The analysis of electrocardiogram (ECG) signals is profoundly affected by the presence of electromyographic (EMG) noise, which can lead to substantial misinterpretations in healthcare applications. To address this challenge, we present ECGDnet, an innovative architecture based on Transformer technology, specifically engineered to denoise multi-channel ECG signals. By leveraging multi-head self-attention mechanisms, positional embeddings, and an advanced sequence-to-sequence processing architecture, ECGDnet effectively captures both local and global temporal dependencies inherent in cardiac signals. Experimental validation on real-world datasets demonstrates ECGDnet’s remarkable efficacy in noise suppression, achieving a Signal-to-Noise Ratio (SNR) of 19.83, a Normalized Mean Squared Error (NMSE) of 0.9842, a Reconstruction Error (RE) of 0.0158, and a Pearson Correlation Coefficient (PCC) of 0.9924. These results represent significant improvements from traditional deep learning approaches while maintaining complex signal morphology and effectively mitigating noise interference.

The analysis of electrocardiogram (ECG) signals is profoundly affected by the presence of electromyographic (EMG) noise, which can lead to substantial misinterpretations in healthcare applications. To address this challenge, we present ECGDnet, an innovative architecture based on Transformer technology, specifically engineered to denoise multi-channel ECG signals. By leveraging multi-head self-attention mechanisms, positional embeddings, and an advanced sequence-to-sequence processing architecture, ECGDnet effectively captures both local and global temporal dependencies inherent in cardiac signals. Experimental validation on real-world datasets demonstrates ECGDnet's remarkable efficacy in noise suppression, achieving a Signal-to-Noise Ratio (SNR) of 19.83, a Normalized Mean Squared Error (NMSE) of 0.9842, a Reconstruction Error (RE) of 0.0158, and a Pearson Correlation Coefficient (PCC) of 0.9924. These results represent significant improvements over traditional deep learning approaches, while maintaining complex signal morphology and effectively mitigating noise interference.

Predictive maintenance (PdM) has emerged as a critical enabler for improving reliability, operational efficiency, and regulatory compliance in pharmaceutical manufacturing. Despite its proven effectiveness in several industrial sectors, PdM adoption within the pharmaceutical industry remains comparatively limited due to sector-specific technical, regulatory, and organizational constraints. This paper presents a structured technical analysis of recent academic and industrial works addressing PdM implementation in pharmaceutical manufacturing systems. The analysis examines applied AI and machine learning techniques, sensor and data acquisition strategies, implementation maturity, and regulatory considerations relevant to highly regulated environments. The findings indicate that, while PdM solutions can significantly improve equipment availability and reduce maintenance-related costs, major barriers persist, including limited failure data, validation and re-validation requirements, and organizational resistance to data-driven maintenance practices. Based on this analysis, the paper argues that hybrid approaches combining physics-based models (PBMs) with data-driven methods and explainable artificial intelligence (XAI), supported by digital twins and robust data governance frameworks, represent a practical and regulation-aware pathway for the broader adoption of predictive maintenance in pharmaceutical manufacturing

Heart murmurs, resulting from turbulent blood flow within the cardiac structure, represent some of the initial acoustic manifestations of potential underlying cardiovascular anomalies, such as arrhythmias. This research presents a deep learning framework aimed at the early detection of potential cardiac anomalies through the analysis of murmur patterns in phonocardiogram (PCG) signals. Our methodology employs a spectro-temporal feature fusion technique that integrates Mel spectrograms, Mel Frequency Cepstral Coefficients (MFCCs), Root Mean Square (RMS) energy, and Power Spectral Density (PSD) representations. The features are derived from segmented 5-second phonocardiogram (PCG) windows and subsequently input into a two-dimensional convolutional neural network (CNN) for the purpose of classification. In order to mitigate class imbalance and enhance generalization, We employ data augmentation techniques, including pitch moving and noise injection. The model under consideration has undergone training and evaluation utilizing a carefully selected subset of the CirCor DigiScope dataset. The experimental findings indicate a robust performance, with a classification accuracy recorded at 92.40% and a cross-entropy loss measured at 0.2242. The results indicate that an analysis of PCG signals informed by murmurs may function as an effective non-invasive method for the early screening of conditions that may include arrhythmias, particularly in clinical environments with limited resources.

