A preliminary deep learning study on automatic segmentation of contrast-enhanced bolus in videofluorography of swallowing

Abstract

Although videofluorography (VFG) is an effective tool for evaluating swallowing functions, its accurate evaluation requires considerable time and effort. This study aimed to create a deep learning model for automated bolus segmentation on VFG images of patients with healthy swallowing and dysphagia using the artificial intelligence deep learning segmentation method, and to assess the performance of the method. VFG images of 72 swallowing of 12 patients were continuously converted into 15 static images per second. In total, 3910 images were arbitrarily assigned to the training, validation, test 1, and test 2 datasets. In the training and validation datasets, images of colored bolus areas were prepared, along with original images. Using a U-Net neural network, a trained model was created after 500 epochs of training. The test datasets were applied to the trained model, and the performances of automatic segmentation (Jaccard index, Sørensen–Dice coefficient, and sensitivity) were calculated. All performance values for the segmentation of the test 1 and 2 datasets were high, exceeding 0.9. Using an artificial intelligence deep learning segmentation method, we automatically segmented the bolus areas on VFG images; our method exhibited high performance. This model also allowed assessment of aspiration and laryngeal invasion.

Automatic Segmentation of Teeth, Crown–Bridge Restorations, Dental Implants, Restorative Fillings, Dental Caries, Residual Roots, and Root Canal Fillings on Orthopantomographs: Convenience and Pitfalls

“…U-Net architecture was used to reveal this. U-Net is a convolutional neural network developed at the Freiburg University Computer Science Department for segmentation in image processing studies in biomedical fields [ 10 , 11 , 12 , 17 , 20 , 21 , 26 , 30 , 31 , 34 , 38 , 50 , 64 , 66 , 67 , 70 , 71 , 82 , 83 , 84 , 85 , 86 ].…”

“…In 1959, Arthur Samuel defined machine learning as “the ability of machines to learn results for which they were not specifically programmed” [ 9 ]. He also developed a checkers game that can work in a computer environment, learn from its own mistakes, and thus improve itself [ 1 , 10 , 11 , 12 ].…”

“…Thus, by recognizing similar sequences, similar motifs, and similar pixels, the machine can detect, segment, and classify what those pictures are. The more data there are, the better artificial intelligence features will be revealed [ 1 , 11 , 12 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ]. Deep neural networks have many hidden layers, with millions of interconnected artificial neurons.…”

“…Deep neural networks have many hidden layers, with millions of interconnected artificial neurons. A number, called weight, represents the connections between one node and another [ 11 , 12 , 30 , 31 , 32 ]. Theoretically, deep neural networks can match any type of input to any type of output; however, they require much more training compared to other machine learning methods.…”

“…Theoretically, deep neural networks can match any type of input to any type of output; however, they require much more training compared to other machine learning methods. They require millions of samples, compared to the hundreds or thousands of training data samples that a simpler network might need [ 1 , 2 , 11 , 12 , 17 , 18 , 19 , 20 , 21 , 30 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ].…”