Background— An open-irrigated radiofrequency (RF) ablation catheter was developed to measure contact force (CF). Three optical fibers measure microdeformation of the catheter tip. The purpose of this study was to (1) validate the accuracy of CF sensor (CFS) (bench test); and (2) determine the relationship between CF and tissue temperatures, lesion size, steam pop, and thrombus during RF ablation using a canine thigh muscle preparation. Methods and Results— CFS measurements (total 1409) from 2 catheters in 3 angles (perpendicular, parallel, and 45°) were compared with a certified balance (range, 0 to 50 g). CFS measurements correlated highly (R2≥0.988; mean error, ≤1.0 g). In 10 anesthetized dogs, a skin cradle over the thigh muscle was superfused with heparinized blood at 37°C. A 7F catheter with 3.5-mm saline-irrigated electrode and CFS (Endosense) was held perpendicular to the muscle at CF of 2, 10, 20, 30, and 40 g. RF was delivered (n=100) for 60 seconds at 30 or 50 W (irrigation 17 or 30 mL/min). Tissue temperature (3 and 7 mm depths), lesion size, thrombus, and steam pop increased significantly with increasing CF at each RF power. Lesion size was greater with applications of lower power (30 W) and greater CF (30 to 40 g) than at high power (50 W) with lower CF (2 to 10 g). Conclusions— This novel ablation catheter, which accurately measures CF, confirmed CF is a major determinant of RF lesion size. Steam pop and thrombus incidence also increases with CF. CFS in an open-irrigated ablation catheter that may optimize the selection of RF power and application time to maximize lesion formation and reduce the risk of steam pop and thrombus

Catheter contact force has an important impact on both ablation lesion size and the incidence of pops.

Lesion size correlates linearly with measured contact FTI. Constant contact produces the largest and intermittent contact the smallest lesions despite constant RF power and identical peak contact forces.

Artificial intelligence (AI) models are playing an increasing role in biomedical research and healthcare services. This review focuses on challenges points to be clarified about how to develop AI applications as clinical decision support systems in the real-world context. Methods: A narrative review has been performed including a critical assessment of articles published between 1989 and 2021 that guided challenging sections. Results: We first illustrate the architectural characteristics of machine learning (ML)/radiomics and deep learning (DL) approaches. For ML/radiomics, the phases of feature selection and of training, validation, and testing are described. DL models are presented as multi-layered artificial/convolutional neural networks, allowing us to directly process images. The data curation section includes technical steps such as image labelling, image annotation (with segmentation as a crucial step in radiomics), data harmonization (enabling compensation for differences in imaging protocols that typically generate noise in non-AI imaging studies) and federated learning. Thereafter, we dedicate specific sections to: sample size calculation, considering multiple testing in AI approaches; procedures for data augmentation to work with limited and unbalanced datasets; and the interpretability of AI models (the so-called black box issue). Pros and cons for choosing ML versus DL to implement AI applications to medical imaging are finally presented in a synoptic way. Conclusions: Biomedicine and healthcare systems are one of the most important fields for AI applications and medical imaging is probably the most suitable and promising domain. Clarification of specific challenging points facilitates the development of such systems and their translation to clinical practice.