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.

No abstract

Objectives: This study presents SmartProbe, an electrical bioimpedance (EBI) sensing system based on a concentric needle electrode (CNE). The system allows the use of commercial CNEs for accurate EBI measurement, and was specially developed for in-vivo real-time cancer detection. Approach: Considering the uncertainties in EBI measurements due to the CNE manufacturing tolerances, we propose a calibration method based on statistical learning. This is done by extracting the correlation between the measured impedance value |Z|, and the material conductivity σ, for a group of reference materials. By utilizing this correlation, the relationship of σ and |Z| can be described as a function and reconstructed using a single measurement on a reference material of known conductivity. Main results: This method simplifies the calibration process, and is verified experimentally. Its effectiveness is demonstrate by results that show less than 6% relative error. An additional experiment is conducted for evaluating the system’s capability to detect cancerous tissue. Four types of ex-vivo human tissue from the head and neck region, including mucosa, muscle, cartilage and salivary gland, are characterized using SmartProbe. The measurements include both cancer and surrounding healthy tissue excised from 10 different patients operated on for head and neck cancer. The measured data is then processed using dimension reduction and analyzed for tissue classification. The final results show significant differences between pathologic and healthy tissues in muscle, mucosa and cartilage specimens. Significance: These results are highly promising and indicate a great potential for SmartProbe to be used in various cancer detection tasks.

No abstract