4/18/2023 0 Comments Manifold definitionRef. grouped pure noise and motion artifacts into uninformative frames of optical endomicroscopy. employed the Shannon entropy to eliminate the uninformative images in their specific endoscopic applications. extracted the features with a local color histogram in the HSV space and isolated non-informative frames with a support vector machine classifier. Similarly, ref. observed that informative frames have higher spatial frequencies than blurred ones. Their method overcame the specific threshold. Ref. selected a set of descriptors followed by a support vector machine to classify the NBI endoscopic frames. However, the frequency spectrum contains superfluous information. Later, the k-means algorithm was employed in the endoscopic video manifolds to cluster the informative frames. Ref. found that the informative and uninformative frames can be distinguished in the frequency domain according to an energy histogram. It also suggests the patterns embedded in the data help develop flexible algorithms that do not require manual labeling. Our results demonstrate the effectiveness of the proposed scheme and the robustness of detecting the informative frames. The overall median recall of the proposed method is currently the highest, 96%. It outperforms the baseline by 12% absolute. Along with the proposed automatic cluster labeling algorithm and cost function in Bayesian optimization, the proposed method coupled with UMAP achieves state-of-the-art performance. We extract feature embedding using a vanilla neural network (VGG16) and introduce a new dimensionality reduction method called UMAP that distinguishes the feature embedding in the lower-dimensional space. Here, we show that a novel unsupervised scheme is comparable to the current benchmarks on the dataset of NBI-InfFrames. However, the definition of the uninformative categories is ambiguous, and tedious labeling still cannot be avoided. This issue is commonly addressed by a classifier with task-specific categories of the uninformative frames. It often takes a lot of time for the physician to manually inspect the informative frames. Removing the uninformative frames is vital to improve detection accuracy and speed up computer-aided diagnosis. Most images produced from the examination are useless, such as blurred, specular reflection, and underexposed. Narrow band imaging is an established non-invasive tool used for the early detection of laryngeal cancer in surveillance examinations.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |