A novel soft sensor calibration method is proposed for minimally invasive surgery, based on our developed gelatin-graphite sensor with high compliance and adaptability. This approach uses convolutional deep learning that accounts for a sensor’s non-linear behavior and reduces noise amplification. This technique offers a smaller minimum detectable force than other approaches and is particularly useful in sensitive surgical scenarios. The sensor’s performance is characterized by its fine resolution (≤1mN) and accurate force estimation, especially for forces below 400 mN of amplitude. The best calibration (Morse) scheme provides high performance, with a Mean Absolute Error of ≤7.9 mN. This work was validated through comparison among other representative studies and offered a path toward future directions for optimizing and implementing soft robotic sensors in minimally invasive surgeries. The application of this sensor can revolutionize surgical procedures and capitalize on the benefits of soft robotics, potentially enhancing precision and reducing trauma in surgeries.

Human-robot interactions are becoming more and more prevalent in various aspects of life, enhancing humans’ mobility, accessibility and health. However, safety measures need to be addressed when applying robotic-generated forces that put human users at risk. One way to improve safety and performance in robotic tasks is to include physiological information, such as damping properties of the arm, in the control system to help regulate the energy that is delivered to the user. In this work, we estimated the energy absorbing capabilities of the human arm, based on the metric Excess of Passivity (EOP), originally defined in [1]. We used data from healthy subjects to generate models that fit different levels of safety and stability. Variability in subjects’ EOP was a major finding in this study. For demanding applications such as robotic rehabilitation therapy, we suggest using a linear model with two EOP points. Such points are the mean values of EOP estimations at relaxed and rigid levels of hand-grasp forces. Two standard deviations were subtracted from each EOP point to consider the variability due to the neuromuscular changes in the human arm.

Visual field (VF) testing is crucial for the management of glaucoma. However, the process is often hindered by technician shortages and reliability issues. In this study, we leveraged machine learning to predict pattern standard deviation (PSD) using clinical inputs. This machine learning retrospective study used publicly accessible data from 743 eyes (541 glaucoma and 202 non-glaucoma controls). An automated neural network (ANN) model was trained using seven clinical input features: mean retinal nerve fiber layer (RNFL), IOP, patient age, CCT, glaucoma diagnosis, study protocol, and laterality. The ANN demonstrated efficient training across 1000 epochs, with consistent error reduction in training and test sets. Mean RMSEs were 1.67 ± 0.05 for training, and 2.27 ± 0.27 for testing. The r was 0.89 ± 0.01 for training, and 0.81 ± 0.04 for testing, indicating strong predictive accuracy with minimal overfitting. The LOFO analysis revealed that the primary contributors to PSD prediction were RNFL, CCT, IOP, glaucoma status, study protocol, and age, listed in order of significance. Our neural network successfully predicted PSD from RNFL and clinical data with strong performance metrics, in addition to demonstrating construct validity. This work demonstrates that neural networks hold the potential to predict or even generate VF estimations based solely on RNFL and clinical inputs.

Ophthalmic surgery requires micrometer-level precision due to the eye’s delicate anatomy, yet manual limitations and restricted 3D visualization make absolute accuracy challenging, driving interest in robotic and Artificial Intelligence technologies to enhance safety and precision. This is a narrative review of experimental and published studies on PubMed and Open Evidence to review the current advances, challenges, and translational potential of robotic-assisted needle insertion in corneal surgery. Topics include robotic corneal surgery platforms such as the da Vinci and custom microsurgical robots, telemanipulation, intraoperative optical coherence tomography (iOCT), and reinforcement learning applications. Recent advancements in the field have demonstrated enhanced needle insertion precision, tremor elimination, and improved visualization of needle trajectory in corneal procedures, including corneal lacerations, pterygium repairs and penetrating keratoplasties (PKs). Nonetheless, significant limitations in the state of the art persist, particularly concerning the integration of robotic systems into clinical practice in in vivo settings. Our results indicate that current studies are mostly conducted in an ex vivo setting, which introduces inherent biases and reduces the generalizability of findings to clinical practice. Additionally, the majority of these studies involve small sample sizes, limiting statistical power and the ability to draw robust conclusions. Together, these limitations highlight the need for larger, well-designed in vivo studies to validate and expand upon existing findings. This review bridges experimental innovation and clinical application, highlighting strategies to overcome current barriers in robotic corneal surgery.