Vision-Based Assistance for Vocal Fold Identification in Laryngoscopy with Knowledge Distillation.

Laryngoscopy images play a vital role in merging computer vision and otorhinolaryngology research. However, limited studies offer laryngeal datasets for comparative evaluation. Hence, this study introduces a novel dataset focusing on vocal fold images. Additionally, we propose a lightweight network utilizing knowledge distillation, with our student model achieving around 98.4% accuracy-comparable to the original EfficientNetB1 while reducing model weights by up to 88%. We also present an AI-assisted smartphone solution, enabling a portable and intelligent laryngoscopy system that aids laryngoscopists in efficiently targeting vocal fold areas for observation and diagnosis. To sum up, our contribution includes a laryngeal image dataset and a compressed version of the efficient model, suitable for handheld laryngoscopy devices.

Towards long-tailed, multi-label disease classification from chest X-ray: Overview of the CXR-LT challenge.

Many real-world image recognition problems, such as diagnostic medical imaging exams, are "long-tailed" - there are a few common findings followed by many more relatively rare conditions. In chest radiography, diagnosis is both a long-tailed and multi-label problem, as patients often present with multiple findings simultaneously. While researchers have begun to study the problem of long-tailed learning in medical image recognition, few have studied the interaction of label imbalance and label co-occurrence posed by long-tailed, multi-label disease classification. To engage with the research community on this emerging topic, we conducted an open challenge, CXR-LT, on long-tailed, multi-label thorax disease classification from chest X-rays (CXRs). We publicly release a large-scale benchmark dataset of over 350,000 CXRs, each labeled with at least one of 26 clinical findings following a long-tailed distribution. We synthesize common themes of top-performing solutions, providing practical recommendations for long-tailed, multi-label medical image classification. Finally, we use these insights to propose a path forward involving vision-language foundation models for few- and zero-shot disease classification.

LifeSeeker: an interactive concept-based retrieval system for lifelog data.

Lifelogging was introduced as the process of passively capturing personal daily events via wearable devices. It ultimately creates a visual diary encoding every aspect of one's life with the aim of future sharing or recollecting. In this paper, we present LifeSeeker, a lifelog image retrieval system participating in the Lifelog Search Challenge (LSC) for 3 years, since 2019. Our objective is to support users to seek specific life moments using a combination of textual descriptions, spatial relationships, location information, and image similarities. In addition to the LSC challenge results, a further experiment was conducted in order to evaluate the power retrieval of our system on both expert and novice users. This experiment informed us about the effectiveness of the user's interaction with the system when involving non-experts.

MANet: Multi-branch attention auxiliary learning for lung nodule detection and segmentation.

BACKGROUND AND OBJECTIVE: Pulmonary nodule detection and segmentation are currently two primary tasks in analyzing chest computed tomography (Chest CT) in order to detect signs of lung cancer, thereby providing early treatment measures to reduce mortality. Even though there are many proposed methods to reduce false positives for obtaining effective detection results, distinguishing between the pulmonary nodule and background region remains challenging because their biological characteristics are similar and varied in size. The purpose of our work is to propose a method for automatic nodule detection and segmentation in Chest CT by enhancing the feature information of pulmonary nodules. METHODS: We propose a new UNet-based backbone with multi-branch attention auxiliary learning mechanism, which contains three novel modules, namely, Projection module, Fast Cascading Context module, and Boundary Enhancement module, to further enhance the nodule feature representation. Based on that, we build MANet, a lung nodule localization network that simultaneously detects and segments precise nodule positions. Furthermore, our MANet contains a Proposal Refinement step which refines initially generated proposals to effectively reduce false positives and thereby produce the segmentation quality. RESULTS: Comprehensive experiments on the combination of two benchmarks LUNA16 and LIDC-IDRI show that our proposed model outperforms state-of-the-art methods in the tasks of nodule detection and segmentation tasks in terms of FROC, IoU, and DSC metrics. Our method reports an average FROC score of 88.11% in lung nodule detection. For the lung nodule segmentation, the results reach an average IoU score of 71.29% and a DSC score of 82.74%. The ablation study also shows the effectiveness of the new modules which can be integrated into other UNet-based models. CONCLUSIONS: The experiments demonstrated our method with multi-branch attention auxiliary learning ability are a promising approach for detecting and segmenting the pulmonary nodule instances compared to the original UNet design.

