arXiv subjects

Jennifer Toth

Publications and source records attributed to Jennifer Toth.

3 recordsLinked to original sources

arXiv2023

Autofluorescence Bronchoscopy Video Analysis for Lesion Frame Detection

Because of the significance of bronchial lesions as indicators of early lung cancer and squamous cell carcinoma, a critical need exists for early detection of bronchial lesions. Autofluorescence bronchoscopy (AFB) is a primary modality used for bronchial lesion detection, as it shows high sensitivity to suspicious lesions. The physician, however, must interactively browse a long video stream to locate lesions, making the search exceedingly tedious and error prone. Unfortunately, limited research has explored the use of automated AFB video analysis for efficient lesion detection. We propose a robust automatic AFB analysis approach that distinguishes informative and uninformative AFB video frames in a video. In addition, for the informative frames, we determine the frames containing potential lesions and delineate candidate lesion regions. Our approach draws upon a combination of computer-based image analysis, machine learning, and deep learning. Thus, the analysis of an AFB video stream becomes more tractable. Tests with patient AFB video indicate that $\ge$97\% of frames were correctly labeled as informative or uninformative. In addition, $\ge$97\% of lesion frames were correctly identified, with false positive and false negative rates $\le$3\%.

eess.IV↗

arXiv2023

Bronchoscopic video synchronization for interactive multimodal inspection of bronchial lesions

With lung cancer being the most fatal cancer worldwide, it is important to detect the disease early. A potentially effective way of detecting early cancer lesions developing along the airway walls (epithelium) is bronchoscopy. To this end, developments in bronchoscopy offer three promising noninvasive modalities for imaging bronchial lesions: white-light bronchoscopy (WLB), autofluorescence bronchoscopy (AFB), and narrow-band imaging (NBI). While these modalities give complementary views of the airway epithelium, the physician must manually inspect each video stream produced by a given modality to locate the suspect cancer lesions. Unfortunately, no effort has been made to rectify this situation by providing efficient quantitative and visual tools for analyzing these video streams. This makes the lesion search process extremely time-consuming and error-prone, thereby making it impractical to utilize these rich data sources effectively. We propose a framework for synchronizing multiple bronchoscopic videos to enable an interactive multimodal analysis of bronchial lesions. Our methods first register the video streams to a reference 3D chest computed-tomography (CT) scan to produce multimodal linkages to the airway tree. Our methods then temporally correlate the videos to one another to enable synchronous visualization of the resulting multimodal data set. Pictorial and quantitative results illustrate the potential of the methods.

eess.IV↗

arXiv2022

ESFPNet: efficient deep learning architecture for real-time lesion segmentation in autofluorescence bronchoscopic video

Lung cancer tends to be detected at an advanced stage, resulting in a high patient mortality rate. Thus, much recent research has focused on early disease detection Bronchoscopy is the procedure of choice for an effective noninvasive way of detecting early manifestations (bronchial lesions) of lung cancer. In particular, autofluorescence bronchoscopy (AFB) discriminates the autofluorescence properties of normal (green) and diseased tissue (reddish brown) with different colors. Because recent studies show AFB's high sensitivity in searching lesions, it has become a potentially pivotal method in bronchoscopic airway exams. Unfortunately, manual inspection of AFB video is extremely tedious and error prone, while limited effort has been expended toward potentially more robust automatic AFB lesion analysis. We propose a real-time (processing throughput of 27 frames/sec) deep-learning architecture dubbed ESFPNet for accurate segmentation and robust detection of bronchial lesions in AFB video streams. The architecture features an encoder structure that exploits pretrained Mix Transformer (MiT) encoders and an efficient stage-wise feature pyramid (ESFP) decoder structure. Segmentation results from the AFB airway-exam videos of 20 lung cancer patients indicate that our approach gives a mean Dice index = 0.756 and an average Intersection of Union = 0.624, results that are superior to those generated by other recent architectures. Thus, ESFPNet gives the physician a potential tool for confident real-time lesion segmentation and detection during a live bronchoscopic airway exam. Moreover, our model shows promising potential applicability to other domains, as evidenced by its state-of-the-art (SOTA) performance on the CVC-ClinicDB, ETIS-LaribPolypDB datasets, and superior performance on the Kvasir, CVC-ColonDB datasets.

eess.IV↗