AIROGS: Artificial Intelligence for Robust Glaucoma Screening Challenge.

The early detection of glaucoma is essential in preventing visual impairment. Artificial intelligence (AI) can be used to analyze color fundus photographs (CFPs) in a cost-effective manner, making glaucoma screening more accessible. While AI models for glaucoma screening from CFPs have shown promising results in laboratory settings, their performance decreases significantly in real-world scenarios due to the presence of out-of-distribution and low-quality images. To address this issue, we propose the Artificial Intelligence for Robust Glaucoma Screening (AIROGS) challenge. This challenge includes a large dataset of around 113,000 images from about 60,000 patients and 500 different screening centers, and encourages the development of algorithms that are robust to ungradable and unexpected input data. We evaluated solutions from 14 teams in this paper and found that the best teams performed similarly to a set of 20 expert ophthalmologists and optometrists. The highest-scoring team achieved an area under the receiver operating characteristic curve of 0.99 (95% CI: 0.98-0.99) for detecting ungradable images on-the-fly. Additionally, many of the algorithms showed robust performance when tested on three other publicly available datasets. These results demonstrate the feasibility of robust AI-enabled glaucoma screening.

Detection of maxillary sinus fungal ball via 3-D CNN-based artificial intelligence: Fully automated system and clinical validation.

BACKGROUND: This study aims to develop artificial intelligence (AI) system to automatically classify patients with maxillary sinus fungal ball (MFB), chronic rhinosinusitis (CRS), and healthy controls (HCs). METHODS: We collected 512 coronal image sets from ostiomeatal unit computed tomography (OMU CT) performed on subjects who visited a single tertiary hospital. These data included 254 MFB, 128 CRS, and 130 HC subjects and were used for training the proposed AI system. The AI system takes these 1024 sets of half CT images as input and classifies these as MFB, CRS, or HC. To optimize the classification performance, we adopted a 3-D convolutional neural network of ResNet 18. We also collected 64 coronal OMU CT image sets for external validation, including 26 MFB, 18 CRS, and 20 HCs from subjects from another referral hospital. Finally, the performance of the developed AI system was compared with that of the otolaryngology resident physicians. RESULTS: Classification performance was evaluated using internal 5-fold cross-validation (818 training and 206 internal validation data) and external validation (128 data). The area under the receiver operating characteristic over the internal 5-fold cross-validation and the external validation was 0.96 ±0.006 and 0.97 ±0.006, respectively. The accuracy of the internal 5-fold cross-validation and the external validation was 87.5 ±2.3% and 88.4 ±3.1%, respectively. As a result of performing a classification test on external validation data from six otolaryngology resident physicians, the accuracy was obtained as 84.6 ±11.3%. CONCLUSIONS: This AI system is the first study to classify MFB, CRS, and HC using deep neural networks to the best of our knowledge. The proposed system is fully automatic but performs similarly to or better than otolaryngology resident physicians. Therefore, we believe that in regions where otolaryngology specialists are scarce, the proposed AI will perform sufficiently effective diagnosis on behalf of doctors.

Quantitative CT-based structural alterations of segmental airways in cement dust-exposed subjects.

BACKGROUND: Dust exposure has been reported as a risk factor of pulmonary disease, leading to alterations of segmental airways and parenchymal lungs. This study aims to investigate alterations of quantitative computed tomography (QCT)-based airway structural and functional metrics due to cement-dust exposure. METHODS: To reduce confounding factors, subjects with normal spirometry without fibrosis, asthma and pneumonia histories were only selected, and a propensity score matching was applied to match age, sex, height, smoking status, and pack-years. Thus, from a larger data set (N = 609), only 41 cement dust-exposed subjects were compared with 164 non-cement dust-exposed subjects. QCT imaging metrics of airway hydraulic diameter (Dh), wall thickness (WT), and bifurcation angle (θ) were extracted at total lung capacity (TLC) and functional residual capacity (FRC), along with their deformation ratios between TLC and FRC. RESULTS: In TLC scan, dust-exposed subjects showed a decrease of Dh (airway narrowing) especially at lower-lobes (p < 0.05), an increase of WT (wall thickening) at all segmental airways (p < 0.05), and an alteration of θ at most of the central airways (p < 0.001) compared with non-dust-exposed subjects. Furthermore, dust-exposed subjects had smaller deformation ratios of WT at the segmental airways (p < 0.05) and θ at the right main bronchi and left main bronchi (p < 0.01), indicating airway stiffness. CONCLUSIONS: Dust-exposed subjects with normal spirometry demonstrated airway narrowing at lower-lobes, wall thickening at all segmental airways, a different bifurcation angle at central airways, and a loss of airway wall elasticity at lower-lobes. The airway structural alterations may indicate different airway pathophysiology due to cement dusts.

Erratum: Structural and Functional Features on Quantitative Chest Computed Tomography in the Korean Asian versus the White American Healthy Non-Smokers.

This corrects the article on p. 20 in vol. 1236, PMID: 31270987.

Structural and Functional Features on Quantitative Chest Computed Tomography in the Korean Asian versus the White American Healthy Non-Smokers.

OBJECTIVE: Considering the different prevalence rates of diseases such as asthma and chronic obstructive pulmonary disease in Asians relative to other races, Koreans may have unique airway structure and lung function. This study aimed to investigate unique features of airway structure and lung function based on quantitative computed tomography (QCT)-imaging metrics in the Korean Asian population (Koreans) as compared with the White American population (Whites). MATERIALS AND METHODS: QCT data of healthy non-smokers (223 Koreans vs. 70 Whites) were collected, including QCT structural variables of wall thickness (WT) and hydraulic diameter (Dh) and functional variables of air volume, total air volume change in the lung (ΔVair), percent emphysema-like lung (Emph%), and percent functional small airway disease-like lung (fSAD%). Mann-Whitney U tests were performed to compare the two groups. RESULTS: As compared with Whites, Koreans had smaller volume at inspiration, ΔVair between inspiration and expiration (p < 0.001), and Emph% at inspiration (p < 0.001). Especially, Korean females had a decrease of ΔVair in the lower lobes (p < 0.001), associated with fSAD% at the lower lobes (p < 0.05). In addition, Koreans had smaller Dh and WT of the trachea (both, p < 0.05), correlated with the forced expiratory volume in 1 second (R = 0.49, 0.39; all p < 0.001) and forced vital capacity (R = 0.55, 0.45; all p < 0.001). CONCLUSION: Koreans had unique features of airway structure and lung function as compared with Whites, and the difference was clearer in female individuals. Discriminating structural and functional features between Koreans and Whites enables exploration of inter-racial differences of pulmonary disease in terms of severity, distribution, and phenotype.