Support of deep learning to classify vocal fold images in flexible laryngoscopy.

PURPOSE: To collect a dataset with adequate laryngoscopy images and identify the appearance of vocal folds and their lesions in flexible laryngoscopy images by objective deep learning models. METHODS: We adopted a number of novel deep learning models to train and classify 4549 flexible laryngoscopy images as no vocal fold, normal vocal folds, and abnormal vocal folds. This could help these models recognize vocal folds and their lesions within these images. Ultimately, we made a comparison between the results of the state-of-the-art deep learning models, and another comparison of the results between the computer-aided classification system and ENT doctors. RESULTS: This study exhibited the performance of the deep learning models by evaluating laryngoscopy images collected from 876 patients. The efficiency of the Xception model was higher and steadier than almost the rest of the models. The accuracy of no vocal fold, normal vocal folds, and vocal fold abnormalities on this model were 98.90 %, 97.36 %, and 96.26 %, respectively. Compared to our ENT doctors, the Xception model produced better results than a junior doctor and was near an expert. CONCLUSION: Our results show that current deep learning models can classify vocal fold images well and effectively assist physicians in vocal fold identification and classification of normal or abnormal vocal folds.

Revisiting natural user interaction in virtual world.

Natural user interaction in virtual environment is a prominent factor in any mixed reality applications. In this paper, we revisit the assessment of natural user interaction via a case study of a virtual aquarium. Viewers with the wearable headsets are able to interact with virtual objects via head orientation, gaze, gesture, and visual markers. The virtual environment is operated on both Google Cardboard and HoloLens, the two popular wireless head-mounted displays. Evaluation results reveal the preferences of users over different natural user interaction methods.

Hyperglycemia in Severe and Critical COVID-19 Patients: Risk Factors and Outcomes.

BACKGROUND: Hyperglycemia is commonly seen in critically ill patients. This disorder was also seen in coronavirus disease 2019 (COVID-19) patients and was associated with a worse prognosis. The current study determined the prevalence, risk factors, and prognostic implications of hyperglycemia in COVID-19 patients. METHOD: This was a retrospective observational study performed in an intensive care unit for COVID-19 patients. Electronic data of COVID-19 patients admitted to the intensive care unit from August 2nd to October 15th, 2021, were collected. Patients were divided into non-hyperglycemia, hyperglycemia in diabetic patients, and hyperglycemia in non-diabetic patients. Primary outcomes were 28-day and in-hospital mortalities. Multinomial logistic regression and multivariable Cox regression models were used to determine the risk factors for hyperglycemia and mortality, respectively. RESULTS: Hyperglycemia was documented in 65.6% of patients: diabetic patients (44.8%) and new-onset hyperglycemia (20.8%). In-hospital and 28-day mortality rates were 30.2% and 26.1%, respectively. Respiratory failure, corticosteroid therapy, and a higher level of procalcitonin were risk factors for hyperglycemia in diabetic patients, whereas cardiovascular diseases, respiratory failure, and higher aspartate aminotransferase/glutamate aminotransferase ratio were risk factors for hyperglycemia in non-diabetic patients. The risk of the 28-day mortality rate was highest in the new-onset hyperglycemia (hazard ratio [HR] 3.535, 95% confidence interval [CI] 1.338-9.338, p=0.011), which was higher than hyperglycemia in type 2 diabetes mellitus patients (HR 1.408, 95% CI 0.513-3.862, p=0.506). CONCLUSION: Hyperglycemia was common in COVID-19 patients in the intensive care unit. Hyperglycemia reflected the disease severity but was also secondary to therapeutic intervention. New-onset hyperglycemia was associated with poorer outcomes than that in diabetic patients.

Surface-based protein domains retrieval methods from a SHREC2021 challenge.