In an era where reliability, efficiency, and sustainability have become essential objectives across all sectors, the healthcare domain faces even greater challenges due to its direct impact on human life. Medical devices represent the technological backbone of modern healthcare systems, supporting activities ranging from diagnosis and monitoring to treatment and therapeutic intervention. Ensuring their continuous availability, safety, and performance is therefore a critical priority. Traditional maintenance strategies, however, often rely on reactive actions that may lead to unexpected downtime, increased operational costs, and potential risks to patient safety. To address these limitations, predictive maintenance has emerged as a promising and innovative solution. This study focuses on the development of a predictive maintenance framework for the RF subsystem of a medical cyclotron, identified as the most critical subsystem through a Failure Modes, Effects, and Criticality Analysis (FMECA). The proposed framework combines real operational log data with synthetic signals generated from a MATLAB Simulink (version 2023) physics-based model of the RF subsystem, and applies a Random Forest classifier trained on a degree-3 polynomial feature expansion of the three raw RF signals. Using a group-separated evaluation protocol that partitions contiguous acquisition sessions into independent fitting, calibration, validation, and test sets, the model achieves perfect fault-state classification (Recall = $100\%$, zero false alarms) on 58,291 held-out real operational samples containing 64 fault-state events, with the decision threshold selected exclusively on an independent validation partition. A complementary short-horizon anticipation task, using only causally available past information, evaluates the model’s ability to flag an impending threshold crossing up to 60 s in advance. The proposed approach provides a scalable and interpretable foundation for real-time predictive maintenance deployment in nuclear medicine facilities, while identifying the acquisition of independently verified maintenance records as a necessary next step toward validating true early failure prediction.

Postoperative atrial fibrillation (POAF) is a leading cause of stroke and mortality after cardiac surgery, motivating continuous long-term ECG monitoring.This paper presents Cardiac-Net, a wearable ECG system (ESP32-S3/AD8232 BLE device with a Flutter mobile gateway; 128 Hz nominal/85 Hz effective BLE sampling rate, 12.6 ms mean latency, zero over-the-air loss of transmitted packets, 3.11 mJ per transmitted sample) combined with a hybrid 1D CNN–Transformer classifier and a proposed cloud-assisted monitoring architecture; the cloud inference and monitoring-dashboard layer is a design-level proposal only and has not been implemented. Bench signal fidelity, validated against a BIOPAC MP36 reference using a Fluke ProSim 8 simulator across five noise conditions, yields a Pearson Correlation Coefficient of 0.9222 and BPM errors below 0.2% under clean conditions, within the 5% heart-rate-error threshold specified by IEC 60601-2-47. The classifier was trained and evaluated on the Long-Term Atrial Fibrillation Database (84 recordings) using a record-wise (patient-wise) data split with verified zero overlap between training, validation, and test recordings; 5-fold record-wise cross-validation spanning all 84 recordings yields AUC $=0.953\pm 0.016$, F1 $=0.894\pm 0.030$, precision $=0.880\pm 0.051$, and recall $=0.915\pm 0.073$; re-evaluating at the device’s actual 85 Hz effective throughput confirms comparable discrimination (AUC $=0.946\pm 0.010$) alongside a real increase in false-alarm burden. A clinical evaluation protocol beyond segment-level accuracy shows that most AF episodes are detected (episode-level sensitivity $0.839\pm 0.222$ by fold; per-patient median 1.000, IQR [0.925, 1.000]), but this is not uniform across patients: 7.2% (6/83) fall below chance-level detection. Combined with a false-alarm burden of $86\pm 38$ alerts per patient-day and a prevalence-adjusted positive predictive value of only 0.28–0.44 at 5–10% assumed AF prevalence, alarm calibration and patient-level reliability both remain barriers to unsupervised clinical deployment. Zero-shot evaluation on the independent MIT-BIH Atrial Fibrillation Database, with no fine-tuning, yields an ensemble AUC of 0.834, indicating generalization beyond the training distribution alongside a measurable domain gap. Post-hoc Grad-CAM and Integrated Gradients attributions are consistent with physiologically grounded decision-making, concentrating on RR-interval irregularity and QRS morphology. Cardiac-Net thus contributes a validated wearable acquisition layer and a rigorously, record-wise cross-validated detection model, together with an honest characterization of the false-alarm and calibration challenges that remain before postoperative clinical deployment.