This study evaluates the performance of deep learning models in the prediction of the end time of procedures performed in the cardiac catheterization laboratory (cath lab). We employed only the clinical phases derived from video analysis as input to the algorithms. Our results show that InceptionTime and LSTM-FCN yielded the most accurate predictions. InceptionTime achieves Mean Absolute Error (MAE) values below 5 min and Symmetric Mean Absolute Percentage Error (SMAPE) under 6% at 60-s sampling intervals. In contrast, LSTM with attention mechanism and standard LSTM models have higher error rates, indicating challenges in handling both long-term and short-term dependencies. CNN-based models, especially InceptionTime, excel at feature extraction across different scales, making them effective for time-series predictions. We also analyzed training and testing times. CNN models, despite higher computational costs, significantly reduce prediction errors. The Transformer model has the fastest inference time, making it ideal for real-time applications. An ensemble model derived by averaging the two best performing algorithms reported low MAE and SMAPE, although needing longer training. Future research should validate these findings across different procedural contexts and explore ways to optimize training times without losing accuracy. Integrating these models into clinical scheduling systems could improve efficiency in cath labs. Our research demonstrates that the models we implemented can form the basis of an automated tool, which predicts the optimal time to call the next patient with an average error of approximately 30 s. These findings show the effectiveness of deep learning models, especially CNN-based architectures, in accurately predicting procedure end times.

Background: The integration of medical technology in the operating room has revolutionized surgical workflows and team dynamics. However, this progress coincides with a critical global shortage of nurses and a high turnover rate within the existing nursing workforce, impacting patient care quality, nurses' well-being, and hospital finances Aim: This study investigates the impact of technological complexity on the workload and job satisfaction of intra-operative nurses, focusing on open surgery, minimally invasive surgery, and robotic-assisted surgery within the gynecology department of a Dutch academic hospital. Method: The study design follows a mixed-methods approach, combining qualitative and quantitative methods to assess nursing experiences across three surgical modalities. Specifically, we conducted 5 interviews, distributed 28 validated questionnaires, performed automated video analysis on 35 recorded surgeries, and analyzed hospital datasets encompassing 411 cases. Data collection took place in 2022 and 2023. Results: Findings show that intra-operative nurses experience varying levels of workload and job satisfaction depending on the level of technology. Open procedures showed the highest job satisfaction, characterized by continuous engagement and manageable workloads. Minimally invasive surgery procedures, while less physically demanding, were associated with reduced involvement and lower satisfaction. Robotic-assisted procedures presented the most significant challenges, with increased workload, reduced involvement, and heightened stress stemming from surgery preparation, technological complexity, and altered team dynamics. Conclusions: Advancements in medical technology improve outcomes and efficiency but often neglect their impact on intra-operative nurses. Communication issues, equipment challenges, and limited technical training contribute to burnout and turnover. This study underscores the need for supportive operating room environments that prioritize nurses’ well-being. By examining the link between technology, workload, and satisfaction, it offers strategies to retain and empower nursing staff. It also shows how automated video analysis can objectively assess nursing roles, highlighting the importance of balancing technology with human-centered care in the operating room.

Deep learning-based object detectors, while offering exceptional performance, are data-dependent and can suffer from generalization issues. In this work, we investigated deep neural networks for detecting people and medical instruments for the vision-based workflow analysis system inside Catheterization Laboratories (Cath Labs). The central problem explored in this paper is the fact that the performance of the detector can degrade drastically if it is trained and tested on data from different Cath Labs. Our research aimed to investigate the underlying causes of this specific performance degradation and find solutions to mitigate this issue. We employed the YOLOv8 object detector and created datasets from clinical procedures recorded at Reinier de Graaf Hospital (RdGG) and Philips Best Campus, supplemented with publicly accessible images. Through a series of experiments complemented by data visualization, we discovered that the performance degradation primarily stems from data distribution shifts in the feature space. Notably, the object detector trained on non-sensitive online images can generalize to unseen Cath Labs, outperforming the model trained on a procedure recording from a different Cath Lab. The detector trained on the online images achieved an mAP@0.5 of 0.517 on the RdGG dataset. Furthermore, by switching to the most suitable camera for each object in the Cath Lab, the multi-camera system can further improve the detection performance significantly. An aggregated L-camera mAP@0.5 of 0.679 is achieved for single-object classes on the RdGG dataset.