Coronary artery disease (CAD) remains the leading cause of death globally and invasive coronary angiography (ICA) is considered the gold standard of anatomical imaging evaluation when CAD is suspected. However, risk evaluation based on ICA has several limitations, such as visual assessment of stenosis severity, which has significant interobserver variability. This motivates to development of a lesion classification system that can support specialists in their clinical procedures. Although deep learning classification methods are well‐developed in other areas of medical imaging, ICA image classification is still at an early stage. One of the most important reasons is the lack of available and high‐quality open‐access datasets. In this paper, we reported a new annotated ICA images dataset, CADICA, to provide the research community with a comprehensive and rigorous dataset of coronary angiography consisting of a set of acquired patient videos and associated disease‐related metadata. This dataset can be used by clinicians to train their skills in angiographic assessment of CAD severity, by computer scientists to create computer‐aided diagnostic systems to help in such assessment, and to validate existing methods for CAD detection. In addition, baseline classification methods are proposed and analysed, validating the functionality of CADICA with deep learning‐based methods and giving the scientific community a starting point to improve CAD detection.

Invasive Coronary Angiography (ICA) images are considered the gold standard for assessing the state of the coronary arteries. Deep learning classification methods are widely used and well-developed in different areas where medical imaging evaluation has an essential impact due to the development of computer-aided diagnosis systems that can support physicians in their clinical procedures. In this paper, a new performance analysis of deep learning methods for binary ICA classification with different lesion degrees is reported. To reach this goal, an annotated dataset of ICA images containing the ground truth, the location of lesions, and seven possible severity degrees ranging between 0% and 100% was employed. The ICA images were divided into “lesion” or “non-lesion” patches. We aim to study how binary classification performance is affected by the different lesion degrees considered in the positive class. Therefore, five Convolutional Neural Network architectures – DenseNet-201, MobileNet-V2, NasNet-Mobile, ResNet-18, and ResNet-50 – were trained with different input images where different lesion degree ranges were gradually incorporated until considering the seven lesion degrees. Besides, four types of experiments with and without data augmentation were designed, whose F-measure and Area Under Curve (AUC) were computed. Reported results achieved an F-measure and AUC of 92.7% and 98.1%, respectively. However, lesion classification is highly affected by the degree of the lesion intended to be classified, with 15% less accuracy when < 99% lesion patches are present.

The study’s purpose was to assess the reliability of the LIS2DH12 in two different positions, using the commercial sensor Actigraph GT9X as a reference instrument. Five participants completed two gait tests on a treadmill. Firstly, both sensors were worn on the wrist and around the thigh. Each test consisted of a 1 min walk for participants to become accustomed to the treadmill, followed by a 2 min trial at ten pre-set speeds. Data from both sensors were collected in real-time. Intraclass correlation coefficient (ICC) was used to evaluate the equality of characteristics obtained by both sensors: maximum peaks, minimum peaks, and the mean of the complete signal (sequence of acceleration values along the time) by each axis and speed were extracted to evaluate the equality of characteristics obtained with LIS2DH12 compared to Actigraph. Intraclass correlation coefficient (ICC) was extracted, and a standard deviation of the mean was obtained from the data. Our results show that LIS2DH12 measurements present more reliability than Actigraph GT9X, ICC > 0.8 at three axes. This study concludes that LIS2DH12 is as reliable and accurate as Actigraph GT9X Link and, therefore, would be a suitable tool for future kinematic studies.

Coronary artery disease (CAD) remains the leading cause of mortality, creating an urgent need for reproducible, image-based decision support. Although YOLOv8-based detectors underpin much of today's state-of-the-art stenosis detection, their accuracy is sensitive to dozens of interacting hyperparameters. We therefore perform a systematic study of hyperparameter optimizers for YOLO-style models, pairing YOLOv8 and its Double Coordinate Attention (DCA) variant with three model-based engines: Covariance-Matrix-Adaptation Evolution Strategy (CMA-ES), Tree-structured Parzen Estimator (TPE), and a Gaussian-process sampler; and contrasting them with Random Search and the mutation-only routine that serves as the default optimizer in the ultralytics package. Optimization targets binary detection and was benchmarked using the CADICA (full sequences) and ARCADE (single key-frames) datasets, maximizing the F1-Score under stratified three-fold cross-validation within a fixed compute budget. Model-based methods consistently lift the F1-speed Pareto frontier: CMA-ES attains 0.355±0.079 on v8l, while Bayesian strategies top the medium and small backbones with 0.346±0.048 (v8m) and 0.304±0.054 (v8s). All surpass the default optimizer and yield more lesion-centric EigenCAM saliency, confirming the value of adaptive probabilistic search for tuning high-dimensional YOLO-based CAD pipelines. The complete code-base is open-source and released at https://github.com/MarioPasc/Coronary_Angiography_Detection.