While supervised learning techniques have demonstrated state-of-the-art performance in many medical image analysis tasks, the role of sample selection is important. Selecting the most informative samples contributes to the system attaining optimum performance with minimum labeled samples, which translates to fewer expert interventions and cost. Active Learning (AL) methods for informative sample selection are effective in boosting performance of computer aided diagnosis systems when limited labels are available. Conventional approaches to AL have mostly focused on the single label setting where a sample has only one disease label from the set of possible labels. These approaches do not perform optimally in the multi-label setting where a sample can have multiple disease labels (e.g. in chest X-ray images). In this paper we propose a novel sample selection approach based on graph analysis to identify informative samples in a multi-label setting. For every analyzed sample, each class label is denoted as a separate node of a graph. Building on findings from interpretability of deep learning models, edge interactions in this graph characterize similarity between corresponding interpretability saliency map model encodings. We explore different types of graph aggregation to identify informative samples for active learning. We apply our method to public chest X-ray and medical image datasets, and report improved results over state-of-the-art AL techniques in terms of model performance, learning rates, and robustness.

Background: Fibrosis in pulmonary Langerhans cell histiocytosis (PLCH) histologically comprises a central scar with septal strands and associated airspace enlargement that produce an octopus-like appearance. The purpose of this study was to identify the octopus sign on high-resolution computed tomography (HRCT) images to determine its frequency and distribution across stages of the disease. Methods: Fifty-seven patients with confirmed PLCH were included. Two experienced chest radiologists assessed disease stages as early, intermediate, or late, as well as the lung parenchyma for nodular, cystic, or fibrotic changes and for the presence of the octopus sign. Statistical analysis included Cohen’s kappa for interrater agreement and Fisher’s exact test for the frequency of the octopus sign. Results: Interobserver agreement was substantial for the octopus sign (kappa = 0.747). Significant differences in distribution of the octopus sign between stages 2 and 3 were found with more frequent octopus signs in stage 2 and fewer in stage 3. In addition, we only found the octopus sign in cases of nodular und cystic lung disease. Conclusions: The octopus sign in PLCH can be identified not only on histological images, but also on HRCT images. Its radiological presence seems to depend on the stage of PLCH.

Data-driven approaches have achieved great success in various medical image analysis tasks. However, fully-supervised data-driven approaches require unprecedentedly large amounts of labeled data and often suffer from poor generalization to unseen new data due to domain shifts. Various unsupervised domain adaptation (UDA) methods have been actively explored to solve these problems. Anatomical and spatial priors in medical imaging are common and have been incorporated into data-driven approaches to ease the need for labeled data as well as to achieve better generalization and interpretation. Inspired by the effectiveness of recent transformer-based methods in medical image analysis, the adaptability of transformer-based models has been investigated. How to incorporate prior knowledge for transformer-based UDA models remains under-explored. In this paper, we introduce a prior knowledge-guided and transformer-based unsupervised domain adaptation (PUDA) pipeline. It regularizes the vision transformer attention heads using anatomical and spatial prior information that is shared by both the source and target domain, which provides additional insight into the similarity between the underlying data distribution across domains. Besides the global alignment of class tokens, it assigns local weights to guide the token distribution alignment via adversarial training. We evaluate our proposed method on a clinical outcome prediction task, where Computed Tomography (CT) and Chest X-ray (CXR) data are collected and used to predict the intubation status of patients in a week. Abnormal lesions are regarded as anatomical and spatial prior information for this task and are annotated in the source domain scans. Extensive experiments show the effectiveness of the proposed PUDA method.

Objectives: The aim of this study was to evaluate the severity of COVID-19 patients' disease by comparing a multiclass lung lesion model to a single-class lung lesion model and radiologists' assessments in chest computed tomography scans. Materials and Methods: The proposed method, AssessNet-19, was developed in 2 stages in this retrospective study. Four COVID-19–induced tissue lesions were manually segmented to train a 2D-U-Net network for a multiclass segmentation task followed by extensive extraction of radiomic features from the lung lesions. LASSO regression was used to reduce the feature set, and the XGBoost algorithm was trained to classify disease severity based on the World Health Organization Clinical Progression Scale. The model was evaluated using 2 multicenter cohorts: a development cohort of 145 COVID-19–positive patients from 3 centers to train and test the severity prediction model using manually segmented lung lesions. In addition, an evaluation set of 90 COVID-19–positive patients was collected from 2 centers to evaluate AssessNet-19 in a fully automated fashion. Results: AssessNet-19 achieved an F1-score of 0.76 ± 0.02 for severity classification in the evaluation set, which was superior to the 3 expert thoracic radiologists (F1 = 0.63 ± 0.02) and the single-class lesion segmentation model (F1 = 0.64 ± 0.02). In addition, AssessNet-19 automated multiclass lesion segmentation obtained a mean Dice score of 0.70 for ground-glass opacity, 0.68 for consolidation, 0.65 for pleural effusion, and 0.30 for band-like structures compared with ground truth. Moreover, it achieved a high agreement with radiologists for quantifying disease extent with Cohen κ of 0.94, 0.92, and 0.95. Conclusions: A novel artificial intelligence multiclass radiomics model including 4 lung lesions to assess disease severity based on the World Health Organization Clinical Progression Scale more accurately determines the severity of COVID-19 patients than a single-class model and radiologists' assessment.

Idiopathic pulmonary fibrosis (IPF), hypersensitivity pneumonitis (HP) and systemic sclerosis (SSc) are among the most common entities that cause pulmonary fibrosis. Alveolar collapse with subsequent collapse induration of lung tissue is thought to contribute to the fibrotic transformation. The purpose of this study was to examine lung tissue in computed tomography (CT) of non-diseased appearance during expiration for signs of increased density suggesting collapsibility in fibrosing lung diseases. We further analyzed the diaphragmatic movements during the respiratory cycle to determine relationships between density differences and the apex–diaphragm diameter. Significant differences in attenuation changes between inspiration and expiration of unaffected lung parenchyma were detected between IPF and controls and between HP and controls for all lung lobes (p < 0.001). Only minor differences were found between SSc and controls. There was no clinically relevant difference between patients with IPF and those with HP. The measured absolute apex–diaphragm diameter in inspiration and expiration demonstrated a statistically significant difference between patients with IPF versus normal controls. However, the diaphragmatic excursions were not different between these groups. Compared to controls, CT lung density increases significantly more during expiration in the fibrotic lungs of IPF and HP patients. The observed increase in density might indicate the collapse of alveoli during expiration and may represent a common pathophysiologic feature of fibrosing lung diseases. The density changes and lung extensions do not have the same ratios across different diseases and controls.