Proteins are essential to nearly all cellular mechanism and the effectors of the cells activities. As such, they often interact through their surface with other proteins or other cellular ligands such as ions or organic molecules. The evolution generates plenty of different proteins, with unique abilities, but also proteins with related functions hence similar 3D surface properties (shape, physico-chemical properties, …). The protein surfaces are therefore of primary importance for their activity. In the present work, we assess the ability of different methods to detect such similarities based on the geometry of the protein surfaces (described as 3D meshes), using either their shape only, or their shape and the electrostatic potential (a biologically relevant property of proteins surface). Five different groups participated in this contest using the shape-only dataset, and one group extended its pre-existing method to handle the electrostatic potential. Our comparative study reveals both the ability of the methods to detect related proteins and their difficulties to distinguish between highly related proteins. Our study allows also to analyze the putative influence of electrostatic information in addition to the one of protein shapes alone. Finally, the discussion permits to expose the results with respect to ones obtained in the previous contests for the extended method. The source codes of each presented method have been made available online.

Camouflaged Instance Segmentation In-the-Wild: Dataset, Method, and Benchmark Suite.

This paper pushes the envelope on decomposing camouflaged regions in an image into meaningful components, namely, camouflaged instances. To promote the new task of camouflaged instance segmentation of in-the-wild images, we introduce a dataset, dubbed CAMO++, that extends our preliminary CAMO dataset (camouflaged object segmentation) in terms of quantity and diversity. The new dataset substantially increases the number of images with hierarchical pixel-wise ground truths. We also provide a benchmark suite for the task of camouflaged instance segmentation. In particular, we present an extensive evaluation of state-of-the-art instance segmentation methods on our newly constructed CAMO++ dataset in various scenarios. We also present a camouflage fusion learning (CFL) framework for camouflaged instance segmentation to further improve the performance of state-of-the-art methods. The dataset, model, evaluation suite, and benchmark will be made publicly available on our project page.

A comprehensive analysis of classification methods in gastrointestinal endoscopy imaging.

Gastrointestinal (GI) endoscopy has been an active field of research motivated by the large number of highly lethal GI cancers. Early GI cancer precursors are often missed during the endoscopic surveillance. The high missed rate of such abnormalities during endoscopy is thus a critical bottleneck. Lack of attentiveness due to tiring procedures, and requirement of training are few contributing factors. An automatic GI disease classification system can help reduce such risks by flagging suspicious frames and lesions. GI endoscopy consists of several multi-organ surveillance, therefore, there is need to develop methods that can generalize to various endoscopic findings. In this realm, we present a comprehensive analysis of the Medico GI challenges: Medical Multimedia Task at MediaEval 2017, Medico Multimedia Task at MediaEval 2018, and BioMedia ACM MM Grand Challenge 2019. These challenges are initiative to set-up a benchmark for different computer vision methods applied to the multi-class endoscopic images and promote to build new approaches that could reliably be used in clinics. We report the performance of 21 participating teams over a period of three consecutive years and provide a detailed analysis of the methods used by the participants, highlighting the challenges and shortcomings of the current approaches and dissect their credibility for the use in clinical settings. Our analysis revealed that the participants achieved an improvement on maximum Mathew correlation coefficient (MCC) from 82.68% in 2017 to 93.98% in 2018 and 95.20% in 2019 challenges, and a significant increase in computational speed over consecutive years.

A Multi-Organ Nucleus Segmentation Challenge.

Generalized nucleus segmentation techniques can contribute greatly to reducing the time to develop and validate visual biomarkers for new digital pathology datasets. We summarize the results of MoNuSeg 2018 Challenge whose objective was to develop generalizable nuclei segmentation techniques in digital pathology. The challenge was an official satellite event of the MICCAI 2018 conference in which 32 teams with more than 80 participants from geographically diverse institutes participated. Contestants were given a training set with 30 images from seven organs with annotations of 21,623 individual nuclei. A test dataset with 14 images taken from seven organs, including two organs that did not appear in the training set was released without annotations. Entries were evaluated based on average aggregated Jaccard index (AJI) on the test set to prioritize accurate instance segmentation as opposed to mere semantic segmentation. More than half the teams that completed the challenge outperformed a previous baseline. Among the trends observed that contributed to increased accuracy were the use of color normalization as well as heavy data augmentation. Additionally, fully convolutional networks inspired by variants of U-Net, FCN, and Mask-RCNN were popularly used, typically based on ResNet or VGG base architectures. Watershed segmentation on predicted semantic segmentation maps was a popular post-processing strategy. Several of the top techniques compared favorably to an individual human annotator and can be used with confidence for nuclear morphometrics.