It is well known in ML that modern techniques such as gradient boosting and deep learning have better capacity than traditional models to solve complex problems in high dimensions. Despite this, most radiomics researchers still do not focus on these models in their research. As access to high-quality and large data sets increase, these high-capacity ML models may become even more relevant. In this article, we use a large dataset of 949 prostate cancer patients to compare the performance of a few of the most promising ML models for tabular data: gradient-boosted decision trees (GBDTs), multilayer perceptions, convolutional neural networks, and transformers. To this end, we predict nine different prostate cancer pathology outcomes of clinical interest. Our goal is to give a rough overview of how these models compare against one another in a typical radiomics setting. We also investigate if multitask learning improves the performance of these models when multiple targets are available. Our results suggest that GBDTs perform well across all targets, and that multitask learning does not provide a consistent improvement.

Breast cancer is still one of the most common cancers in women, and it is also the leading cause of mortality among women. Breast cancer detection has been improved using a variety of Machine Learning and Deep Learning techniques. Federated Learning is a distributed learning technique that allows models to be trained on a vast set of decentralized data. In this paper we present a new Multiple Criteria Optimization approach for Federated Learning. We use Goal Programming in particular to address the Federated Learning problem's diverse set of objectives. Numerical experiments also demonstrate the efficacy of this strategy in terms of performance.

The application of Machine Learning in Human Resources is becoming increasingly common. The Human Resources industry relies on Machine Learning to automate processes, improve efficiency, and enhance decision making. However, using Machine Learning in Human Resources is not without its challenges. One of these challenges is the lack of interpretability. Interpretability is essential in Human Resources because many of the decisions made in the field have a direct impact on people's lives. As such, these decisions must be made transparently and understandably. In this paper, we apply a model-agnostic technique for providing post-hoc explanations, known as Anchors, to a real-world dataset from the Human Resources industry. The dataset was used to create a predictive pipeline for employee attrition. The result is a Human Capital Management system capable of explaining the reasons behind employee attrition, allowing the Human Resource professionals to enact retention policies promptly. Moreover, the results suggest that Anchors can be used to create a prescriptive pipeline that can be used to explain the reasons behind every single decision to leave the company, as they are easily interpretable by a non-expert. This system has the advantage of allowing the Decision Maker to act in a prescriptive way and retain valuable resources.