Objectives: To assess a new application of artificial intelligence for real-time detection of laryngeal squamous cell carcinoma (LSCC) in both white light (WL) and narrow-band imaging (NBI) videolaryngoscopies based on the You-Only-Look-Once (YOLO) deep learning convolutional neural network (CNN).Study Design: Experimental study with retrospective data. Methods: Recorded videos of LSCC were retrospectively collected from in-office transnasal videoendoscopies and intraoperative rigid endoscopies. LSCC videoframes were extracted for training, validation, and testing of various YOLO models. Different techniques were used to enhance the image analysis: contrast limited adaptive histogram equalization, data augmentation techniques, and test time augmentation (TTA). The best-performing model was used to assess the automatic detection of LSCC in six videolaryngoscopies.Results: Two hundred and nineteen patients were retrospectively enrolled. A total of 624 LSCC videoframes were extracted. The YOLO models were trained after random distribution of images into a training set (82.6%), validation set (8.2%), and testing set (9.2%). Among the various models, the ensemble algorithm (YOLOv5s with YOLOv5m-TTA) achieved the best LSCC detection results, with performance metrics in par with the results reported by other state-of-the-art detection models: 0.66 Precision (positive predicted value), 0.62 Recall (sensitivity), and 0.63 mean Average Precision at 0.5 intersection over union. Tests on the six videolaryngoscopies demonstrated an average computation time per videoframe of 0.026 seconds. Three demonstration videos are provided.Conclusion: This study identified a suitable CNN model for LSCC detection in WL and NBI videolaryngoscopies. Detection performances are highly promising. The limited complexity and quick computational times for LSCC detection make this model ideal for real-time processing.

A medical robotic system for teleoperated laser microsurgery based on a concept we have called "virtual scalpel" is presented in this paper. This system allows surgeries to be safely and precisely performed using a graphics pen directly over a live video from the surgical site. This is shown to eliminate hand-eye coordination problems that affect other microsurgery systems and to make full use of the operator's manual dexterity without requiring extra training. The implementation of this system, which is based on a tablet PC and a new motorized laser micromanipulator offering 1µm aiming accuracy within the traditional line-of-sight 2D operative space, is fully described. This includes details on the system's hardware and software structures and on its calibration process, which is essential for guaranteeing precise matching between a point touched on the live video and the laser aiming point at the surgical site. Together, the new hardware and software structures make both the calibration parameters and the laser aiming accuracy (on any plane orthogonal to the imaging axis) independent of the target distance and of its motions. Automatic laser control based on new intraoperative planning software and safety improvements based on virtual features are also described in this paper, which concludes by presenting results from sets of path following evaluation experiments conducted with 10 different subjects. These demonstrate an error reduction of almost 50% when using the virtual scalpel system versus the traditional laser microsurgery setup, and an 80% error reduction when using the automatic laser control routines, evidencing great improvements in terms of precision and controllability, and suggesting that the technological advances presented herein will lead to a significantly enhanced capacity for treating a variety of internal human pathologies.

Twin-to-Twin Transfusion Syndrome (TTTS) is commonly treated with minimally invasive laser surgery in fetoscopy. The inter-foetal membrane is used as a reference to find abnormal anastomoses. Membrane identification is a challenging task due to small field of view of the camera, presence of amniotic liquid, foetus movement, illumination changes and noise. This paper aims at providing automatic and fast membrane segmentation in fetoscopic images. We implemented an adversarial network consisting of two Fully-Convolutional Neural Networks (FCNNs). The former (the segmentor) is a segmentation network inspired by U-Net and integrated with residual blocks, whereas the latter acts as critic and is made only of the encoding path of the segmentor. A dataset of 900 images acquired in 6 surgical cases was collected and labelled to validate the proposed approach. The adversarial networks achieved a median Dice similarity coefficient of 91.91% with Inter-Quartile Range (IQR) of 4.63%, overcoming approaches based on U-Net (82.98%-IQR : 14.41%) and U-Net with residual blocks (86.13%-IQR : 13.63%). Results proved that the proposed architecture could be a valuable and robust solution to assist surgeons in providing membrane identification while performing fetoscopic surgery.